neuroscience

Dopaminergic Neurons in the Ventral Tegmental Area as a Target for Treatment-Resistant Depression

August 28, 2025By logancollins No Comments

Prelude

This essay is intended as an intuitive examination of a reward system neural circuit which may serve as a useful target for new therapies aimed at fighting treatment-resistant depression (TRD). My purpose here is not to introduce an entirely novel concept, but rather to compile in one place a set of important explanations on how information flow in the reward system relates to TRD and how these reward system mechanisms may have clinical relevance.

Treatment Resistant Depression

Treatment resistant depression (TRD) is a widespread and debilitating condition. Patients with TRD are defined to have failed to adequately respond to two or more treatments for depression.¹ As a broader category, depression affects about 280 million people worldwide.² TRD affects roughly 30% of these patients³ (~84 million). In the USA, it has been estimated that about 2.8 million adults suffer from TRD.¹ A common symptom associated with TRD is anhedonia, the inability to feel positive emotions. It is thought that defects in the brain’s reward pathway are central to the neurobiology of TRD since this pathway contains the neural circuitry necessary to encode positive emotional experiences.

Reward Circuits

When sensory recognition of a potential reward occurs, various pathways inhibit activity of the lateral habenula (LHb), which in turn inhibits the rostromedial tegmental nucleus (RMTg). This disinhibits the ventral tegmental area (VTA).⁴ The VTA’s dopaminergic projections then spike in phasic bursts, sending dopamine to the nucleus accumbens (NAc) (mesolimbic pathway, a part of the medial forebrain bundle or MFB) and prefrontal cortex (PFC) (mesocortical pathway).⁵ NAc GABAergic medium spiny neurons (MSNs) generally express either the dopamine 1 receptor (D1R) or express the dopamine 2 receptor (D2R). D1R MSNs are excited by dopamine while D2R MSNs are inhibited by dopamine. Mesolimbic inputs bias the NAc to output from the D1R MSNs, which stimulate the direct motor pathway to respond to the reward. The GABAergic MSN activity furthermore inhibits the ventral pallidum (VP), which in turn lifts its own GABAergic inhibition on targets such as the mediodorsal thalamus, lateral hypothalamus, and VTA.⁶ This increases arousal and helps with motor processes. The mediodorsal thalamus projects to the PFC and triggers circuits that represent the value of the reward.⁷

These circuits facilitate reward learning by a comparative mechanism called reward prediction error (RPE). The pedunculopontine tegmental nucleus (PPTg) receives inputs about the actual reward from brainstem sensory signals (and other brain areas) and projects glutamatergic and cholinergic synapses into the VTA (and elsewhere) to activate the dopaminergic neurons.⁸ If the reward is less valuable than expected, the LHb activates, which triggers firing of GABAergic neurons in the RMTg onto the VTA dopamine neurons, shutting down the mesolimbic activity.⁴ If the actual reward remains valuable, then the mesolimbic activity continues. This process where the inhibitory LHb-RMTg signal is “subtracted” from the stimulatory PPTg signal determines the RPE comparison’s outcome and whether the VTA continues its dopaminergic signals.⁹ All of this facilitates reward learning, where mesolimbic long-term potentiation (LTP) occurs if the reward is as strong as expected (or stronger) and mesolimbic long-term depression (LTD) (not the same as psychiatric depression) occurs if the reward is not as strong as expected.

Dopaminergic Neurons and Treatment Resistant Depression

As a central driver within the reward system, dopaminergic VTA neurons have high potential as a target for combatting TRD. Activation of these neurons may alleviate anhedonia and increase motivation. There already exists clinical evidence that stimulation of VTA dopaminergic neurons has significant benefits. As mentioned, the mesolimbic pathway projections of VTA dopaminergic neurons make up a major part of the MFB. Multiple clinical studies on deep brain stimulation (DBS) of the MFB (specifically the supero-lateral MFB or slMFB) have shown long-term beneficial effects for patients with TRD.^10–12 Functional imaging evidence suggests this works indirectly through activation of descending glutamatergic fibers from the PFC which activate the VTA’s dopamine neurons.¹⁰ Dopamine axons themselves are small in diameter, which make them not as responsive to conventional DBS. It should be noted that the VTA is a highly heterogeneous structure with dopaminergic, GABAergic, and glutamatergic neurons,¹³ so DBS of the VTA in general might have off-target effects and/or partially mitigate the benefits of the stimulation. Activation of the VTA’s GABAergic and glutamatergic neurons can have markedly different effects compared to activation of only its dopaminergic neurons.¹⁴ In mice, GABAergic VTA neuronal activity particularly has been found to occur in response to aversive stimuli and stimuli predicting the absence of reward.^15,16 In rats, optogenetic stimulation of VTA dopamine neurons promotes motivated behavior while optogenetic stimulation of VTA GABA neurons disrupts reward and promotes aversion.¹⁷ Clinical and animal model evidence thus supports the idea that selective activation of VTA dopamine neurons might act as a potent therapy for TRD.

Conclusion

Based on the literature, raising the basal level of VTA dopaminergic neuron activity might demonstrate a strong ameliorative effect on TRD. Extensive preclinical and clinical testing will of course be crucial to establish safety. Possible addictiveness of treatments which activate this circuit will need careful examination in particular. Depending on the modality of treatment, different forms of neurological adaptation may occur, so ways of mitigating this issue should be explored. VTA dopaminergic neurons represent a promising target for next-generation therapies aimed at overcoming TRD.

References

1. Zhdanava, M. et al. The Prevalence and National Burden of Treatment-Resistant Depression and Major Depressive Disorder in the United States. J. Clin. Psychiatry 82, (2021).

2. World Health Organization – Depressive disorder (depression). https://www.who.int/news-room/fact-sheets/detail/depression (2023).

3. McIntyre, R. S. et al. Treatment-resistant depression: definition, prevalence, detection, management, and investigational interventions. World Psychiatry 22, 394–412 (2023).

4. Hong, S., Jhou, T. C., Smith, M., Saleem, K. S. & Hikosaka, O. Negative Reward Signals from the Lateral Habenula to Dopamine Neurons Are Mediated by Rostromedial Tegmental Nucleus in Primates. J. Neurosci. 31, 11457 LP – 11471 (2011).

5. Juarez, B. & Han, M.-H. Diversity of Dopaminergic Neural Circuits in Response to Drug Exposure. Neuropsychopharmacology 41, 2424–2446 (2016).

6. Root, D. H., Melendez, R. I., Zaborszky, L. & Napier, T. C. The ventral pallidum: Subregion-specific functional anatomy and roles in motivated behaviors. Prog. Neurobiol. 130, 29–70 (2015).

7. Haber, S. N. & Knutson, B. The Reward Circuit: Linking Primate Anatomy and Human Imaging. Neuropsychopharmacology 35, 4–26 (2010).

8. Skvortsova, V. et al. A Causal Role for the Pedunculopontine Nucleus in Human Instrumental Learning. Curr. Biol. 31, 943-954.e5 (2021).

9. Eshel, N. et al. Arithmetic and local circuitry underlying dopamine prediction errors. Nature 525, 243–246 (2015).

10. Fenoy, A. J. et al. Deep brain stimulation of the “medial forebrain bundle”: sustained efficacy of antidepressant effect over years. Mol. Psychiatry 27, 2546–2553 (2022).

11. Schlaepfer, T. E., Bewernick, B. H., Kayser, S., Mädler, B. & Coenen, V. A. Rapid Effects of Deep Brain Stimulation for Treatment-Resistant Major Depression. Biol. Psychiatry 73, 1204–1212 (2013).

12. Fenoy, A. J., Quevedo, J. & Soares, J. C. Deep brain stimulation of the “medial forebrain bundle”: a strategy to modulate the reward system and manage treatment-resistant depression. Mol. Psychiatry 27, 574–592 (2022).

13. Faget, L. et al. Afferent Inputs to Neurotransmitter-Defined Cell Types in the Ventral Tegmental Area. Cell Rep. 15, 2796–2808 (2016).

14. Root, D. H. et al. Distinct Signaling by Ventral Tegmental Area Glutamate, GABA, and Combinatorial Glutamate-GABA Neurons in Motivated Behavior. Cell Rep. 32, (2020).

15. van Zessen, R., Phillips, J. L., Budygin, E. A. & Stuber, G. D. Activation of VTA GABA Neurons Disrupts Reward Consumption. Neuron 73, 1184–1194 (2012).

16. Tan, K. R. et al. GABA Neurons of the VTA Drive Conditioned Place Aversion. Neuron 73, 1173–1183 (2012).

17. Tong, Y., Pfeiffer, L., Serchov, T., Coenen, V. A. & Döbrössy, M. D. Optogenetic stimulation of ventral tegmental area dopaminergic neurons in a female rodent model of depression: The effect of different stimulation patterns. J. Neurosci. Res. 100, 897–911 (2022).

Notes on Ultrasound Physics and Instrumentation

May 16, 2025October 25, 2025By logancollins No Comments

PDF version: Notes on Ultrasound Physics and Instrumentation – by Logan Thrasher Collins

Fundamentals of ultrasound waves

Sound waves such as those in ultrasound are longitudinal waves where particles oscillate backwards and forwards along the wave’s direction of propagation. This forms regions of higher (compression) and lower (rarefaction) density. Sound waves occur through the exchange of kinetic energy from molecular movement with the potential energy from elastic compression and stretching of bonds.

The speed of sound c depends on the medium. For example, in air c = 330 m/s while in water c = 1480 m/s. The frequency of a sound wave depends on the source producing it. Frequency is measured in Hertz (Hz) where 1 Hz = 1 cycle/second. Sound waves with frequencies greater than or equal to 20 kHz are referred to as ultrasound. The wavelength of a sound wave is λ = c/f. It is measured in mm (or other units of length). A sound wave’s phase describes what position the wave starts at within a cycle of oscillation and is measured in degrees or radians. Phase shifts are typically measured relative to a phase of 0° (or 0 rad). A sound wave’s amplitude corresponds to its “loudness” and describes the height of the wave’s peaks (or depth of its troughs). Cosine-based or complex exponential equations can describe sound waves in a similar way as they describe electromagnetic waves. These equations include amplitude as a coefficient, frequency as a factor multiplied by time, and phase shift as a term added to time.

Ultrasound pressure, power, and intensity

A sound wave’s excess pressure is the difference between its peak amplitude and the normal ambient pressure of the medium. When a medium is compressed, the excess pressure is positive and when it is rarified, the excess pressure is negative. In practice, the ambient pressure is often quite low and can be ignored, in which case pressure is used rather than excess pressure. Excess pressure and pressure are measured in Pascals (Pa), which are equivalent to N/m².

When an ultrasound wave passes through a medium, it deposits energy (measured in Joules) into said medium. The rate at which a source produces this energy is the power, which is measured in Watts or J/s. An ultrasound wave’s power is unevenly distributed across the beam and often is more concentrated near the beam’s center. Intensity is a measure of the power flowing through a unit area perpendicular (or normal) to the wave’s direction of propagation and is measured in W/m² or W/cm². Intensity is proportional to the wave’s pressure squared (I ∝ p²).

