Journal of Neural Engineering

This research demonstrates that the trans-aqueduct approach is a feasible, minimally invasive access pathway to the third ventricle, offering a potential route to the deep brain for therapeutic technologies. Further pre-clinical investigation is required to thoroughly evaluate physiological tolerance, trauma risk, and the long-term implications of intraventricular implantation. The third ventricle is a high-value site for neuromodulation due to its proximity to deep-brain targets, including the subthalamic nucleus (STN) and globus pallidus internus (GPi). This study defined the anatomical pathway; and evaluated the technical feasibility of retrograde access to the third ventricle via the cerebral aqueduct using minimally invasive interventional techniques. Evaluation was conducted in three phases using human MRI datasets (n=16; mean age 48.4 years) and cadaveric specimens (n=6; mean age 88.2 years). Phase 1 involved morphometric MRI analysis of the aqueduct and ventricles. Phase 2 tested trans-aqueduct access on cadaver specimens via fluoroscopically guided guidewires and catheters. Phase 3 utilized direct anatomical dissections on cadaver specimens (n=3) to morphometrically measure the third ventricular cavity and its relationship to deep-brain nuclei. Measurements across the sample groups showed a mean aqueduct diameter of 1.6 mm (SD=0.14). Third ventricle dimensions averaged 27.6 mm (ventral-dorsal), 19.9 mm (caudal-cranial), and 5.7 mm (lateral). Successful access to the third ventricle was achieved in 83% (5/6) of cadaveric specimens. The optimal technical configuration utilized a 0.018'' angled-tip guidewire and 5-6 Fr catheters; the aqueduct accommodated diameters up to 2.0 mm with minimal resistance. The STN and GPi were localized within 5-20 mm of the ventricular volumetric centroid. The trans-aqueduct approach is a technically feasible, minimally invasive pathway for accessing the third ventricle. This route offers a potential alternative for the delivery of therapeutic neurotechnologies. Further research is required to assess physiological tolerance, trauma risk, and the long-term safety of intraventricular implantation.

Electroencephalography (EEG) preprocessing is a critical yet time-consuming step that often relies on expert-driven, semi-automatic pipelines, limiting scalability and reproducibility across large datasets. In this work, we present sEEGnal, a fully automated and modular pipeline for EEG preprocessing designed to produce outputs comparable to expert-driven analyses while ensuring consistency and computational efficiency. The pipeline integrates three main modules: data standardization following the EEG extension of the Brain Imaging Data Structure (BIDS), bad channel detection, and artifact identification, combining physiologically grounded criteria with independent component analysis and ICLabel-based classification. Performance was evaluated against manual preprocessing performed by EEG experts at two complementary levels: preprocessing metadata (bad channels, artifact duration, and rejected components) and EEG-derived measures. In addition, test-retest analyses were conducted to assess the stability of the pipeline across repeated recordings. Results show that sEEGnal achieves performance comparable to expert-driven preprocessing while preserving key neurophysiological features. Furthermore, the pipeline demonstrates reduced variability and increased consistency compared to human experts. These findings support sEEGnal as a robust and scalable solution for automated EEG preprocessing in both research and large-scale applications. HighlightsFully automated and modular EEG preprocessing pipeline. Benchmarked against expert-driven preprocessing. Comparable performance in metadata and EEG-derived measures. Demonstrates stable performance in test-retest recordings. BIDS-based framework for reproducible EEG data handling.

Hidden (semi-) Markov Models (HsMMs) are increasingly being used to segment neurophysiological signals into sequences of latent cognitive processes. The idea: different processes will leave distinct traces in trial-level recordings of (multivariate) neuro-physiological signals. Markov models, equipped with an emission model of these traces and a latent process model describing the progression through the different latent processes involved in a task, can then be used to infer the most likely process for any time-point and trial. However, the currently used HsMMs remain limited in two important ways. First, they cannot account for subject-level heterogeneity in the latent and emission process. Instead, a single group-level model is assumed to explain the entire data. Second, they cannot account for the potentially non-linear effects of experimental covariates on the latent and emission process. To address these problems, we present a modeling framework in which the HsMM parameters of the emission and latent process are replaced with mixed additive models, including smooth functions of experimental covariates and random effects. We derive all necessary quantities for empirical Bayes and fully Bayesian inference for all parameters and provide a Python implementation of all estimation algorithms. To demonstrate the advantages offered by this framework, we apply such a multi-level model to an existing lexical decision dataset. We show that, even in such a simple task, not all subjects rely on the same processes equally and that at least two semi-Markov states, previously believed to reflect distinct processes, might actually relate to the same cognitive process.

