Facial and Vocal Markers of Schizophrenia Measured Using Remote Smartphone Assessments: Observational Study.

BACKGROUND: Machine learning-based facial and vocal measurements have demonstrated relationships with schizophrenia diagnosis and severity. Demonstrating utility and validity of remote and automated assessments conducted outside of controlled experimental or clinical settings can facilitate scaling such measurement tools to aid in risk assessment and tracking of treatment response in populations that are difficult to engage. OBJECTIVE: This study aimed to determine the accuracy of machine learning-based facial and vocal measurements acquired through automated assessments conducted remotely through smartphones. METHODS: Measurements of facial and vocal characteristics including facial expressivity, vocal acoustics, and speech prevalence were assessed in 20 patients with schizophrenia over the course of 2 weeks in response to two classes of prompts previously utilized in experimental laboratory assessments: evoked prompts, where subjects are guided to produce specific facial expressions and speech; and spontaneous prompts, where subjects are presented stimuli in the form of emotionally evocative imagery and asked to freely respond. Facial and vocal measurements were assessed in relation to schizophrenia symptom severity using the Positive and Negative Syndrome Scale. RESULTS: Vocal markers including speech prevalence, vocal jitter, fundamental frequency, and vocal intensity demonstrated specificity as markers of negative symptom severity, while measurement of facial expressivity demonstrated itself as a robust marker of overall schizophrenia symptom severity. CONCLUSIONS: Established facial and vocal measurements, collected remotely in schizophrenia patients via smartphones in response to automated task prompts, demonstrated accuracy as markers of schizophrenia symptom severity. Clinical implications are discussed.

Digitally Enabled, Patient-Centric Clinical Trials: Shifting the Drug Development Paradigm.

The rapidly advancing field of digital health technologies provides a great opportunity to radically transform the way clinical trials are conducted and to shift the clinical trial paradigm from a site-centric to a patient-centric model. Merck's (Kenilworth, NJ) digitally enabled clinical trial initiative is focused on introduction of digital technologies into the clinical trial paradigm to reduce patient burden, improve drug adherence, provide a means of more closely engaging with the patient, and enable higher quality, faster, and more frequent data collection. This paper will describe the following four key areas of focus from Merck's digitally enabled clinical trials initiative, along with corresponding enabling technologies: (i) use of technologies that can monitor and improve drug adherence (smart dosing), (ii) collection of pharmacokinetic (PK), pharmacodynamic (PD), and biomarker samples in an outpatient setting (patient-centric sampling), (iii) use of digital devices to collect and measure physiological and behavioral data (digital biomarkers), and (iv) use of data platforms that integrate digital data streams, visualize data in real-time, and provide a means of greater patient engagement during the trial (digital platform). Furthermore, this paper will discuss the synergistic power in implementation of these approaches jointly within a trial to enable better understanding of adherence, safety, efficacy, PK, PD, and corresponding exposure-response relationships of investigational therapies as well as reduced patient burden for clinical trial participation. Obstacle and challenges to adoption and full realization of the vision of patient-centric, digitally enabled trials will also be discussed.

A Deep Learning Approach for Automated Sleep-Wake Scoring in Pre-Clinical Animal Models.

BACKGROUND: Experimental investigation of sleep-wake dynamics in animals is an important part of pharmaceutical development. Typically, it involves recording of electroencephalogram, electromyogram, locomotor activity, and electrooculogram. Visual identification, or scoring, of the sleep-wake states from these recordings is time-consuming. We sought to develop software for automated sleep-wake scoring capable of processing large databases of multi-channel signal recordings in a range of species. NEW METHOD: We used a large historical database of signal recordings and scores in non-human primates, dogs, mice, and rats, to develop a deep Convolutional Neural Network (CNN) classification algorithm for automatically scoring sleep-wake states. We compared the performance of the CNN algorithm with that of a widely used Machine Learning algorithm, Random Forest (RF). RESULTS: CNN accuracy in sleep-wake scoring of data in non-human primates and dogs was significantly higher than RF accuracy (0.75 vs. 0.66 for non-human primates and 0.73 vs. 0.64 for dogs). In rodents, the difference between CNN and RF was smaller: 0.83 vs. 0.81 for mice and 0.78 vs. 0.77 for rats. The variability of CNN accuracy was lower than that of RF for non-human primates, dogs and mice but similar for rats. COMPARISON WITH EXISTING METHODS: Deep Learning algorithms have not been previously evaluated across a range of species for animal sleep-wake scoring. CONCLUSIONS: We recommend use of CNN for sleep-wake scoring in non-human primates and dogs, and RF for sleep-wake scoring in rodents.

Molecular Fingerprinting of On-Off Direction-Selective Retinal Ganglion Cells Across Species and Relevance to Primate Visual Circuits.