Ultrasound and its medium

As mentioned, the medium (material) an ultrasound wave passes through determines the sound’s speed of propagation. Specifically, the material’s density and stiffness determine the speed of sound propagation. Density ρ is often measured in kg/m³. For instance, bone’s density ρ_bone = 1850 kg/m³ and water’s density ρ_water = 1000 kg/m³ (or 1 g/cm³). Stiffness k describes how much a material resists deformation by force F. It equals the amount of pressure (stress) needed to change the thickness of a material by a given fraction and is measured in Pa. The ratio of change in thickness ΔL over original thickness L₀ is called tensile strain ε. The increase (or decrease) in length of a material due to applied force is denoted ΔL = L – L₀. Stress σ or fractional change in thickness is given by change in thickness divided by original thickness of the sample. Stress over strain is the elastic modulus E (also called Young’s modulus). Here, A is the cross-sectional area of a sample perpendicular to the applied force. Note that these formulas apply only for tensile stress (which induces tensile strain), the type of stress where a material is compressed or elongated.

The relationship between a sound wave’s speed and the properties of density and thickness can be modeled by masses on springs where ρ = m = density (corresponding to masses in the model) and k = stiffness (corresponding to the spring stiffness in the model). The relationship between c, ρ, and k is given by the following equation.

As a result of these properties, the speed of sound varies in different tissues. A table of the approximate speeds of sound across selected tissues is provided.

Acoustic impedance z measures the response of a medium’s particles in terms of their velocity v to a sound wave of a given pressure p. Acoustic impedance can also be expressed in terms of density and stiffness or in terms of density and the sound’s speed. The latter form of definition is called the characteristic acoustic impedance of the tissue. It has units of kg m^–2 s^–1, also called rayl.

Ultrasound reflection and refraction

When an ultrasound wave traveling through a medium of acoustic impedance z₁ encounters a new medium of different acoustic impedance z₂, some of the wave is transmitted and some is reflected back. The equations in the following discussion will initially assume that the wave approaches the interface at a 90° angle (perpendicular) until stated otherwise.

To maintain continuity across the interface, the following equations must hold. Here, p_i and v_i are initial pressure and velocity, p_r and v_r are reflected pressure and velocity, and p_t and v_t are transmitted pressure and velocity.

Additionally, the intensity transmitted I_t across the interface equals the incident intensity I_i minus the reflected intensity I_r.

By using the equation for the definition of acoustic impedance z = p/v, the following two equations are true. Algebraically manipulating these equations leads to the third equation below. R_p is called the amplitude (or pressure) reflection coefficient of the interface. It is important because it decides the amplitude of the echoes produced at various interfaces within the tissue.

Note that if the acoustic impedance of the first medium is greater than the acoustic impedance of the second medium (i.e. z₁ > z₂), then R_p is negative and the reflected wave is inverted (across the x axis). For most interfaces between soft tissues, R_p is quite small and so most of the wave is transmitted to produce further echoes deeper within the tissue. This is useful for ultrasound imaging. Since the interface between air and soft tissue has a much larger R_p, the ultrasound source must be placed in direct contact with the patient’s skin to avoid air blocking transmission. This also means that gas-containing tissues like lungs and gut effectively block imaging beyond the region with the gas.

Another way to describe reflection is in terms of the intensity reflection coefficient R_I. Since I ∝ p², this means that R_I = R_p².

Since the incident intensity is I_i = I_t + I_r and the transmitted intensity is I_t = I_i – I_r, one can also define an intensity transmission coefficient T_I as the first equation below. Since the transmitted pressure is p_t = p_i + p_r, one can also define a pressure transmission coefficient T_p as the second equation below.

It is useful to realize that, because energy at interfaces must be conserved, the following equation holds true.

But ultrasound waves often do not approach interfaces at a 90° angle, so the equations in this section must be modified via trigonometry to account for other angles. First, note that the incident angle θ_i will equal the backscattered angle θ_r (this is true in the 90° case as well).

When a sound wave in a less dense tissue (slower sound wave) crosses into a tissue of greater density (faster sound wave), the transmitted wave bends away from the normal and thus θ_t > θ_i.

Likewise, when a sound wave in a tissue of greater density (faster sound wave) crosses into a tissue of less density (slower sound wave), the transmitted wave bends towards the normal and thus θ_t < θ_i.

Ultrasound scattering

When ultrasound encounters an object which is small compared to the wavelength, it scatters in all directions, though slightly more energy is typically backscattered towards the transducer than away from it. Scattering specifics depend on the shape, size, and acoustic properties (z, k, ρ) of the object.

When many similar small objects are close together (e.g. red blood cells), constructive interference can occur, which is useful. In the case of the blood, this is the basis for Doppler ultrasound, which measures blood flow. By comparison, when many small objects are far apart, complex interference patterns occur. This leads to a phenomenon in images known as “speckle”, which is usually (but not always) considered a form of undesirable noise.

Absorption and relaxation of ultrasound

As ultrasound moves through tissue, it loses energy due to absorption, resulting in heat. There are two ways this occurs: relaxation absorption and classical absorption. The effects of relaxation absorption are typically much more dominant.

Relaxation absorption depends on the elastic properties of tissue, occurring when the tissue returns to its original state after rarefaction or compression by ultrasound. This is quantified by the relaxation time τ = 1/f_r, which is how long the tissue takes to return to its original state after the ultrasound’s effect.

Relaxation is characterized by a relaxation absorption coefficient β_r which is given by the first two equations below. Here, f_r is the frequency of relaxation, f is the frequency of ultrasound, and B₀ is a material-specific constant. In practice, tissues contain a range of values of τ and f_r, so the third equation below is a more general formula where the overall relaxation absorption coefficient is proportional to the sum of the various contributions. Higher values of β_r mean more energy is absorbed into the tissue.

It is useful to note that in tissues, the relationship between the relaxation absorption coefficient β_r and the frequency f is approximately linear.

Classical absorption is less important in tissues since, at clinical frequencies, the relaxation absorption is strongly dominant as mentioned. That said, overall absorption does consist of a combination of relaxation and classical absorption (though the latter may be approximated away sometimes). Classical absorption occurs because of friction between particles as they are displaced by ultrasound, causing loss of energy to heat. This loss is characterized by the classical absorption coefficient β_class ∝ f².

Attenuation coefficients

When an ultrasound beam propagates through tissue, the sum of the absorption and scattering is described as attenuation, which causes an exponential decrease in the pressure and intensity of the ultrasound as a function of the propagation distance x through tissue.

The following equations describe the loss of ultrasound intensity and pressure as the wave moves through tissue. Here, µ is the intensity attenuation coefficient and α is the pressure attenuation coefficient. The value of µ is equal to twice the value of α (so, µ = 2α). Both have units of cm^–1, though the value of µ is often given in units of decibels (dB) per cm, where the conversion factor is µ(dB cm^–1) = 4.343µ(cm^–1). It is useful to note that each 3 dB decrease corresponds to a decrease in intensity by a factor of 2.

Approximate frequency dependences of µ are given in the table below. As an example, in soft tissue, the value of µ = 1 dB cm^–1 for 1 MHz ultrasound and µ = 2 dB cm^–1 for 2 MHz ultrasound. Note that the value of µ for fat is calculated differently than the others via the equation µ(f) = 0.7f^1.5 dB.

Ultrasound transducers

Ultrasound imaging is performed using ultrasound transducers. A gated frequency generator first produces short periodic voltage pulses, which are then amplified and fed into the transducer via a transmit-receive switch. Because the transducer transmits high power pulses and receives low intensity signals from the reflected ultrasound waves, the transmit and receive circuits must be isolated from each other. Amplified voltage is converted by a shaped piezoelectric material (typically lead zirconate titanate, which is abbreviated PZT) in the transducer into a mechanical pressure wave which is transmitted into the tissue.

After reflecting and scattering from boundaries within the tissue, pressure wave signals return to the transducer and are converted back into voltages by the piezoelectric material. The voltages must pass through low-noise preamplifier before digitization. Further amplification and signal processing facilitates display of images on a computer.

When oscillating voltage is applied to one end of shaped PZT material, the thickness of the PZT element oscillates at the same frequency as the voltage. By placing this element in contact with skin, mechanical pressure waves are transmitted into tissue. The element has a resonant frequency f₀ which is determined by its thickness T and the speed of the ultrasound wave in the PZT material c_PZT. The value of c_PZT is ~4000 m/s.

For most ultrasound devices, the transducer element’s thickness must be designed to equal one half the wavelength of ultrasound in the PZT material λ_PZT, so T = λ_PZT/2. This facilitates use of the resonant frequency.

Because of the much higher acoustic impedance of PZT material compared to skin (~18 times higher), a large amount of the energy would be reflected if the PZT was placed directly onto the skin’s surface. This would mean the mechanical wave traveling into the tissue would lose most of its energy.

To prevent this energy loss from happening, transducers possess a matching layer with a z_matchingvalue between z_skin and z_PZT as given by the equation below. The thickness of the matching layer is typically made to be 1/4 of the ultrasound’s wavelength in its material T = λ_matching/4. All this improves the transmission and reception efficiency. Sometimes multiple matching layers are used to further improve efficiency.

At the back of a PZT element, there is a damping layer, typically made of some backing material and epoxy. This damping layer prevents the PZT from continuing to oscillate (at a decaying rate) after each voltage pulse. This continued oscillation would blur the boundaries between the short pulses, which can decrease axial resolution (as will be described soon).

Although transducers have a central frequency f₀, they typically cover a range of frequencies (e.g. a 3 MHz transducer might cover a range of 1-5 MHz). Higher mechanical damping leads to broader transducer bandwidth. Transducer bandwidth is described as the frequency range over which the sensitivity is greater than half the maximum sensitivity level. The relationship between bandwidth and f₀ is often quantified by the quality factor Q, which is the ratio of f₀ to the bandwidth. Low values of Q mean larger bandwidths. Note that 2f₀ is the second harmonic frequency.

Beam geometry and resolution

FUS transducers produce a very complicated wave pattern close to the face of the transducer (the near-field or Fresnel zone). This complicated pattern is not usually useful since it has many parts where the intensity is zero. Beyond the near-field zone, the wave pattern is much simpler and decays exponentially with distance (the far-field or Fraunhofer zone). The boundary between the two zones is called the near-field boundary (NFB) and occurs at a distance Z_NFB away from the face of the transducer. The following equation (where r is the radius of the transducer) can be used to calculate the Z_NFB value.

After the NFB, the FUS beam diverges (spreads out laterally) with an angle of deviation θ which is given by the equation below.

For the far-field zone, the lateral shape of the beam approximates a Gaussian function. The full width at half maximum (FWHM) defines the lateral resolution of the beam. It is given by the equation below, where σ is the standard deviation of the Gaussian. This value is unique to each FUS beam at the specific desired depth. It can be calculated by the following equation.

Single element transducers also produce ultrasound side lobes where the first zero of the side lobe at angle φ is the same equation as the FUS beam’s divergence angle equation. In ultrasound imaging, the side lobes can cause artifacts if they are backscattered from tissue outside of the imaged region.

Axial resolution is the closest distance two boundaries can be relative to each other (in a direction parallel to the FUS beam’s propagation) while still allowing them to be resolved as two distinct features. It is given by the equation below where p_d is the pulse duration and c represents the speed of the ultrasound in the tissue.

The reason that axial resolution works this way is because the echoes of beams returning from two different boundaries are distinguishable so long as these boundaries are spaced widely enough that they do not overlap in time.

Some typical values of axial resolution are 1.5 mm at a frequency (1/c) of 1 MHz or 0.3 mm at a frequency of 5 MHz. But it should be noted that attenuation of the FUS increases at higher frequencies, so there is an important tradeoff between penetration depth and axial resolution. (Very high frequencies such as 40 MHz can be used for imaging the skin at high resolution).