ObjectivePathological beta-band oscillations (13 to 30 Hz) in the subthalamic nucleus (STN) are a hallmark of Parkinsons disease and a primary target for deep brain stimulation therapy, yet the specific pattern of synaptic reorganization that drives their emergence remains incompletely understood. We developed a GPU-accelerated computational framework to systematically investigate combinations of synaptic changes across basal ganglia pathways that produce Parkinsonian beta oscillations while satisfying literature-based electrophysiology constraints. ApproachWe implemented a biophysically detailed spiking network model of the STN, external globus pallidus (GPe), and internal globus pallidus (GPi) in JAX (a high-performance numerical computing Python library), achieving a 490-fold speedup over conventional CPU-based simulation. Using the Covariance Matrix Adaptation Evolution Strategy (CMA-ES) we optimized 10 network parameters across two stages: first establishing a healthy baseline matching primate electrophysiology data, then searching within biologically motivated bounds for synaptic modifications that reproduce Parkinsonian firing rates and beta power. Fixed in-degree connectivity ensured optimized parameters produced scale-invariant dynamics from 450 to 45000 neurons. All simulations ran on a single cloud GPU instance at 84 cents per hour. Main ResultsThe optimizer converged on a coordinated pattern of synaptic reorganization dominated by asymmetric changes within the STN-GPe reciprocal loop: STN to GPe excitation increased 2.21-fold while GPe to STN inhibition collapsed to 0.11-fold of its healthy value. STN to GPi and GPe to GPi pathways changed minimally (1.06-fold and 1.45-fold respectively). This configuration transformed asynchronous firing (beta: 0.4 percent of spectral power) into synchronized bursting with prominent beta oscillations (49.4 percent), with firing rate changes matching experimental observations. Network dynamics were invariant across a 100-fold range of network sizes (firing rate deviation less than 2.4 Hz; all metrics p less than 0.001 across 10 random seeds at 45000 neurons). We implemented a simplified deep brain stimulation model for validation purposes, which achieved complete beta suppression (49.4 percent to 0.0 percent) and restored GPi output to healthy levels. SignificanceThese results suggest that pathological beta oscillations emerge from a specific pattern of synaptic reorganization, namely the reduction of GPe inhibitory feedback to STN. The GPU-accelerated optimization framework, running on commodity cloud infrastructure, demonstrates an accessible platform for parameter exploration in neural circuit models and a foundation for generating synthetic training data for adaptive deep brain stimulation algorithms.

There is more to musculotendon path modeling than aligning a cable to reflect the geometric features of a muscle-tendon unit. From the perspective of simulation accuracy, the key is to replicate the length- and moment arm-joint angle relations of the target muscle. In this study, we propose an effect-oriented approach of automated path modeling, via the hybrid calibration based on muscle surface mesh and moment arm. The task is formulated as an optimization problem with a threefold objective for the path to: 1) pass through multiple ellipses representing muscle cross-sections, 2) yield moment arms that match experimental measurements, and 3) yield moment arms with the designated signs. The performance of our optimization framework is demonstrated with the musculoskeletal surface mesh from the Visible Human Male and moment arm datasets from literature--producing 42 paths that are anatomically realistic and biomechanically accurate in 20.1 min. Our optimization framework is gradient-specified, which is faster and more accurate than using the default numerical gradient, making it applicable for large-scale subject-specific uses.