The ability to detect moving objects is an ethologically salient function. Direction-selective neurons have been identified in the retina, thalamus, and cortex of many species, but their homology has remained unclear. For instance, it is unknown whether direction-selective retinal ganglion cells (DSGCs) exist in primates and, if so, whether they are the equivalent to mouse and rabbit DSGCs. Here, we used a molecular/circuit approach in both sexes to address these issues. In mice, we identify the transcription factor Satb2 (special AT-rich sequence-binding protein 2) as a selective marker for three RGC types: On-Off DSGCs encoding motion in either the anterior or posterior direction, a newly identified type of Off-DSGC, and an Off-sustained RGC type. In rabbits, we find that expression of Satb2 is conserved in On-Off DSGCs; however, it has evolved to include On-Off DSGCs encoding upward and downward motion in addition to anterior and posterior motion. Next, we show that macaque RGCs express Satb2 most likely in a single type. We used rabies virus-based circuit-mapping tools to reveal the identity of macaque Satb2-RGCs and discovered that their dendritic arbors are relatively large and monostratified. Together, these data indicate Satb2-expressing On-Off DSGCs are likely not present in the primate retina. Moreover, if DSGCs are present in the primate retina, it is unlikely that they express Satb2.SIGNIFICANCE STATEMENT The ability to detect object motion is a fundamental feature of almost all visual systems. Here, we identify a novel marker for retinal ganglion cells encoding directional motion that is evolutionarily conserved in mice and rabbits, but not in primates. We show in macaque monkeys that retinal ganglion cells (RGCs) that express this marker comprise a single type and are morphologically distinct from mouse and rabbit direction-selective RGCs. Our findings indicate that On-Off direction-selective retinal neurons may have evolutionarily diverged in primates and more generally provide novel insight into the identity and organization of primate parallel visual pathways.

Representation of Rapid Image Sequences in V4 Networks.

Natural viewing often consists of sequences of brief fixations to image patches of different structure. Whether and how briefly presented sequential stimuli are encoded in a temporal-position manner is poorly understood. Here, we performed multiple-electrode recordings in the visual cortex (area V4) of nonhuman primates (Macaca mulatta) viewing a sequence of 7 briefly flashed natural images, and measured correlations between the cue-triggered population response in the presence and absence of the stimulus. Surprisingly, we found significant correlations for images occurring at the beginning and the end of a sequence, but not for those in the middle. The correlation strength increased with stimulus exposure and favored the image position in the sequence rather than image identity. These results challenge the commonly held view that images are represented in visual cortex exclusively based on their informational content, and indicate that, in the absence of sensory information, neuronal populations exhibit reactivation of stimulus-evoked responses in a way that reflects temporal position within a stimulus sequence.

Focal Changes to Human Electrocorticography With Drowsiness: A Novel Measure of Local Sleep.

Drowsiness may be defined as the progressive loss of cortical processing efficiency that occurs with time passing while awake. This loss of cortical processing efficiency is reflected in focal changes to the electroencephalogram, including islands of increased delta power concurrent with drop-offs in neuronal activity (i.e., focal cortical inactivity). The authors hypothesized that these focal changes are evidenced at individual electrodes by combination of increased instantaneous amplitude in delta band and decreased instantaneous frequency in theta-alpha band, permitting their categorization as "active" and "inactive." An analysis of records from six patients with refractory epilepsy undergoing video-electrocorticographic monitoring was conducted. Feature extraction and state classification on multiple recordings revealed focal changes consistent with the hypothesis, as well as progressively increased numbers of inactive electrodes with time awake. The implications of these findings on the study of sleep, and particularly local sleep, are discussed.

Reactivation of visual-evoked activity in human cortical networks.

In the absence of sensory input, neuronal networks are far from being silent. Whether spontaneous changes in ongoing activity reflect previous sensory experience or stochastic fluctuations in brain activity is not well understood. Here we demonstrate reactivation of stimulus-evoked activity that is distributed across large areas in the human brain. We performed simultaneous electrocorticography recordings from occipital, parietal, temporal, and frontal areas in awake humans in the presence and absence of sensory stimulation. We found that, in the absence of visual input, repeated exposure to brief natural movies induces robust stimulus-specific reactivation at individual recording sites. The reactivation sites were characterized by greater global connectivity compared with those sites that did not exhibit reactivation. Our results indicate a surprising degree of short-term plasticity across multiple networks in the human brain as a result of repeated exposure to unattended information.

A wireless transmission neural interface system for unconstrained non-human primates.

OBJECTIVE: Studying the brain in large animal models in a restrained laboratory rig severely limits our capacity to examine brain circuits in experimental and clinical applications. APPROACH: To overcome these limitations, we developed a high-fidelity 96-channel wireless system to record extracellular spikes and local field potentials from the neocortex. A removable, external case of the wireless device is attached to a titanium pedestal placed in the animal skull. Broadband neural signals are amplified, multiplexed, and continuously transmitted as TCP/IP data at a sustained rate of 24 Mbps. A Xilinx Spartan 6 FPGA assembles the digital signals into serial data frames for transmission at 20 kHz though an 802.11n wireless data link on a frequency-shift key-modulated signal at 5.7-5.8 GHz to a receiver up to 10 m away. The system is powered by two CR123A, 3 V batteries for 2 h of operation. MAIN RESULTS: We implanted a multi-electrode array in visual area V4 of one anesthetized monkey (Macaca fascicularis) and in the dorsolateral prefrontal cortex (dlPFC) of a freely moving monkey (Macaca mulatta). The implanted recording arrays were electrically stable and delivered broadband neural data over a year of testing. For the first time, we compared dlPFC neuronal responses to the same set of stimuli (food reward) in restrained and freely moving conditions. Although we did not find differences in neuronal responses as a function of reward type in the restrained and unrestrained conditions, there were significant differences in correlated activity. This demonstrates that measuring neural responses in freely moving animals can capture phenomena that are absent in the traditional head-fixed paradigm. SIGNIFICANCE: We implemented a wireless neural interface for multi-electrode recordings in freely moving non-human primates, which can potentially move systems neuroscience to a new direction by allowing one to record neural signals while animals interact with their environment.