Single flat ultrasound transducers possess relatively poor lateral resolution. Concave curved transducers can achieve better resolutions. (It should be noted that the transducer equations above may be somewhat altered in the case of curved transducers rather than flat transducers). To make a curved transducer, one can add a curved plastic lens in front of the piezoelectric element or the piezoelectric element itself can be made in a concave curved shape.

The shape of a transducer’s curvature can be described by an “f-number”, which is a value equal to the focal distance divided by the aperture dimension where the aperture dimension is determined by the size of the transducer element.

Lateral resolution for a bowl-shaped curved (focused) transducer is calculated using the following equation below where λ is the ultrasound wavelength, F is the focal distance, and D is the diameter of the transducer. The focal distance F is where the lateral beamwidth is narrowest and is approximated as the radius of curvature (ROC) of the lens or PZT element. This approximation is valid except in the case of very high curvature.

When deciding on the focusing power of a transducer, there is a compromise between high spatial resolution and depth over which good spatial resolution is achievable. For a strongly focused transducer, locations further away from the beam’s focal plane diverge more sharply than for a weakly focused transducer. This can be quantified by the on-axis depth-of-focus (DOF) which equals the axial distance over which the beam’s intensity is at least 50% of its maximum value.

Transducer arrays

Contemporary FUS systems typically use arrays consisting of many small piezoelectric elements (rather than single-element transducers). These arrays allow 2D imaging via electronic steering of the beam through tissue while the transducer is held at a fixed position. Sophisticated electronics produce a dynamically changing focus during pulse transmission and signal reception, which maintains high resolution throughout the image. Linear and phased array transducers represent the two main types of arrays.

A linear array consists of many (often 128-512) rectangular piezoelectric elements where the space between elements is called kerf and the distance between their centers is referred to as pitch. Each element is mechanically and electrically isolated from its neighbors by filling the kerf regions with acoustically isolating material. The elements are not focused. Pitch is designed to range from λ/2 to 3λ/2 (where λ is ultrasound wavelength in tissue). Linear arrays are usually about 1 cm wide and 10-15 cm long.

Linear arrays work by using separate voltage pulses to excite a small number of elements at slightly different times where the outer elements are excited first and the inner elements are excited after a short delay. This creates an effectively curved wavefront with a focal point at a certain distance from the array. After all backscattered echoes have been received, another beam consisting of a distinct subset of elements performs the same steps. This is repeated in sequence until all of the groups of elements have completed the procedure. If even numbers of elements were used for each group, the entire process can be repeated again with odd numbers of elements to cover the focal points between those acquired before.

Linear array focusing occurs only in one dimension. By contrast, the elevation plane (the direction perpendicular to the image plane) cannot be focused unless a curved lens is included to produce focus in this dimension. Linear arrays are most often used for applications involving large fields of view and relatively low penetration depth.

Phased arrays are typically around 1-3 cm in length and 1 cm in width. They are used in applications like cardiac imaging where there is only a small region of the body through which the ultrasound can enter without running into bone or air.

As with linear arrays, phased arrays apply voltage pulses at slightly varying times to excite elements and produce an effective wavefront with a certain focal length. But phased arrays must employ beam steering to reconstruct a full 2D image. This occurs by changing the pattern of excitation to sweep the effective wavefront beam across a range of directions to cover the image plane.

Phased arrays also employ a process called dynamic focusing to optimize lateral resolution over the full depth of imaged tissue. This involves dynamically changing the number of elements used to produce a wavefront with varying focal lengths. At deeper regions in the tissue, the number of elements needed to position the focus (where optimal lateral resolution is achieved) is higher than at shallower regions in the tissue. Dynamic focusing allows high lateral resolution across the full depth of the scan. However, dynamic focusing is relatively slow since multiple scans are needed to build up a single line of the image. It should be noted that the length of each element determines the “slice thickness” for the image’s elevation dimension.

There are also multidimensional transducer arrays which include extra rows of transducer elements. These multidimensional arrays can focus in the elevation dimension without the need for curved elements or lenses (though they are more complicated devices). Multidimensional arrays with a small number of extra rows (e.g. 3-10) are referred to as 1.5 dimensional arrays. These 1.5D arrays can facilitate some level of focusing in the elevation dimension, though to a limited extent. When multidimensional arrays possess a large number of extra rows (up to the number of elements in each row), they are referred to as 2 dimensional arrays. These can acquire full 3D image data without needing to be moved from their initial position.

Annular arrays represent another class of transducer array. They are useful at very high frequencies (>20 MHz) since linear or phased arrays are quite difficult to create for these frequencies. Annular arrays consist of concentric rings of piezoelectric material alternating with rings acoustically isolating material. Beam forming is accomplished using an analogous strategy to that of phased arrays. The outermost rings are excited first and the innermost rings last, producing an effective focus. Because annular arrays require mechanical motion to sweep the beam through tissue for reconstruction of images, commercially available devices have been developed to precisely control this motion.

When transducer arrays receive signals, they pass through an amplifier to strengthen them before digitization. However, such amplifiers do not provide linear gain for signals with a dynamic range that exceeds 40-50 dB. This is an issue because very strong signals appear from tissue boundaries near the transducer while much weaker signals appear from tissue boundaries deeper in the body. Weak signals can thus be lost when attempting to receive over larger dynamic ranges. A process called time-gain compensation (TGC) is employed to circumvent this issue. TGC increases the amplification factor as a function of time after transmission of an ultrasound pulse. As a result, the weaker backscattered echoes which come later are amplified to a greater degree than the stronger backscattered echoes which come sooner. TGC is controlled by the operator of the instrument, which usually comes with a variety of preset values for clinical imaging protocols.

Parameters for focused ultrasound in practice

There is no universally accepted definition of FUS dose, so various metrics of exposure are used to quantify how much ultrasound is delivered during a therapeutic session. Examples of such metrics (which will be discussed further below) include acoustic pressure or peak negative pressure, mechanical index, frequency, pulse repetition frequency (PRF), and intensity.

Peak negative pressure is often measured in MPa and describes the degree of rarefaction caused by the ultrasound wave in tissue. For low intensity focused ultrasound (LIFU), the greatest mechanical safety risk is from cavitation (bubble formation and collapse). To measure cavitation risk, the mechanical index (MI) is used. MI can be computed using the following equation where P_n is the peak negative pressure, f₀ is the fundamental frequency, and the derating constant of 0.3 adjusts for tissue attenuation (~7% loss per cm per MHz) and has units dBcm^–1MHz^–1. After derating, 0.3P_n has units of MPa. FDA guidelines specify that MI should not exceed 1.9. MI itself is unitless.

An ultrasound wave’s pulse’s duration (PD) is the number of cycles divided by the frequency. For instance, a pulse with 500 cycles of 500 kHz ultrasound would last for 1 ms. The pulse repetition interval (PRI) is the amount of time between the start of one pulse and the start of the next pulse (so it includes both the pulse and the pause after the pulse). Pulse repetition rate (PRR) also known as the pulse repetition frequency (PRF) equals 1/PRI. The pulse duty cycle (PDC) equals PD/PRI and is expressed as a percentage. PD typically ranges from microseconds to seconds, PRI from milliseconds to seconds, and PDC from <1% up to 70%.

One’s choice of a particular PD and PDC comes from two main factors: (i) the duty cycle can have varying neuromodulatory effects (excitatory or inhibitory) depending on its value and (ii) lower PDC values can be leveraged to limit total energy and heat deposition.

A pulse train is a series of pulses, for which the pulse train duration (PTD) equals the total number of pulses times the PRI. Typical PTDs range from less than 1 second to several minutes. The amount of time between the start of one pulse train and the start of the next is called the pulse train repetition interval (PTRI). The amount of time between pulse trains is called the interstimulus interval (ISI). The pulse train duty cycle (PTDC) equals PTD/PTRI.

It should be noted that the PTDC does not have a major influence on neuromodulatory effect, so the ratio is driven by safety such that the ISI is long enough to limit cumulative heating to reasonable levels.

Multiplying PDC by PTDC gives an overall duty cycle equal to (PD/PRI)(PTD/PTRI) which can be further multiplied by the average intensity of the pulses I_avg to obtain average temporal intensity I_{avg_tp}. FDA diagnostic safety guidelines state that average intensity should fall below 720 mW/cm². So, I_{avg_tp}= (PD/PRI)(PTD/PTRI)I_avg generally should not exceed 720 mW/cm².

Total ultrasound application time is the sum of the durations of all pulse trains plus ISIs. It typically ranges from less than 1 minute to over 60 minutes. Longer total ultrasound application time is thought to usually improve efficacy by depositing more energy, though this may not always be true. Energy per unit time might play a more significant role in efficacy, but this is an ongoing area of investigation.

Frequency (or fundamental frequency f₀) is the primary frequency of FUS passing through the tissue. It is typically measured in kilohertz (kHz) or megahertz (MHz). In human neuromodulation applications, frequency typically ranges from 200-700 kHz (or 0.2-0.7 MHz), providing an acceptable tradeoff between amount of energy entering the brain and the size of the focal region. The reason that the upper limit of frequency for human neuromodulation is typically ~700 kHz is because FUS energy attenuation by the skull at 700 kHz is ~75% (though this varies depending on skull morphology) and keeps increasing at higher frequencies.

Recall from the equations in the earlier discussion that lateral resolution involves wavelength λ (and f = c/λ), so frequency influences the focal region’s size. Also discussed earlier, frequency influences the distance from the transducer to the near-field boundary (Z_NFB). Frequency itself is generally not believed to contribute to neuromodulatory effects in a direct fashion, though this is still under investigation. So, the f₀ value is usually selected to create a focal volume of a desired size at a given depth.

Intensity is defined as power per unit area (and recall the unit of power is Watts or J/s) and is the rate at which energy is transferred by the FUS wave. For ultrasound at any given point in time during the wave cycle, intensity is proportional to the square of the acoustic pressure as described by the equation below where P is the acoustic pressure, ρ is the density of the medium, and c is the ultrasound speed in the medium. Recall that ρc = z, the acoustic impedance. Acoustic intensity is usually measured in watts per square centimeter (W/cm²).

Beyond instantaneous intensity, FUS is often measured by spatial peak pulse average intensity (I_SPPA) and by spatial peak temporal average intensity (I_SPTA). I_SPPA is the average intensity experienced during a single ultrasound pulse. Note that I does not equal P_n²/2z in the case of a ramped pulse. To determine average intensity (I_SPPA) for ramped pulses, the integral of intensity across the pulse is divided by the pulse’s duration PD. Ramped pulses distribute energy more smoothly and help mitigate auditory confounds for LIFU applications.

I_SPTA represents the average intensity of the FUS beam at the point where it is strongest averaged over the pulse duration while accounting for any off periods. It is described by the following equation consisting of the I_SPPA multiplied by the PDC (which is the fraction of PRI that the pulse is turned on).

References:

1. Legon, W. & Strohman, A. Low-intensity focused ultrasound for human neuromodulation. Nat. Rev. Methods Prim. 4, 91 (2024).

2. Smith, N. B. & Webb, A. Introduction to Medical Imaging: Physics, Engineering and Clinical Applications. (Cambridge University Press, 2010).

Reexamining the neurobiological correlates of subjective experience for whole-brain emulation (slides)

February 11, 2025February 11, 2025By logancollins No Comments

I presented these slides (PDF and images below) during the Workshop on Philosophy and Ethics of Brain Emulation (January 28th-29th, 2025) at the Mimir Center for Long Term Futures Research in Stockholm, Sweden. In my talk, I explored how various biological phenomena beyond standard neuronal electrophysiology may exert noticeable effects on the computations underlying subjective experiences. I emphasized the importance of the large range of timescales that such phenomena operate over (milliseconds to years). If we are to create emulations which think and feel like human beings, we must carefully consider the numerous tunable regulatory mechanisms the brain uses to enhance the complexity of its computational repertoire. At the workshop, we also discussed how the peripheral nervous system, enteric nervous system, endocrine system, musculoskeletal system, sensory systems, and perhaps even immune system may or may not play roles in subjective experience. To better understand what is needed to support proper emulations, I recommend recruitment of more fundamental neurobiology specialists to the whole-brain emulation community.