ObjectiveIn individuals with drug-resistant epilepsy, accurately identifying the brain regions where seizures originate is a critical prerequisite to guide surgical treatment and achieve seizure freedom. To accomplish this, intracranial EEG is considered the gold standard, providing the spatiotemporal high-resolution data necessary to pinpoint epileptogenic activity. However, this precision is achieved through an invasive procedure with significant patient burden, which is fundamentally limited by the electrode placement and spatial coverage. MethodsIn this study, we investigated the potential utility of preoperative resting-state fMRI to non-invasively map alterations in brain dynamics at the whole brain level. Region-wise brain dynamics were quantified with complementary measures of local autocorrelation decay rates. We assessed the capacity of these derived features to effectively identify intracranial EEG confirmed seizure onset zones in 18 individuals with drug-resistant medial temporal lobe epilepsy. Overall, the study cohort contained 3867 implanted electrodes of which 159 classified as seizure onset zones by two independent board-certified epileptologists. ResultsOverall, our findings reveal more constrained temporal dynamics for brain regions associated with seizure onsets compared to non-seizure onset zones. Individual-level prediction showed a performance better than chance in 15 of the 18 patients. The overall predictive performance across all patients yielded a median AUC of 0.81, a median true positive rate of 0.75, and a median true negative rate of 0.83. Furthermore, in a subset of 13 patients, those with negative seizure outcomes showed higher probabilities of seizure onset zone predictions outside the resection area compared to those with good outcomes. SignificanceOverall, our findings suggest that altered temporal dynamics derived from preoperative resting-state fMRI represent a promising non-invasive approach for delineating epileptogenic tissue, potentially informing intervention strategies and guiding electrode placement.

Hippocampal sclerosis (HS) is the most common pathology in drug-resistant temporal lobe epilepsy (TLE). However, clinical diagnosis, prevalent epileptogenicity, and drug drug-resistance in individuals with HS remain an ongoing challenge demanding multidisciplinary research efforts. In this study, we examined the mechanical properties of neurosurgically en bloc resected HS specimens (n=8) ex vivo under compression, tension, and torsional shear. We fitted a two-term Ogden hyperelastic model to the measured mechanical responses to quantify nonlinear mechanical tissue properties. The resulting parameters revealed higher strain stiffening under compression in HS compared to hippocampus obtained post mortem (n=7). The distinction was most noticeable in the large-strain regime, which has important implications for using mechanical tissue properties as valuable diagnostic biomarker. Furthermore, we correlated the tissue microstructure with mechanical parameters. We trained a deep-learning histopathology classifier to detect and classify neurons and glial cells from hematoxylin-stained whole slide images (WSI). We identified a strong association between the small-strain stiffness (shear modulus {micro}) and the overall cell density as well as the glial cell density. The negative relationship between the neuron-to-glia ratio and shear modulus is consistent with the hypothesis that neuronal cell loss and gliosis drives tissue stiffening, respectively. Magnetic resonance imaging (MRI) analysis of the specimens confirmed the previously reported negative association between MRI-derived fractional anisotropy and shear modulus {micro}. Taken together, our study establishes a direct link between tissue mechanics and microstructure, suggesting nonlinear continuum mechanics models as promising new tools for clinical diagnosis and novel research strategies.

Background: Transcranial temporal interference stimulation (tTIS) is an emerging noninvasive neuromodulation approach that enables focal, frequency-specific modulation of deep brain regions, offering a novel method for investigating therapeutic mechanisms underlying brain and mental health disorders. Pain is a key target because it is a feature of multiple disorders and is increasingly understood to depend on brain circuits. Here, we tested the effects of tTIS on bilateral evoked pain, capitalizing on converging evidence from human and animal studies indicating that the primary motor cortex (M1) contains body-wide inter-effector regions and has descending projections to regions implicated in nociceptive, motivational, and autonomic processing, making it a key cortical target for pain modulation. Methods: We conducted a pre-registered, triple-blind, randomized crossover study (N = 32, 160 study sessions), investigating frequency-dependent effects of tTIS applied to the left M1 on experimentally evoked thermal pain in healthy adults. We tested four stimulation frequencies (10 Hz, 20 Hz, 70 Hz, and sham) on separate days (>10,000 pain trials total). Noxious heat was applied to both the right and left forearms using individually calibrated temperatures both pre- and post-stimulation. Results: Active tTIS produced significant analgesia at all stimulation frequencies (10 Hz, 20 Hz, and 70 Hz) relative to sham (Cohens d = 0.46-0.82, all p < 0.05). 10 Hz produced the greatest reduction (d = 0.82), and both 10 Hz and 20 Hz produced more analgesia than 70 Hz (d = 0.44 and 0.38, respectively; p < 0.05). Stimulation-related sensations were equivalent across frequencies, and participants were blind to condition. Pain reductions remained stable over a [~]40-min post-stimulation period and were bilateral, consistent with stimulation of body-wide inter-effector regions. Conclusions: These results provide the first evidence that tTIS can reliably reduce experimental pain perception in humans in a frequency-dependent manner, providing a foundation for noninvasive pain modulation with tTIS.