Correlated variability in laminar cortical circuits.

Despite the fact that strong trial-to-trial correlated variability in responses has been reported in many cortical areas, recent evidence suggests that neuronal correlations are much lower than previously thought. Here, we used multicontact laminar probes to revisit the issue of correlated variability in primary visual (V1) cortical circuits. We found that correlations between neurons depend strongly on local network context--whereas neurons in the input (granular) layers showed virtually no correlated variability, neurons in the output layers (supragranular and infragranular) exhibited strong correlations. The laminar dependence of noise correlations is consistent with recurrent models in which neurons in the granular layer receive intracortical inputs from nearby cells, whereas supragranular and infragranular layer neurons receive inputs over larger distances. Contrary to expectation that the output cortical layers encode stimulus information most accurately, we found that the input network offers superior discrimination performance compared to the output networks.

Examining local network processing using multi-contact laminar electrode recording.

Cortical layers are ubiquitous structures throughout neocortex that consist of highly recurrent local networks. In recent years, significant progress has been made in our understanding of differences in response properties of neurons in different cortical layers, yet there is still a great deal left to learn about whether and how neuronal populations encode information in a laminar-specific manner. Existing multi-electrode array techniques, although informative for measuring responses across many millimeters of cortical space along the cortical surface, are unsuitable to approach the issue of laminar cortical circuits. Here, we present our method for setting up and recording individual neurons and local field potentials (LFPs) across cortical layers of primary visual cortex (V1) utilizing multi-contact laminar electrodes (Figure 1; Plextrode U-Probe, Plexon Inc). The methods included are recording device construction, identification of cortical layers, and identification of receptive fields of individual neurons. To identify cortical layers, we measure the evoked response potentials (ERPs) of the LFP time-series using full-field flashed stimuli. We then perform current-source density (CSD) analysis to identify the polarity inversion accompanied by the sink-source configuration at the base of layer 4 (the sink is inside layer 4, subsequently referred to as granular layer). Current-source density is useful because it provides an index of the location, direction, and density of transmembrane current flow, allowing us to accurately position electrodes to record from all layers in a single penetration.

Adaptation-induced synchronization in laminar cortical circuits.

A fundamental feature of information processing in neocortex is the ability of individual neurons to adapt to changes in incoming stimuli. It is increasingly being understood that cortical adaptation is a phenomenon that requires network interactions. The fact that the structure of local networks depends critically on cortical layer raises the possibility that adaptation could induce specific effects in different layers. Here we show that brief exposure (300 ms) to a stimulus of fixed orientation modulates the strength of synchronization between individual neurons and local population activity in the gamma-band frequency (30-80 Hz) in macaque primary visual cortex (V1) and influences the ability of individual neurons to encode stimulus orientation. Using laminar probes, we found that although stimulus presentation elicits a large increase in the gamma synchronization of rhythmic neuronal activity in the input (granular) layers of V1, adaptation caused a pronounced increase in synchronization in the cortical output (supragranular) layers. The increase in gamma synchronization after adaptation was significantly correlated with an improvement in neuronal orientation discrimination performance only in the supragranular layers. Thus, synchronization between the spiking activity of individual neurons and their local population may enhance sensory coding to optimize network processing across laminar circuits.

Attention alters visual plasticity during exposure-based learning.

It is generally believed that attention enhances the processing of sensory information during perception and learning. Here we report that, contrary to common belief, attention limits the degree of plasticity induced by repeated exposure to image features. Specifically, daily exposure to oriented stimuli that are not linked to a specific task causes an orientation-specific improvement in perceptual performance along the "exposed" axes. This effect is modulated by attention: human subjects showed a larger improvement in orientation discrimination when attention is directed toward the location where stimuli are presented. However, the capacity to perform discriminations away from the exposed orientation is enhanced when the exposure stimuli are unattended. Importantly, the improvement in orientation discrimination at the unattended location leads to a robust enhancement in the discrimination of complex stimuli, such as natural texture images, with orientation components along the exposed axes, whereas the improvement in orientation discrimination at the attended location exhibits only weak transfer to complex stimuli. These results indicate that sensory adaptation by passive stimulus exposure should be viewed as a form of perceptual learning that is complementary to practice-based learning in that it reduces constraints on generalization.