PDF: Reexamining the neurobiological correlates of subjective experience for whole-brain emulation

My abstract:

How much biological detail must a whole-brain emulation (WBE) incorporate to accurately preserve human subjective experience such that living as a WBE would truly feel like existing as a human? I ask this question independently of whether a nonbiological substrate can support subjective experience. Sandberg and Bostrom’s whitepaper Whole Brain Emulation: A Roadmap, briefly explores how levels of emulation detail ranging from abstract brain modules to quantum interactions between molecules may influence success criteria. They estimate that the necessary detail may at most involve an emulation of a connectome with multicompartmental models of neurons, dynamical synaptic states, and concentrations of metabolites and neurotransmitters in each compartment. For simplicity, I will refer to this as a multicompartmental emulation.

Although a multicompartmental emulation might produce an approximation of a human mind, I argue that the lack of additional biological layers of regulation could result in a subjective experience which has significant inaccuracies or missing pieces. Biological systems possess a massive number of tunable regulatory phenomena extending beyond multicompartmental electrophysiology. Some of these phenomena include but are not limited to morphological plasticity of brain cells, glial influence on computation, neurovascular coupling, adult neurogenesis, intercellular RNA transport, gap junctions, volume transmission, influence of perineuronal nets and other extracellular matrix (ECM) components, mechanical influences (e.g. crowding of synapses) on neuronal computation, ephaptic coupling, temporal evolution of genomic-transcriptomic state, and co-transmission of multiple neurotransmitters from the same synaptic bouton. In particular, experiences which depend on long-timescale changes across the brain may not be properly captured by a model which focuses on the fast electrophysiological dynamics of multicompartmental models with fixed connectomic and morphological properties.

To move towards accurately reproducing the feeling of humanness, I propose a first step of rigorously surveying neuroscience literature and evaluating how biological regulatory phenomena contribute to subjectively observed conscious experiences. This may facilitate construction of a draft catalogue where putative links between aspects of conscious experience and neurobiological phenomena can be established. Such an examination of the neural correlates of subjective experiences may serve as an initial guide for future efforts towards WBEs which preserve the feeling of humanness.

Reasons for Panpsychism

June 27, 2024June 27, 2024By logancollins 3 Comments

I have long suspected that panpsychism represents the most likely explanation of how consciousness works. My evidence for this claim is laid out below. That said, I am not an expert in philosophy of mind, so take this with a grain of salt. I am certainly open to constructive critiques, questions, and discussion as well!

Supporting points

Organism complexity gradient argument

There appears to be a hierarchy of cognitive complexity ranging roughly down from humans to other primates to other large mammals (e.g. dogs and pigs) to smaller organisms (e.g. mice and rats) to insects to microorganisms to subcellular molecular systems to molecules to atoms to subatomic particles. At what point does consciousness end? There is no clear dividing line between “conscious” and “not conscious”. Thus, some matter may have exceedingly simple cognition, but there is likely not any matter lacking some form of primitive consciousness or experiential qualia.

Lack of real boundaries between brain, body, and environment

Approaching the complexity gradient from a different perspective, one can see that the brain is made of matter and is physically embedded in a body that is also made of matter. Furthermore, the body is physically embedded in an environment made of matter. Where does the “conscious part” end? How does one distinguish between atoms at the edge of the brain which may be thought to participate in conscious processes and atoms at the edge of the pia mater which some may argue do not participate in conscious processes? Furthermore, if a properly configured brain-brain interface was built (think of an electronic cord that physically bridges two people’s brains), it is highly plausible that the two people would experience some of each other’s qualia. Thus, the only real barrier to conscious experiences “spreading” between different organisms seems to be the accuracy of data transfer.

Dualism/supernatural is implied if panpsychism is false

Assume that panpsychism is false. Also assume that the complexity gradient argument holds, implying that there is no clear boundary between conscious and non-conscious material. In this case, where would consciousness exist? It would need to either occupy a sharply defined subset of the universe or need to exist outside of the material universe (i.e. as a supernatural force). But if the gradient argument does not allow us to define a specific subset where consciousness exists, then the only option remaining is for consciousness to exist outside of the material universe. If monist physicalism holds, then panpsychism must be true.

Panpsychism may address the hard problem through physical equivalency

The hard problem confronts many theories of consciousness. In one form of the hard problem, there is the argument that you could have complete mechanistic understanding of how the brain gives rise to a given conscious percept such as the percept of seeing the color red without actually knowing anything about the subjective experience of seeing the color red. However, this may not be true if everything is conscious since it would not be truly possible to have a complete mechanistic understanding of a percept without being physically identical to the matter (e.g. a human) experiencing said percept. Therefore, panpsychism may at least partially address the hard problem of consciousness.

Addressing objections

Reportedly non-conscious parts of brain may actually have qualia

There is neuroscientific evidence that some parts of the brain are active during conscious processing and some parts are not active (subconscious). Yet this is based on the idea that the patient who reports conscious awareness of stimuli represents a unitary entity. What if subcomponents of the brain are instead like different “people”. Perhaps your cerebellum does experience qualia, but just doesn’t transfer most of the data needed to perceive these qualia to your prefrontal cortex. In this way, the conscious parts of the brain would be the ones that have detailed “conversations” with the part of the brain directly involved in the patient’s reporting to the examiner. This goes back to the idea that information transfer may be the only limiting factor in preventing the whole universe from acting as a hive mind. Some parts of the universe do not transfer accurate information to other parts of the universe, but this does not mean that any part of the universe is unconscious.

Anatomy of a rock’s central processing

Some contend that panpsychism is intuitively ludicrous by pointing to the idea that a rock could not possibly be conscious. But consider the following scenario. A rock is illuminated by sunlight on one half of its surface while a shadow from a tree covers the other half. The surface of the rock acts a sensory organ. The rate of diffusion of heat through the rock is governed by factors like the shapes of dense granules packed into the rock’s interior and the composition of the different parts of the rock. The interior of the rock thus acts as a cognitive processor. When the heat comes out from the shadowed side of the rock, different parts of the surface will emit heat at different rates due to the processing that happened inside the rock. The shadowed side of the rock thus acts as a motor output. Certainly, the rock may not have a very accurate model of the world or a system for remembering, predicting, and reflecting. The rock is thus unlikely to have much for self-awareness. Yet it seems plausible that the rock still experiences some form of primitive and noisy qualia. Thinking of the rock in this way makes panpsychism seem less ludicrous.

Illustration of “sensory, integrative/cognitive, and motor” processes that may occur within a rock.

Global Highlights in Neuroengineering 2005-2018

January 14, 2019January 24, 2019By logancollins No Comments

PDF version: global highlights in neuroengineering 2005-2018 – logan thrasher collins

Optogenetic stimulation using ChR2

(Boyden, Zhang, Bamberg, Nagel, & Deisseroth, 2005)

Ed Boyden, Karl Deisseroth, and colleagues developed optogenetics, a revolutionary technique for stimulating neural activity.
Optogenetics involves engineering neurons to express light-gated ion channels. The first channel used for this purpose was ChR2 (a protein originally found in bacteria which responds to blue light). In this way, a neuron exposed to an appropriate wavelength of light will be stimulated.
Over time, optogenetics has gained a place as an essential experimental tool for neuroscientists across the world. It has been expanded upon and improved in numerous ways and has even allowed control of animal behavior via implanted fiber optics and other light sources. Optogenetics may eventually be used in the development of improved brain-computer interfaces.

optogenetics

Blue Brain Project cortical column simulation

(Markram, 2006)

In the early stages of the Blue Brain Project, neuronal cell types from the layers of the rat neocortex were reconstructed. Furthermore, their electrophysiology was experimentally characterized.
Next, a virtual neocortical column with about 10,000 multicompartmental Hodgkin-Huxley-type neurons and over ten million synapses was built. Its connectivity was defined according the patterns of connectivity found in biological rats, (though this involved the numbers of inputs and outputs quantified for given cell types rather than explicit wiring). In addition, the spatial distributions of boutons forming synaptic terminals upon target cells reflected biological data.
The cortical column was emulated using the Blue Gene/L supercomputer and the dynamics of the emulation reflected its biological counterpart.

Optogenetic silencing using halorhodopsin

(Han & Boyden, 2007)

Ed Boyden continued developing optogenetic tools to manipulate neural activity. Along with Xue Han, he expressed a codon-optimized version of a bacterial halorhodopsin (along with the ChR2 protein) in neurons.
Upon exposure to yellow light, halorhodopsin pumps chloride ions into the cell, hyperpolarizing the membrane and inhibiting neural activity.
Using halorhodopsin and ChR2, neurons could be easily activated and inhibited using yellow and blue light respectively.

halorhodopsin and chr2 wavelengths

Brainbow

(Livet et al., 2007)

Lichtman and colleagues used Cre/Lox recombination tools to create genes which express a randomized set of three or more differently-colored fluorescent proteins (XFPs) in a given neuron, labeling the neuron with a unique combination of colors. About ninety distinct colors were emitted across a population of genetically modified neurons.
The detailed structures within neural tissue equipped with the Brainbow system can be imaged much more easily since neurons can be distinguished via color contrast.
As a proof-of-concept, hundreds of synaptic contacts and axonal processes were reconstructed in a selected volume of the cerebellum. Several other neural structures were also imaged using Brainbow.
The fluorescent proteins expressed by the Brainbow system are usable in vivo.

High temporal precision optogenetics

(Gunaydin et al., 2010)

Karl Deisseroth, Peter Hegemann, and colleagues used protein engineering to improve the temporal resolution of optogenetic stimulation.
Glutamic acid at position 123 in ChR2 was mutated to threonine, producing a new ion channel protein (dubbed ChETA).
The ChETA protein allows for induction of spike trains with frequencies up to 200 Hz and greatly decreases the incidence of unintended spikes. Furthermore, ChETA eliminates plateau potentials (a phenomenon which interferes with precise control of neural activity).

ultrafast optogenetics

Hippocampal prosthesis in rats

(Berger et al., 2012)

Theodore Berger and his team developed an artificial replacement for neurons which transmit information from the CA3 region to the CA1 region of the hippocampus.
This cognitive prosthesis employs recording and stimulation electrodes along with a multi-input multi-output (MIMO) model to encode the information in CA3 and transfer it to CA1.
The hippocampal prosthesis was shown to restore and enhance memory in rats as evaluated by behavioral testing and brain imaging.

In vivo superresolution microscopy for neuroimaging

(Berning, Willig, Steffens, Dibaj, & Hell, 2012)

Stefan Hell (2014 Nobel laureate in chemistry) developed stimulated emission depletion microscopy (STED), a type of superresolution fluorescence microscopy which allows imaging of synapses and dendritic spines.
STED microscopy uses a torus-shaped de-excitation laser that interferes with the excitation laser to deplete fluorescence except in a very small spot. In this way, the diffraction limit is surpassed since the resulting light illuminates extremely small regions of the sample.
Neurons in transgenic mice (equipped with glass-sealed holes in their skulls) were imaged using STED. Synapses and dendritic spines were observed up to fifteen nanometers below the surface of the brain tissue.