International guidelines for low-frequency electromagnetic field exposure (LF EMF) are primarily intended to prevent substantiated adverse effects. In the frameworks, limits on internal electric fields are linked to external exposure levels through computational dosimetry. However, the relationship between internal electric fields and these adverse effects remains incompletely understood. In particular, current approaches often overlook the morphological complexity and diversity of cortical neurons, which may limit the realism of neuronal activation estimates used to support these assessments. This study evaluates LF EMF-induced neural activation using 25 morphologically realistic neuron models spanning all cortical layers, embedded within 11 detailed human head models. The internal electric fields were simulated for uniform magnetic field exposures (100 Hz-100 kHz) along the three anatomical directions, and excitation thresholds were computed using a multi-scale framework combining voxel-based dosimetry with biophysical neuron simulations. A real-world exposure scenario involving a child near an acousto-magnetic article-surveillance deactivator was also analyzed. Thresholds varied across cell type, morphology, cortical location, subject anatomy, frequency, and exposure direction, with L2/3 pyramidal, L4 basket, and L5 thick-tufted pyramidal cells showing the lowest thresholds. Despite this variability, all simulated thresholds were conservative with respect to the basic restrictions and dosimetric reference limits set by IEEE ICES and ICNIRP. The smallest margin occurred at 100 kHz, where the threshold remained a factor of 2.8 above the corresponding limit. These findings indicate that current LF EMF exposure limits remain conservative when evaluated using highly detailed, morphology-based CNS activation models.

Neuropathy caused by chemotherapy is a common and debilitating side-effect of cancer treatment. With 30% of patients experiencing chronic neuropathy and with no good evidence-based treatments; early detection triggering chemotherapy regime modification remains the best option for prevention. Early detection is challenging because of a lack of diagnostic tools with sufficient longitudinal temporal precision and convenience for patient/clinical adoption. To tackle this problem, we developed SenseCheQ; enabling self-administered autonomous sensory testing which can be used by patients at home. We present the instrumental engineering approach taken to address the challenge, including haptic self-calibration combined with skin thermal-clamping protocols, and demonstrate robustly reliable performance in the face of environmental and user-related variance in home settings. We present prospective case studies of people having chemotherapy treatment for cancer, conducting regular unsupervised quantitative sensory testing to monitor their nerve function at home. These proof-of-principle studies show SenseCheQ can detect sub-clinical changes in nerve function, matching patient reported outcomes and lab-based sensory testing. This highlights SenseCheQs promise as a scalable biomarker platform for neuropathy-detection and therapeutic development.

Salient sensory stimuli are known to evoke neural activations across distributed brain regions. However, the temporal dynamics of these responses over sub-second timescales remain poorly understood, in part due to limitations in the temporal resolution of non-invasive neuroimaging methods. We examined the spatiotemporal dynamics of neural activations evoked by salient sensory stimuli (rare sounds) using 1,194 widely distributed intracranial electrodes in 5 neurosurgical patients. Salient stimuli preferentially activated 263 of 1,194 electrodes (22%), with responses segregating into two largely distinct spatiotemporal patterns: (1) phasic activation in sensorimotor regions, and (2) sustained activation within the salience network. Cross-correlation analysis revealed that phasic sensorimotor activation preceded sustained salience network activation on a trial-by-trial basis. These findings support an updated view of salience processing in the human brain, revealing that salient stimuli evoke two sequential stages of neural activation--phasic sensorimotor responses followed by sustained salience network activity--rather than simultaneous widespread activation.