In vivo three-photon microscopy

(Horton et al., 2013)

Multi-photon excitation uses pulsed lasers to excite fluorophores with two or more photons of light with long wavelengths. During the excitation, the photons undergo a nonlinear recombination process, yielding a single emitted photon with a much shorter wavelength. Because the excitation photons possess long wavelengths, they can penetrate tissue much more deeply than traditional microscopy allows.
Horton and colleagues developed a three-photon excitation method to facilitate even deeper tissue penetration than the commonly used two-photon microscopic techniques.
Since three photons were involved per excitation event, even longer excitation wavelengths (about 1,700 nm) were usable, allowing the construction of a 3-dimensional image stack that reached a depth of up to 1.4 mm within the living mouse brain.
Blood vessels and RFP-labeled neurons were imaged using this approach. Furthermore, the depth was sufficient to enable imaging of neurons within the mouse hippocampus.

3-photon microscopy

Whole-brain functional recording from larval zebrafish

(Ahrens, Orger, Robson, Li, & Keller, 2013)

Laser-scanning light-sheet microscopy was used to volumetrically image the entire brains of larval zebrafish (an optically transparent organism).
The genetically encoded calcium sensor GCaMP5G facilitated functional recording at single-cell resolution from about 80% of the total neurons in the larval zebrafish brains. Computational methods were used to distinguish between individual neurons.
Populations of neurons that underwent correlated activity patterns were identifiedto show the technique’s utility for uncovering the dynamics of neural circuits. These populations included hindbrain neurons that were functionally linked to neural activity in the spinal cord and a population of neurons which showed coupled oscillations on the left and right halves.

whole-brain recording from larval zebrafish

Eyewire: crowdsourcing method for retina mapping

(Marx, 2013)

The Eyewire project was created by Sebastian Seung’s research group. It is a crowdsourcing initiative for connectomic mapping within the retina towards uncovering neural circuits involved in visual processing.
Laboratories first collect data via serial electron microscopy as well as functional data from two-photon microscopy.
In the Eyewire game, images of tissue slices are provided to players who then help reconstruct neural morphologies and circuits by “coloring in” the parts of the images which correspond to cells and stacking many images on top of each other to generate 3D maps. Artificial intelligence tools help provide initial “best guesses” and guide the players, but the people ultimately perform the task of reconstruction.
By November 2013, around 82,000 participants had played the game. Its popularity continues to grow.

eyewire

The BRAIN Initiative

(“Fact Sheet: BRAIN Initiative,” 2013)

The BRAIN Initiative (Brain Research through Advancing Innovative Technologies) provided neuroscientists with $110 million in governmental funding and $122 million in funding from private sources such as the Howard Hughes Medical Institute and the Allen Institute for Brain Science.
The BRAIN Initiative focused on funding research which develops and utilizes new technologies for functional connectomics. It helped to accelerate research on tools for decoding the mechanisms of neural circuits in order to understand and treat mental illness, neurodegenerative diseases, and traumatic brain injury.
The BRAIN Initiative emphasized collaboration between neuroscientists and physicists. It also pushed forward nanotechnology-based methods to image neural tissue, record from neurons, and otherwise collect neurobiological data.

The CLARITY method for making brains translucent

(Chung & Deisseroth, 2013)

Karl Deisseroth and colleagues developed a method called CLARITY to make samples of neural tissue optically translucent without damaging the fine cellular structures in the tissue. Using CLARITY, entire mouse brains have been turned transparent.
Mouse brains were infused with hydrogel monomers (acrylamide and bisacrylamide) as well as formaldehyde and some other compounds for facilitating crosslinking. Next, the hydrogel monomers were crosslinked by incubating the brains at 37°C. Lipids in the hydrogel-stabilized mouse brains were extracted using hydrophobic organic solvents and electrophoresis.
CLARITY allows antibody labeling, fluorescence microscopy, and other optically-dependent techniques to be used for imaging entire brains. In addition, it renders the tissue permeable to macromolecules, which broadens the types of experimental techniques that these samples can undergo (i.e. macromolecule-based stains, etc.)

X-ray microtomography used to reconstruct Drosophila brain hemisphere

(Mizutani, Saiga, Takeuchi, Uesugi, & Suzuki, 2013)

Mizutani and colleagues stained Drosophila brains with silver nitrate and tetrachloroaurate (a gold-containing compound), facilitating 3-dimensional imaging using X-ray microtomography at a voxel size of 220 × 328 × 314 nm.
To generate the X-rays, a synchrotron source was used. It should be noted that synchrotron sources require large facilities to operate.
Neuronal tracing was performed manually on the 3-dimensional X-ray images of the fly brain, a process which took about 1,700 person-hours. Some neuronal processes were too dense to be resolved, so they were “fused” into unified structures. Furthermore, some neuronal traces were fragmented and most of the cell bodies were not considered. This decreased the number of traces to one third of the estimated number of actual processes in the hemisphere.
Mizutani’s investigation represents an early effort at large-scale connectomics that sets the stage for further initiatives as neuronal tracing, sample preparation, and X-ray microtomography technologies continue to improve.

traced drosophila brain hemisphere

Telepathic rats engineered using hippocampal prosthesis

(S. Deadwyler et al., 2013)

Berger’s hippocampal prosthesis was implanted in pairs of rats. When “donor” rats were trained to perform a task, they developed neural representations (memories) which were recorded by their hippocampal prostheses.
The donor rat memories were run through the MIMO model and transmitted to the stimulation electrodes of the hippocampal prostheses implanted in untrained “recipient” rats. After receiving the memories, the recipient rats showed significant improvements on the task that they had not been trained to perform.

Integrated Information Theory 3.0

(Oizumi, Albantakis, & Tononi, 2014)

Integrated information theory (IIT) was originally proposed by Giulio Tononi in 2004. IIT is a quantitative theory of consciousness which may help explain the hard problem of consciousness.
IIT begins by assuming the following phenomenological axioms; each experience is characterized by how it differs from other experiences, an experience cannot be reduced to interdependent parts, and the boundaries which distinguish individual experiences are describable as having defined “spatiotemporal grains.”
From these phenomenological axioms and the assumption of causality, IIT identifies maximally irreducible conceptual structures (MICS) associated with individual experiences. MICS represent particular patterns of qualia that form unified percepts.
IIT also outlines a mathematical measure of an experience’s quantity. This measure is called integrated information or ϕ.

Openworm

(Szigeti et al., 2014)

The anatomical elegans connectome was originally mapped in 1976 by Albertson and Thomson. More data has since been collected on neurotransmitters, electrophysiology, cell morphology, and other characteristics.
Szigeti, Larson, and their colleagues made an online platform for crowdsourcing research on elegans computational neuroscience, with the goal of completing an entire “simulated worm.”
The group also released software called Geppetto, a program that allows users to manipulate both multicompartmental Hodgkin-Huxley models and highly efficient soft-body physics simulations (for modeling the worm’s electrophysiology and anatomy).

Expansion microscopy

(F. Chen, Tillberg, & Boyden, 2015)

The Boyden group developed expansion microscopy, a method which enlarges neural tissue samples (including entire brains) with minimal structural distortions and so facilitates superior optical visualization of the scaled-up neural microanatomy. Furthermore, expansion microscopy greatly increases the optical translucency of treated samples.
Expansion microscopy operates by infusing a swellable polymer network into brain tissue samples along with several chemical treatments to facilitate polymerization and crosslinking and then triggering expansion via dialysis in water. With 4.5-fold enlargement, expansion microscopy only distorts the tissue by about 1% (computed using a comparison between control superresolution microscopy of easily-resolvable cellular features and the expanded version).
Before expansion, samples can express various fluorescent proteins to facilitate superresolution microscopy of the enlarged tissue once the process is complete. Furthermore, expanded tissue is highly amenable to fluorescent stains and antibody-based labels.

expansion microscopy

Japan’s Brain/MINDS project

(Okano, Miyawaki, & Kasai, 2015)

In 2014, the Brain/MINDS (Brain Mapping by Integrated Neurotechnologies for Disease Studies) project was initiated to further neuroscientific understanding of the brain. This project received nearly $30 million in funding for its first year alone.
Brain/MINDS focuses on studying the brain of the common marmoset (a non-human primate abundant in Japan), developing new technologies for brain mapping, and understanding the human brain with the goal of finding new treatments for brain diseases.

The TrueNorth chip from DARPA and IBM

(Akopyan et al., 2015)

The TrueNorth neuromorphic computing chip was constructed and validated by DARPA and IBM. TrueNorth uses circuit modules which mimic neurons. Inputs to these fundamental circuit modules must overcome a threshold in order to trigger “firing.”
The chip can emulate up to a million neurons with over 250 million synapses while requiring far less power than traditional computing devices.

Human Brain Project cortical mesocircuit reconstruction and simulation

(Markram et al., 2015)

The Human Brain Project reconstructed a 0.29 mm³ region of rat cortical tissue including about 31,000 neurons and 37 million synapses based on morphological data, statistical connectivity rules (rather than exact connectivity), and other datasets. The cortical mesocircuit was emulated using the Blue Gene/Q supercomputer.
This emulation was sufficiently accurate to reproduce emergent neurological processes and yield insights on the mechanisms of their computations.

Recording from C. elegans neurons reveals motor operations

(Kato et al., 2015)

Live elegans worms were immobilized in microfluidic devices and the neurons in their head ganglia as well as some of their motor systems were imaged and recorded from using the calcium indicator GCaMP. As the C. elegans connectome is well-characterized, Kato and colleagues were able to determine the identities of most of the cells that underwent imaging (with the help of computational segmentation techniques).
Principal component analysis was used to reduce the dimensionality of the neural activity datasets since over 100 neurons per worm were recorded from simultaneously.
Next, phase space analysis was utilized to visualize the patterns formed by the recording data. Motor behaviors including dorsal turns, ventral turns, forward movements, and backward movements were found to correspond to specific sequences of neural events as uncovered by examining the patterns found in the phase plots. Further analyses revealed various insights about these brain dynamics and their relationship to motor actions.

c. elegans brain dynamics

Neural lace

(Liu et al., 2015)

Charles Lieber’s group developed a syringe-injectable electronic mesh made of submicrometer-thick wiring for neural interfacing.
The meshes were constructed using novel soft electronics for biocompatibility. Upon injection, the neural lace expands to cover and record from centimeter-scale regions of tissue.
Neural lace may allow for “invasive” brain-computer interfaces to circumvent the need for surgical implantation. Lieber has continued to develop this technology towards clinical application.

BigNeuron initiative towards standardized neuronal morphology acquisition

(Peng et al., 2015)

Because of the inconsistencies between neuronal reconstruction methods and lack of standardization found in neuronal morphology databases, BigNeuron was established as a community effort to improve the situation.
BigNeuron tests as many automated neuronal reconstruction algorithms as possible using large-scale microscopy datasets (from several types of light microscopy). It uses the Vaa3D neuronal reconstruction software as a central platform. Reconstruction algorithms are added to Vaa3D as plugins. These computational tests are performed on supercomputers.
BigNeuron aims to create a superior community-oriented neuronal morphology database, a set of greatly improved tools for neuronal reconstruction, a standardized protocol for future neuronal reconstructions, and a library of morphological feature definitions to facilitate classification.

Human telepathy during a 20 questions game

(Stocco et al., 2015)

Using an interactive question-and-answer setup, Stocco and colleagues demonstrated real-time telepathic communication between pairs of individuals via EEG and transcranial magnetic stimulation. Five pairs of participants played games of 20 questions and attempted to identify unknown objects.
EEG data were recorded from the respondent, computationally processed, and transmitted as transcranial magnetic stimulation signals into the mind (occipital lobe stimulation) of a respondent. The respondent’s answers were translated into higher-intensity transcranial magnetic stimulation pulses corresponding to “yes” answers or lower-intensity transcranial magnetic stimulation pulses corresponding to “no” answers.
When compared to control trials in which sham interfaces were used, the people using the brain-brain interfaces were significantly more successful at playing 20 questions games.

Expansion FISH

(F. Chen et al., 2016)

Boyden, Chen, Marblestone, Church, and colleagues combined fluorescent in situ hybridization (FISH) with expansion microscopy to image the spatial localization of RNA in neural tissue.
The group developed a chemical linker to covalently attach intracellular RNA to the infused polymer network used in expansion microscopy. This allowed for RNAs to maintain their relative spatial locations within each cell post-expansion.
After the tissue was enlarged, FISH was used to fluorescently label targeted RNA molecules. In this way, RNA localization was more effectively resolved.
As a proof-of-concept, expansion FISH was used to reveal the nanoscale distribution of long noncoding RNAs in nuclei as well as the locations of RNAs within dendritic spines.

Neural dust

(Seo et al., 2016)

Michel Maharbiz’s group invented implantable, ~ 1 mm biosensors for wireless neural recording and tested them in rats.
This neural dust could be miniaturized to less than 0.5 mm or even to microscale dimensions using customized electronic components.
Neural dust motes consist of two recording electrodes, a transistor, and a piezoelectric crystal.
The neural dust received external power from ultrasound. Neural signals were recorded by measuring disruptions to the piezoelectric crystal’s reflection of the ultrasound waves. Signal processing mathematics allowed precise detection of activity.

The China Brain Project

(Poo et al., 2016)

The China Brain Project was launched to help understand the neural mechanisms of cognition, develop brain research technology platforms, develop preventative and diagnostic interventions for brain disorders, and to improve brain-inspired artificial intelligence technologies.
This project will be take place from 2016 until 2030 with the goal of completing mesoscopic brain circuit maps.
China’s population of non-human primates and preexisting non-human primate research facilities give the China Brain Project an advantage. The project will focus on studying rhesus macaques.

Somatosensory cortex stimulation for spinal cord injuries

(Flesher et al., 2016)

Gaunt, Flesher, and colleagues found that microstimulation of the primary somatosensory cortex (S1) partially restored tactile sensations to a patient with a spinal cord injury.
Electrode arrays were implanted into the S1 regions of a patient with a spinal cord injury. The array performed intracortical microstimulation over a period of six months.
The patient reported locations and perceptual qualities of the sensations elicited by microstimulation. The patient did not experience pain or “pins and needles” from any of the stimulus trains. Overall, 93% of the stimulus trains were reported as “possibly natural.”
Results from this study might be used to engineer upper-limb neuroprostheses which provide somatosensory feedback.

Simulation of rat CA1 region

(Bezaire, Raikov, Burk, Vyas, & Soltesz, 2016)

Detailed computational models of 338,740 neurons (including pyramidal cells and various types of interneurons) were equipped with connectivity patterns based on data from the biological CA1 region. External inputs were also estimated using biological data and incorporated into the simulation. It is important to note that these connectivity patterns described the typical convergence and divergence of neurites to and from particular cell types rather than explicitly representing the exact connections found in the biological rat.
Each neuron was simulated using a multicompartmental Hodgkin-Huxley-type model with its morphological structure based on biological data from the given cell type. Furthermore, different cell types received different numbers of presynaptic terminals at specified distances from the soma. In total, over five billion synapses were present within the CA1 model.
The simulation was implemented on several different supercomputers. Due to the model’s complexity, a four second simulation took about four hours to complete.
As with the biological CA1 region, the simulation gave rise to gamma oscillations and theta oscillations as well as other biologically consistent phenomena. In addition, parvalbumin-expressing interneurons and neurogliaform cells were identified as drivers of the theta oscillations, demonstrating the utility of detailed neuronal simulations for uncovering biological insights.

ca1 simulation

UltraTracer enhances existing neuronal tracing software

(Peng et al., 2017)

UltraTracer is an algorithm that can improve the efficiency of existing neuronal tracing software for handling large datasets while maintaining accuracy.
Datasets with hundreds of billions of voxels were utilized to test UltraTracer. Ten existing tracing algorithms were augmented.
For most of the existing algorithms, the performance improvements were around 3-6 times, though a few showed improvements of 10-30 times. Even when using computers with smaller memory, UltraTracer was consistently able to enhance conventional software.
UltraTracer was made opensource and is available as a plugin for the Vaa3D tracing software suite.

Whole-brain electron microscopy in larval zebrafish

(Hildebrand et al., 2017)

Serial electron microscopy facilitated imaging of the entire brain of a larval zebrafish at 5.5 days post-fertilization.
Neuronal tracing software (a modified version of the CATMAID software) was used to reconstruct all the myelinated axons found in the larval zebrafish brain.
The reconstructed dataset included 2,589 myelinated axon segments along with some of the associated soma and dendrites. It should be noted that only 834 of the myelinated axons were successfully traced back to their cell bodies.

Hippocampal prosthesis in monkeys

(S. A. Deadwyler et al., 2017)

Theodore Berger continued developing his cognitive prosthesis and tested it in Rhesus Macaques.
As with the rats, monkeys with the implant showed substantially improved performance on memory tasks.

The $100 billion Softbank Vision Fund

(Lomas, 2017)

Masayoshi Son, the CEO of Softbank (a Japanese telecommunications corporation), announced a plan to raise $100 billion in venture capital to invest in artificial intelligence. This plan involved partnering with multiple large companies in order to raise this enormous amount of capital.
By the end of 2017, the Vision Fund successfully reached its $100 billion goal. Masayoshi Son has since announced further plans to continue raising money with a new goal of over $800 billion.
Masayoshi Son’s reason for these massive investments is the Technological Singularity. He agrees with Kurzweil that the Singularity will likely occur at around 2045 and he hopes to help bring the Singularity to fruition. Though Son is aware of the risks posed by artificial superintelligence, he feels that superintelligent AI’s potential to tackle some of humanity’s greatest challenges (such as climate change and the threat of nuclear war) outweighs those risks.

Bryan Johnson launches Kernel

(Regalado, 2017)

Entrepreneur Bryan Johnson invested $100 million to start Kernel, a neurotechnology company.
Kernel plans to develop implants that allow for recording and stimulation of large numbers of neurons at once. The company’s initial goal is to develop treatments for mental illnesses and neurodegenerative diseases. Its long-term goal is to enhance human intelligence.
Kernel originally partnered with Theodore Berger and intended to utilize his hippocampal prosthesis. Unfortunately, Berger and Kernel parted ways after about six months because Berger’s vision was reportedly too long-range to support a financially viable company (at least for now).
Kernel was originally a company called Kendall Research Systems. This company was started by a former member of the Boyden lab. In total, four members of Kernel’s team are former Boyden lab members.

Elon Musk launches NeuraLink

(Etherington, 2017)

Elon Musk (CEO of Tesla, SpaceX, and a number of other successful companies) initiated a neuroengineering venture called NeuraLink.
NeuraLink will begin by developing brain-computer interfaces (BCIs) for clinical applications, but the ultimate goal of the company is to enhance human cognitive abilities in order to keep up with artificial intelligence.
Though many of the details around NeuraLink’s research are not yet open to the public, it has been rumored that injectable electronics similar to Lieber’s neural lace might be involved.

Facebook announces effort to build brain-computer interfaces

(Constine, 2017)

Facebook revealed research on constructing non-invasive brain-computer interfaces (BCIs) at a company-run conference in 2017. The initiative is run by Regina Dugan, Facebook’s head of R&D at division building 8.
Facebook’s researchers are working on a non-invasive BCI which may eventually enable users to type one hundred words per minute with their thoughts alone. This effort builds on past investigations which have been used to help paralyzed patients.
The building 8 group is also developing a wearable device for “skin hearing.” Using just a series of vibrating actuators which mimic the cochlea, test subjects have so far been able to recognize up to nine words. Facebook intends to vastly expand this device’s capabilities.

DARPA funds research to develop improved brain-computer interfaces

(Hatmaker, 2017)

The U.S. government agency DARPA awarded $65 million in total funding to six research groups.
The recipients of this grant included five academic laboratories (headed by Arto Nurmikko, Ken Shepard, Jose-Alain Sahel and Serge Picaud, Vicent Pieribone, and Ehud Isacoff) and one small company called Paradromics Inc.
DARPA’s goal for this initiative is to develop a nickel-sized bidirectional brain-computer interface (BCI) which can record from and stimulate up to one million individual neurons at once.

Human Brain Project analyzes brain computations using algebraic topology

(Reimann et al., 2017)

Investigators at the Human Brain Project utilized algebraic topology to analyze the reconstructed ~ 31,000 neuron cortical microcircuit from their earlier work.
The analysis involved representing the cortical network as a digraph, finding directed cliques (complete directed subgraphs belonging to a digraph), and determining the net directionality of information flow (by computing the sum of the squares of the differences between in-degree and out-degree for all the neurons in a clique). In algebraic topology, directed cliques of n neurons are called directed simplices of dimension n-1.
Vast numbers of high-dimensional directed cliques were found in the cortical microcircuit (as compared to null models and other controls). Spike correlations between pairs of neurons within a clique were found to increase with the clique’s dimension and with the proximity of the neurons to the clique’s sink. Furthermore, topological metrics allowed insights into the flow of neural information among multiple cliques.
Experimental patch-clamp data supported the significance of the findings. In addition, similar patterns were found within the elegans connectome, suggesting that the results may generalize to nervous systems across species.

hbp algebraic topology

Early testing of hippocampal prosthesis algorithm in humans

(Song, She, Hampson, Deadwyler, & Berger, 2017)

Dong Song (who was working alongside Berger) tested the MIMO algorithm on human epilepsy patients using implanted recording and stimulation electrodes. The full hippocampal prosthesis was not implanted, but the electrodes acted similarly, though in a temporary capacity. Although only two patients were tested in this study, many trials were performed to compensate for the small sample size.
Hippocampal spike trains from individual cells in CA1 and CA3 were recorded from the patients during a delayed match-to-sample task. The patients were shown various images while neural activity data were recorded by the electrodes and processed by the MIMO model. The patients were then asked to recall which image they had been shown previously by picking it from a group of “distractor” images. Memories encoded by the MIMO model were used to stimulate hippocampal cells during the recall phase.
In comparison to controls in which the same two epilepsy patients were not assisted by the algorithm and stimulation, the experimental trials demonstrated a significant increase in successful pattern matching.