Research in cognitive neuroscience has relied on simple, highly controlled stimuli due to the difficulty in developing standardized, ecologically valid stimulus sets. However, there is a consensus that using ecologically valid stimuli is imperative to generalize results beyond controlled laboratory settings. The current study introduces a naturalistic audio stimulus database, consisting of short, recognizable, and emotionally rated stimuli. To create such a database, the current study collected 291 audio files from a wide range of sources. 361 participants rated the audio clips on emotionality, arousal, and recognizability, and subsequently freely described the audios by typing what they believed the sound to be. The text responses of the participants were embedded and clustered using an unsupervised machine-learning algorithm to derive a participant-grounded organization of auditory object categories. The results indicate audio clips were easily recognizable, while emotionality and arousal ratings showed broad variability, making the database suitable for diverse experimental needs. Furthermore, the final database comprises 10 distinct semantic categories, providing a diverse set of auditory stimuli.

Routine ICU data offers valuable insights into daily physiological rhythms. While traditional methods assume these cycles maintain fixed periods and amplitudes, their inherent variability requires dynamic estimation of instantaneous trends. Wavelet transform effectively resolves circadian oscillations, especially for frequently measured vital parameters. We present novel extensions to the Continuous Wavelet Transform (CWT) power spectral analysis to better detect and segment subtle temporal patterns. Using this approach, we uncover hidden circadian patterns in cardiovascular vitals such as Heart Rate (HR) and Mean Blood Pressure (MBP) measured over five days in a retrospective cohort of 855 ICU patients. By quantifying non-stationary rhythms, we identified diurnal and semi-diurnal oscillations varying in period and power according to delirium and deep sedation. Notably, HR exhibits a clear diurnal and semi-diurnal rhythm when delirium is absent. Overall, our framework supports the CWT as a powerful tool for analyzing complex physiological signals, particularly vital signs. Crucially, our findings suggest that cardiovascular rhythm disruption can be associated with ICU-related delirium and deep sedation.

ObjectiveVagus nerve stimulation (VNS) is an established neuromodulation therapy used in the management of drug-resistant epilepsy (DRE), or when other intracranial surgical modalities have not reduced seizure burden. We evaluated whether prior intracranial surgery for epilepsy influences safety and effectiveness outcomes with adjunctive VNS, using real-world data from the CORE-VNS study. MethodsCORE-VNS (NCT03529045), a prospective, multicenter, international observational study, was designed to collect data on seizure and non-seizure outcomes in patients with DRE treated with VNS. Participants were identified as having or not having undergone prior intracranial brain surgery for epilepsy (ICSE) and received an initial VNS implant. Baseline seizure frequency data and patient-reported outcome measures were collected at 3, 6, 12, 24, and 36 months. This analysis compared the baseline data for VNS therapy and safety outcomes at 36 months. ResultsAmong 531 participants implanted with VNS, prior ICSE was performed in 84. Median percentage seizure reductions at 36 months for all seizures (76.6% and 76.3%), all focal seizures (83.3% and 71.8%), and all generalized seizures (77.8% and 76.2%) were found to be similar between those without and with a history of ICSE, respectively. The 50% responder rate for all seizures reported at baseline was similar, 64.8% and 61.8%, in both groups and complete seizure freedom was reported by 17.9% and 8.8%, respectively. Implant-related adverse events (AE) and serious AE rates were similar between groups. ConclusionVNS was associated with clinically meaningful seizure reductions and showed a consistent safety profile irrespective of the history of ICSE. Prior ICSE should not be a contraindication to the consideration of VNS.

Connectome-constrained neural network models aim to link anatomical connectivity with functional computation by training networks whose architectures reflect biological circuits. Because such models are increasingly used to infer neural mechanisms, it is important to assess their robustness to variations in training conditions and model selection criteria. Here we retrain ensembles of connectome-constrained models under nominally identical conditions and compare their correspondence to experimentally measured response properties in the Drosophila motion pathway. While task performance remains similar across models, the identification of biologically plausible circuit solutions is unstable across retraining runs. In particular, model clusters selected by lowest validation task error do not reliably correspond to experimentally observed neural tuning, and small variations in performance metrics can reorder cluster rankings. These results indicate that, in this framework, similar task performance does not reliably identify biologically plausible circuit solutions. Task error alone is therefore insufficient for mechanistic identification, and additional model-selection criteria are needed.