Brain imaging factory in China

(Cyranoski, 2017)

Qingming Luo started the HUST-Suzhou Institute for Brainsmatics, a brain imaging “factory.” Each of the numerous machines in Luo’s facility performs automated processing and imaging of tissue samples. The devices make ultrathin slices of brain tissue using diamond blades, treat the samples with fluorescent stains or other contrast-enhancing chemicals, and image then using fluorescence microscopy.
The institute has already demonstrated its potential by mapping the morphology of a previously unknown neuron which “wraps around” the entire mouse brain.

china brain mapping image

Automated patch-clamp robot for in vivo neural recording

(Suk et al., 2017)

Ed Boyden and colleagues developed a robotic system to automate patch-clamp recordings from individual neurons. The robot was tested in vivo using mice and achieved a data collection yield similar to that of skilled human experimenters.
By continuously imaging neural tissue using two-photon microscopy, the robot can adapt to a target cell’s movement and shift the pipette to compensate. This adaptation is facilitated by a novel algorithm called an imagepatching algorithm. As the pipette approaches its target, the algorithm adjusts the pipette’s trajectory based on the real-time two-photon microscopy.
The robot can be used in vivo so long as the target cells express a fluorescent marker or otherwise fluoresce corresponding to their size and position.

automated patch clamp system

Genome editing in the mammalian brain

(Nishiyama, Mikuni, & Yasuda, 2017)

Precise genome editing in the brain has historically been challenging because most neurons are postmitotic (non-dividing) and the postmitotic state prevents homology-directed repair (HDR) from occurring. HDR is a mechanism of DNA repair which allows for targeted insertions of DNA fragments with overhangs homologous to the region of interest (by contrast, non-homologous end-joining is highly unpredictable).
Nishiyama, Mikuni, and Yasuda developed a technique which allows genome editing in postmitotic mammalian neurons using adeno-associated viruses (AAVs) and CRISPR-Cas9.
The AAVs delivered ssDNA sequences encoding a single guide RNA (sgRNA) and an insert. Inserts encoding a hemagglutinin tag (HA) and inserts encoding EGFP were both tested. Cas9 was encoded endogenously by transgenic host cells and in transgenic host animals.
The technique achieved precise genome editing in vitro and in vivo with a low rate of off-target effects. Inserts did not cause deletion of nearby endogenous sequences for 98.1% of infected neurons.

Neuropixels probe

(Jun et al., 2017)

Jun and colleagues created the Neuropixels probe to facilitate simultaneous recording from hundreds of individual neurons with high spatiotemporal resolution. Previous extracellular probes were only able to record from a few dozen individual neurons.
The Neuropixels recording shank is one centimeter long and includes 384 recording channels. Due to the small size of the accompanying apparatus (a 6×9 mm base and a data transmission cable), it enables high-throughput recording in freely moving animals. Because the shank is quite long, Neuropixels can record from multiple brain regions at once.
Voltage signals are processed directly on the base of the Neuropixels apparatus, allowing for noise-free data transmission along the cable for further analysis.

neuropixels

EEG-based facial image reconstruction

(Nemrodov, Niemeier, Patel, & Nestor, 2018)

EEG data associated with viewing images of faces was collected and used to determine the neural correlates of facial processing. In this way, the images were computationally reconstructed in a fashion resembling “mind reading.”
It should be noted that the images reconstructed using data taken from multiple people were more accurate than the images reconstructed using single individuals. Nonetheless, the single individual data still yielded statistically significant accuracy.
In addition to reconstructing the images themselves, the process gave insights on the cognitive steps involved in perceiving faces.

eeg reconstructions of faces

Near-infrared light and upconversion nanoparticles for optogenetic stimulation

(S. Chen et al., 2018)

Upconversion nanoparticles absorb two or more low-energy photons and emit a higher energy photon. For instance, multiple near-infrared photons can be converted into a single visible spectrum photon.
Shuo Chen and colleagues injected upconversion nanoparticles into the brains of mice and used them to convert externally applied near-infrared (NIR) light into visible light within the brain tissue. In this way, optogenetic stimulation was performed without the need for surgical implantation of fiber optics or similarly invasive procedures.
The authors demonstrated stimulation via upconversion of NIR to blue light (to activate ChR2) and inhibition via upconversion of NIR to green light (to activate a rhodopsin called Arch).
As a proof-of-concept, this technology was used to alter the behavior of the mice by activating hippocampally-encoded fear memories.

upconversion nanoparticles and nir

Map of all neuronal cell bodies within mouse brain

(Murakami et al., 2018)

Ueda, Murakami, and colleagues combined methods from expansion microscopy and CLARITY to develop a protocol called CUBIC-X which both expands and clears entire brains. Light-sheet fluorescence microscopy was used to image the treated brains and a novel algorithm was developed to detect individual nuclei.
Although expansion microscopy causes some increased tissue transparency on its own, CUBIC-X greatly improved this property in the enlarged tissues, facilitating more detailed whole-brain imaging.
Using CUBIC-X, the spatial locations of all the cell bodies (but not dendrites, axons, or synapses) within the mouse brain were mapped. This process was performed upon several adult mouse brains as well as several developing mouse brains to allow for comparative analysis.
The authors made the spatial atlas publicly available in order to facilitate global cooperation towards annotating connectivity among the neural cell bodies within the atlas.

cubic-x

Clinical testing of hippocampal prosthesis algorithm in humans

(Hampson et al., 2018)

Further clinical tests of Berger’s hippocampal prosthesis were performed. Twenty-one patients took part in the experiments. Seventeen patients underwent CA3 recording so as to facilitate training and optimization of the MIMO model. Eight patients received CA1 stimulation so as to improve their memories.
Electrodes with the ability to record from single neurons (10-24 single-neuron recording sites) and via EEG (4-6 EEG recording sites) were implanted such that recording and stimulation could occur at CA3 and CA1 respectively.
Patients performed behavioral memory tasks. Both short-term and long-term memory showed an average improvement of 35% across the patients who underwent stimulation.

Precise optogenetic manipulation of fifty neurons

(Mardinly et al., 2018)

Mardinly and colleagues engineered a novel excitatory optogenetic ion channel called ST-ChroME and a novel inhibitory optogenetic ion channel called IRES-ST-eGtACR1. The channels were localized to the somas of host neurons and generated stronger photocurrents over shorter timescales than previously existing opsins, allowing for powerful and precise optogenetic stimulation and inhibition.
3D-SHOT is an optical technique in which light is tuned by a device called a spatial light modulator along with several other optical components. Using 3D-SHOT, light was precisely projected upon targeted neurons within a volume of 550×550×100 μm³.
By combining novel optogenetic ion channels and the 3D-SHOT technique, complex patterns of neural activity were created in vivo with high spatial and temporal precision.
Simultaneously, calcium imaging allowed measurement of the induced neural activity. More custom optoelectronic components helped avoid optical crosstalk of the fluorescent calcium markers with the photostimulating laser.

Whole-brain Drosophila connectome data acquired via serial electron microscopy

(Zheng et al., 2018)

Zheng, Bock, and colleagues collected serial electron microscopy data on the entire adult Drosophila connectome, providing the data necessary to reconstruct a complete structural map of the fly’s brain at the resolution of individual synapses, dendritic spines, and axonal processes.
The data are in the form of 7050 transmission electron microscopy images (187500 x 87500 pixels and 16 GB per image), each representing a 40nm-thin slice of the fly’s brain. In total the dataset requires 106 TB of storage.
Although much of the the data still must be processed to reconstruct a 3-dimensional map of the Drosophila brain, the authors did create 3-dimensional reconstructions of selected areas in the olfactory pathway of the fly. In doing so, they discovered a new cell type as well as several other previously unrealized insights about the organization of Drosophila’s olfactory biology.

Human telepathy using BrainNet

(Jiang et al., 2018)

EEG recordings were taken from two individuals (termed senders) while they played a Tetris-like game. Next, the recordings were converted into transcranial magnetic stimulation signals that acted to provide a third individual (called a receiver) with the necessary information to make decisions in the game without seeing the screen. The occipital cortex was stimulated. Fifteen people (five groups of three) took part in the study.
To convey their information, the senders were told to focus upon either a higher or a lower intensity light corresponding to commands within the game (the two lights were placed on different sides of the computer screen). In the receiver’s mind, this translated to perceiving a flash of light. The receiver was able to distinguish the intensities and implement the correct command within the game.
Using only the telepathically provided stimulation, the receiver made the correct game-playing decisions 81% of the time.

Transcriptomic cell type classification across mouse neocortex

(Tasic et al., 2018)

Single-cell RNA sequencing was used to characterize gene expression across 23,822 cells from the primary visual cortex and the anterior lateral motor cortex of mice.
Using dimensionality reduction and clustering methods, the resulting data were used to classify the neurons into 133 transcriptomic cell types.
Injections of adeno associated viruses (engineered to express fluorescent markers) facilitated retrograde tracing of neuronal projections within a subset of the sequenced cells. In this way, correspondences between projection patterns and transcriptomic identities were established.

References

Ahrens, M. B., Orger, M. B., Robson, D. N., Li, J. M., & Keller, P. J. (2013). Whole-brain functional imaging at cellular resolution using light-sheet microscopy. Nature Methods, 10, 413. Retrieved from https://doi.org/10.1038/nmeth.2434

Akopyan, F., Sawada, J., Cassidy, A., Alvarez-Icaza, R., Arthur, J., Merolla, P., … Modha, D. S. (2015). TrueNorth: Design and Tool Flow of a 65 mW 1 Million Neuron Programmable Neurosynaptic Chip. IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, 34(10), 1537–1557. https://doi.org/10.1109/TCAD.2015.2474396

Berger, T. W., Song, D., Chan, R. H. M., Marmarelis, V. Z., LaCoss, J., Wills, J., … Granacki, J. J. (2012). A Hippocampal Cognitive Prosthesis: Multi-Input, Multi-Output Nonlinear Modeling and VLSI Implementation. IEEE Transactions on Neural Systems and Rehabilitation Engineering, 20(2), 198–211. https://doi.org/10.1109/TNSRE.2012.2189133

Berning, S., Willig, K. I., Steffens, H., Dibaj, P., & Hell, S. W. (2012). Nanoscopy in a Living Mouse Brain. Science, 335(6068), 551 LP-551. Retrieved from http://science.sciencemag.org/content/335/6068/551.abstract

Bezaire, M. J., Raikov, I., Burk, K., Vyas, D., & Soltesz, I. (2016). Interneuronal mechanisms of hippocampal theta oscillations in a full-scale model of the rodent CA1 circuit. ELife, 5, e18566. https://doi.org/10.7554/eLife.18566

Boyden, E. S., Zhang, F., Bamberg, E., Nagel, G., & Deisseroth, K. (2005). Millisecond-timescale, genetically targeted optical control of neural activity. Nature Neuroscience, 8, 1263. Retrieved from http://dx.doi.org/10.1038/nn1525

Chen, F., Tillberg, P. W., & Boyden, E. S. (2015). Expansion microscopy. Science, 347(6221), 543 LP-548. Retrieved from http://science.sciencemag.org/content/347/6221/543.abstract

Chen, F., Wassie, A. T., Cote, A. J., Sinha, A., Alon, S., Asano, S., … Boyden, E. S. (2016). Nanoscale imaging of RNA with expansion microscopy. Nature Methods, 13, 679. Retrieved from http://dx.doi.org/10.1038/nmeth.3899

Chen, S., Weitemier, A. Z., Zeng, X., He, L., Wang, X., Tao, Y., … McHugh, T. J. (2018). Near-infrared deep brain stimulation via upconversion nanoparticle–mediated optogenetics. Science, 359(6376), 679 LP-684. Retrieved from http://science.sciencemag.org/content/359/6376/679.abstract

Chung, K., & Deisseroth, K. (2013). CLARITY for mapping the nervous system. Nature Methods, 10, 508. Retrieved from http://dx.doi.org/10.1038/nmeth.2481

Constine, J. (2017). Facebook is building brain-computer interfaces for typing and skin-hearing. TechCrunch. Retrieved from https://techcrunch.com/2017/04/19/facebook-brain-interface/

Cyranoski, D. (2017). China launches brain-imaging factory. Nature, 548(7667), 268–269. https://doi.org/10.1038/548268a

Deadwyler, S. A., Hampson, R. E., Song, D., Opris, I., Gerhardt, G. A., Marmarelis, V. Z., & Berger, T. W. (2017). A cognitive prosthesis for memory facilitation by closed-loop functional ensemble stimulation of hippocampal neurons in primate brain. Experimental Neurology, 287, 452–460. https://doi.org/https://doi.org/10.1016/j.expneurol.2016.05.031

Deadwyler, S., Hampson, R., Sweat, A., Song, D., Chan, R., Opris, I., … Berger, T. (2013). Donor/recipient enhancement of memory in rat hippocampus. Frontiers in Systems Neuroscience. Retrieved from https://www.frontiersin.org/article/10.3389/fnsys.2013.00120

Etherington, D. (2017). Elon Musk’s Neuralink wants to boost the brain to keep up with AI. TechCrunch. Retrieved from techcrunch.com/2017/03/27/elon-musks-neuralink-wants-to-boost-the-brain-to-keep-up-with-ai/

Fact Sheet: BRAIN Initiative. (2013). Retrieved from https://obamawhitehouse.archives.gov/the-press-office/2013/04/02/fact-sheet-brain-initiative

Flesher, S. N., Collinger, J. L., Foldes, S. T., Weiss, J. M., Downey, J. E., Tyler-Kabara, E. C., … Gaunt, R. A. (2016). Intracortical microstimulation of human somatosensory cortex. Science Translational Medicine. Retrieved from http://stm.sciencemag.org/content/early/2016/10/12/scitranslmed.aaf8083.abstract

Gunaydin, L. A., Yizhar, O., Berndt, A., Sohal, V. S., Deisseroth, K., & Hegemann, P. (2010). Ultrafast optogenetic control. Nature Neuroscience, 13, 387. Retrieved from http://dx.doi.org/10.1038/nn.2495

Hampson, R. E., Song, D., Robinson, B. S., Fetterhoff, D., Dakos, A. S., Roeder, B. M., … Deadwyler, S. A. (2018). Developing a hippocampal neural prosthetic to facilitate human memory encoding and recall. Journal of Neural Engineering, 15(3), 36014. https://doi.org/10.1088/1741-2552/aaaed7

Han, X., & Boyden, E. S. (2007). Multiple-Color Optical Activation, Silencing, and Desynchronization of Neural Activity, with Single-Spike Temporal Resolution. PLOS ONE, 2(3), e299. Retrieved from https://doi.org/10.1371/journal.pone.0000299

Hatmaker, T. (2017). DARPA awards $65 million to develop the perfect, tiny two-way brain-computer interface. TechCrunch. Retrieved from techcrunch.com/2017/07/10/darpa-nesd-grants-paradromics/

Hildebrand, D. G. C., Cicconet, M., Torres, R. M., Choi, W., Quan, T. M., Moon, J., … Engert, F. (2017). Whole-brain serial-section electron microscopy in larval zebrafish. Nature, 545, 345. Retrieved from https://doi.org/10.1038/nature22356

Horton, N. G., Wang, K., Kobat, D., Clark, C. G., Wise, F. W., Schaffer, C. B., & Xu, C. (2013). In vivo three-photon microscopy of subcortical structures within an intact mouse brain. Nature Photonics, 7, 205. Retrieved from https://doi.org/10.1038/nphoton.2012.336

Jiang, L., Stocco, A., Losey, D. M., Abernethy, J. A., Prat, C. S., & Rao, R. P. N. (2018). BrainNet: a multi-person brain-to-brain interface for direct collaboration between brains. ArXiv Preprint ArXiv:1809.08632.

Jun, J. J., Steinmetz, N. A., Siegle, J. H., Denman, D. J., Bauza, M., Barbarits, B., … Harris, T. D. (2017). Fully integrated silicon probes for high-density recording of neural activity. Nature, 551, 232. Retrieved from https://doi.org/10.1038/nature24636

Kato, S., Kaplan, H. S., Schrödel, T., Skora, S., Lindsay, T. H., Yemini, E., … Zimmer, M. (2015). Global Brain Dynamics Embed the Motor Command Sequence of Caenorhabditis elegans. Cell, 163(3), 656–669. https://doi.org/10.1016/j.cell.2015.09.034

Liu, J., Fu, T.-M., Cheng, Z., Hong, G., Zhou, T., Jin, L., … Lieber, C. M. (2015). Syringe-injectable electronics. Nature Nanotechnology, 10, 629. Retrieved from http://dx.doi.org/10.1038/nnano.2015.115

Livet, J., Weissman, T. A., Kang, H., Draft, R. W., Lu, J., Bennis, R. A., … Lichtman, J. W. (2007). Transgenic strategies for combinatorial expression of fluorescent proteins in the nervous system. Nature, 450, 56. Retrieved from http://dx.doi.org/10.1038/nature06293

Lomas, N. (2017). Superintelligent AI explains Softbank’s push to raise a $100BN Vision Fund. TechCrunch. Retrieved from https://techcrunch.com/2017/02/27/superintelligent-ai-explains-softbanks-push-to-raise-a-100bn-vision-fund/

Mardinly, A. R., Oldenburg, I. A., Pégard, N. C., Sridharan, S., Lyall, E. H., Chesnov, K., … Adesnik, H. (2018). Precise multimodal optical control of neural ensemble activity. Nature Neuroscience, 21(6), 881–893. https://doi.org/10.1038/s41593-018-0139-8

Markram, H. (2006). The Blue Brain Project. Nature Reviews Neuroscience, 7, 153. Retrieved from http://dx.doi.org/10.1038/nrn1848

Markram, H., Muller, E., Ramaswamy, S., Reimann, M. W., Abdellah, M., Sanchez, C. A., … Schürmann, F. (2015). Reconstruction and Simulation of Neocortical Microcircuitry. Cell, 163(2), 456–492. https://doi.org/10.1016/j.cell.2015.09.029

Marx, V. (2013). Neuroscience waves to the crowd. Nature Methods, 10, 1069. Retrieved from http://dx.doi.org/10.1038/nmeth.2695

Mizutani, R., Saiga, R., Takeuchi, A., Uesugi, K., & Suzuki, Y. (2013). Three-dimensional network of Drosophila brain hemisphere. Journal of Structural Biology, 184(2), 271–279. https://doi.org/https://doi.org/10.1016/j.jsb.2013.08.012

Murakami, T. C., Mano, T., Saikawa, S., Horiguchi, S. A., Shigeta, D., Baba, K., … Ueda, H. R. (2018). A three-dimensional single-cell-resolution whole-brain atlas using CUBIC-X expansion microscopy and tissue clearing. Nature Neuroscience, 21(4), 625–637. https://doi.org/10.1038/s41593-018-0109-1

Nemrodov, D., Niemeier, M., Patel, A., & Nestor, A. (2018). The Neural Dynamics of Facial Identity Processing: Insights from EEG-Based Pattern Analysis and Image Reconstruction. Eneuro, 5(1), ENEURO.0358-17.2018. https://doi.org/10.1523/ENEURO.0358-17.2018

Nishiyama, J., Mikuni, T., & Yasuda, R. (2017). Virus-Mediated Genome Editing via Homology-Directed Repair in Mitotic and Postmitotic Cells in Mammalian Brain. Neuron, 96(4), 755–768.e5. https://doi.org/10.1016/j.neuron.2017.10.004

Oizumi, M., Albantakis, L., & Tononi, G. (2014). From the Phenomenology to the Mechanisms of Consciousness: Integrated Information Theory 3.0. PLOS Computational Biology, 10(5), e1003588. Retrieved from https://doi.org/10.1371/journal.pcbi.1003588

Okano, H., Miyawaki, A., & Kasai, K. (2015). Brain/MINDS: brain-mapping project in Japan. Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences, 370(1668). https://doi.org/10.1098/rstb.2014.0310

Peng, H., Hawrylycz, M., Roskams, J., Hill, S., Spruston, N., Meijering, E., & Ascoli, G. A. (2015). BigNeuron: Large-Scale 3D Neuron Reconstruction from Optical Microscopy Images. Neuron, 87(2), 252–256. https://doi.org/10.1016/j.neuron.2015.06.036

Peng, H., Zhou, Z., Meijering, E., Zhao, T., Ascoli, G. A., & Hawrylycz, M. (2017). Automatic tracing of ultra-volumes of neuronal images. Nature Methods, 14, 332. Retrieved from https://doi.org/10.1038/nmeth.4233

Poo, M., Du, J., Ip, N. Y., Xiong, Z.-Q., Xu, B., & Tan, T. (2016). China Brain Project: Basic Neuroscience, Brain Diseases, and Brain-Inspired Computing. Neuron, 92(3), 591–596. https://doi.org/10.1016/j.neuron.2016.10.050

Regalado, A. (2017). The Entrepreneur with the $100 Million Plan to Link Brains to Computers. MIT Technology Review. Retrieved from https://www.technologyreview.com/s/603771/the-entrepreneur-with-the-100-million-plan-to-link-brains-to-computers/

Reimann, M. W., Nolte, M., Scolamiero, M., Turner, K., Perin, R., Chindemi, G., … Markram, H. (2017). Cliques of Neurons Bound into Cavities Provide a Missing Link between Structure and Function. Frontiers in Computational Neuroscience. Retrieved from https://www.frontiersin.org/article/10.3389/fncom.2017.00048

Seo, D., Neely, R. M., Shen, K., Singhal, U., Alon, E., Rabaey, J. M., … Maharbiz, M. M. (2016). Wireless Recording in the Peripheral Nervous System with Ultrasonic Neural Dust. Neuron, 91(3), 529–539. https://doi.org/10.1016/j.neuron.2016.06.034

Song, D., She, X., Hampson, R. E., Deadwyler, S. A., & Berger, T. W. (2017). Multi-resolution multi-trial sparse classification model for decoding visual memories from hippocampal spikes in human. In 2017 39th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC) (pp. 1046–1049). https://doi.org/10.1109/EMBC.2017.8037006

Stocco, A., Prat, C. S., Losey, D. M., Cronin, J. A., Wu, J., Abernethy, J. A., & Rao, R. P. N. (2015). Playing 20 Questions with the Mind: Collaborative Problem Solving by Humans Using a Brain-to-Brain Interface. PLOS ONE, 10(9), e0137303. Retrieved from https://doi.org/10.1371/journal.pone.0137303

Suk, H.-J., van Welie, I., Kodandaramaiah, S. B., Allen, B., Forest, C. R., & Boyden, E. S. (2017). Closed-Loop Real-Time Imaging Enables Fully Automated Cell-Targeted Patch-Clamp Neural Recording In Vivo. Neuron, 95(5), 1037–1047.e11. https://doi.org/https://doi.org/10.1016/j.neuron.2017.08.011

Szigeti, B., Gleeson, P., Vella, M., Khayrulin, S., Palyanov, A., Hokanson, J., … Larson, S. (2014). OpenWorm: an open-science approach to modeling Caenorhabditis elegans. Frontiers in Computational Neuroscience. Retrieved from https://www.frontiersin.org/article/10.3389/fncom.2014.00137

Tasic, B., Yao, Z., Graybuck, L. T., Smith, K. A., Nguyen, T. N., Bertagnolli, D., … Zeng, H. (2018). Shared and distinct transcriptomic cell types across neocortical areas. Nature, 563(7729), 72–78. https://doi.org/10.1038/s41586-018-0654-5

Zheng, Z., Lauritzen, J. S., Perlman, E., Robinson, C. G., Nichols, M., Milkie, D., … Bock, D. D. (2018). A Complete Electron Microscopy Volume of the Brain of Adult Drosophila melanogaster. Cell, 174(3), 730–743.e22. https://doi.org/10.1016/j.cell.2018.06.019