Short-interval intracortical inhibition (SICI) is the most widely used neurophysiological index of GABAergic inhibition in the human cortex. However, it is an indirect measure, inferring synaptic inhibition from suppression of peripherally recorded motor-evoked potentials (MEPs) elicited by transcranial magnetic stimulation (TMS). In the standard protocol, a subthreshold conditioning pulse suppresses the MEP evoked by a suprathreshold test pulse delivered 1-5 ms later. Interpretation is further complicated by temporal overlap with short-interval intracortical facilitation (SICF), reflecting excitatory interactions at interstimulus intervals of [~]1.5 and 2.7 ms. To overcome these limitations, we recorded immediate TMS-evoked EEG potentials (iTEPs; 1-10 ms post-stimulus) as a more direct measure of motor cortical activity in 16 healthy volunteers (20-35 years; 7 male). The conventional SICI protocol suppressed only later components of the iTEP, likely corresponding to late corticospinal volleys previously identified in epidural spinal recordings after suprathreshold TMS, while the earliest iTEP component was unaffected. Importantly, later iTEPs were suppressed to a similar extent whether conditioning-test intervals coincided with SICF peaks or troughs, and the magnitude of iTEP suppression correlated with concurrently recorded paired-pulse MEP suppression. SICI also reduced an early TEP component (N15; 10-20 ms), but paired-pulse N15 suppression showed a different dependence on stimulus intensity and did not correlate with MEP suppression. These findings demonstrate that SICI measured via MEPs does not reflect a global index of cortical GABAergic motor cortical inhibition but instead reflects inhibition within specific cortical circuits that can be investigated directly with iTEPs.

BackgroundTranscatheter aortic valve replacement has transformed the management of aortic stenosis; however, adverse outcomes such as leaflet thrombosis and hypoattenuating leaflet thickening remain clinically significant concerns. Flow disturbances resulting from valve canting may alter local hemodynamics and promote thrombogenic conditions. We investigated how modest transcatheter heart valve canting alters cusp-specific sinus flow and washout and promotes localized thrombogenic microenvironments associated with leaflet surface thrombus formation using particle image velocimetry, a physiologic blood loop, and tissue analysis. MethodsA patient-derived aortic root model was used to evaluate the hemodynamic and thrombogenic effects of THV canting at -10{degrees} (anti-curvature), 0{degrees} (neutral), and +10{degrees} (along-curvature). High-resolution particle image velocimetry quantified sinus flow fields and washout characteristics, and complementary whole-blood loop experiments enabled histologic assessment of leaflet-associated thrombus formation. ResultsCanting redistributed systolic jet orientation and sinus recirculation in a direction-dependent manner while preserving global hemodynamic measurements. The most spatially constrained cusp showed the largest increase in stasis and the slowest washout. In the right coronary cusp, anti-curvature canting increased the fraction of sinus area with velocity magnitude <0.05 m/s to 92% versus 43% in neutral and 10% in along-curvature deployments, and prolonged neo-sinus (T90) washout to 4.7 cycles versus 2.9 and 1.8 cycles, respectively. Histology localized surface-adherent platelet/fibrin thrombus to these poorly washed regions, most prominently on the right coronary cusp leaflet in anti-curvature deployments. Left and noncoronary cusp responses shifted with tilt direction, indicating redistribution rather than uniform worsening of thrombogenic conditions. ConclusionsEven modest noncoaxial deployment is sufficient to create sinus-resolved throm-bogenic microenvironments that are not captured by global gradient or effective orifice area. Deployment configuration is therefore a modifiable determinant of post-TAVR leaflet throm-bosis risk and may contribute to HALT.

A distinctive feature of bodily experience is its transparency. During skilled action, our limbs recede from awareness and function as the medium of interaction rather than perceptual objects1. This is reflected in systematic perceptual biases: humans reliably underestimate the weight of their own hands2, potentially reflecting predictive motor processes that modulate self-generated sensory signals. Wearable technologies may test the limits of this perceptual transparency. Exoskeletons and other augmentative devices attach directly to the body, adding mass that must be integrated into sensorimotor control3; yet little is known about how such devices are experienced as they become integrated into the sensorimotor system. Here, we tested whether training with finger-extending exoskeletons alters their perceived weight and whether such changes depend on active use. We developed a Bayesian analytic framework combining individual psychometric modelling with a regression-based decomposition of perceived weight, to partition contributions of the biological hand and attached exoskeletal device. Thirty-four right-handed adults completed a weight-perception task before and after 20 minutes of training with either finger-extending or non-augmenting control devices. Participants compared the perceived weight of their right hand, with or without the exoskeleton, to reference weights suspended from the opposite wrist. Before training, the weight of both the biological hand and the exoskeleton were underestimated to a similar degree ([~]25- 30%), suggesting rapid perceptual integration following attachment. Training selectively increased attenuation of the perceived weight of the finger-extending exoskeleton, with no corresponding change for the biological hand and little evidence for a general training effect. These findings support a two-stage embodiment process in which passive attachment initiates perceptual updating, while sensorimotor training consolidates integration through functional interaction with the device. Perceived weight thus provides a behavioral marker of embodiment, offering insight into how the sensorimotor system integrates wearable augmentative technologies.

Subcallosal cingulate cortex (SCC) deep brain stimulation (DBS) can provide relief for individuals with Treatment Resistant Depression (TRD), but ongoing clinical management remains challenging due to nonspecific symptom fluctuations that can obscure core depression recovery on standard rating scales. Objective, stable biomarkers that selectively track the therapeutic effects of SCC DBS are therefore essential for developing principled decision support systems to guide stimulation adjustments. Recent bidirectional DBS systems enable chronic recording of local field potentials (LFPs) and prior work using the Activa PC+S device identified an electrophysiological signature of stable clinical recovery. However, translation to practical clinical deployment requires demonstrating that this biomarker is robustly generalizable, specific to the impact of the DBS therapy, and deployable in real-world recording contexts. To address this need, we developed an at-home SCC LFP data collection platform (built on the Medtronic Summit RC+S system) enabling at home data collection for a new cohort of ten SCC DBS participants with TRD (ClinicalTrials.gov identifier NCT04106466). Using longitudinal LFP recordings collected from this system, we report findings demonstrating that the previously reported biomarker of stable recovery generalizes across subject cohorts and devices, is robust to common potential confounds (including time of day and stimulation status), and shows symptom specificity, sensitivity and stability necessary to support clinical decision making. Across both cohorts, biomarker changes show relationships to pre-DBS white matter structure and network function measured using diffusion MRI and resting-state functional MRI (rsFMRI). These findings replicating and extending previous findings support the biomarkers utility as a foundation for scalable, electrophysiology-informed decision support in SCC DBS.

Automated glaucoma screening from optical coherence tomography (OCT) faces two persistent challenges: scarcity of expert-labeled data and unreliable model predictions on diagnostically ambiguous cases. We present a two-tier diagnostic pipeline that addresses both. In the first tier, an EfficientNetV2-S classifier trained under a semi-supervised pseudo supervisor framework achieves 0.84 AUC on 150 held-out test patients from the Harvard Glaucoma Detection and Progression dataset, using only 350 labeled training samples out of 700. In the second tier, 124 flagged cases are routed to a multi-agent system built on MedGemma 4B, where three specialist agents deliberate over three rounds before rendering a final diagnosis. On these flagged cases, the agent system achieves 100% sensitivity--detecting all 55 glaucoma cases with zero missed diagnoses--and 89.5% overall accuracy (111/124), compared to the classifiers 73.4% (91/124). Uncertainty analysis confirms that the classifiers output probability reliably separates confident predictions (96.3% accuracy, n = 27) from uncertain ones (74.0%, n = 123), producing a 22-percentage-point gap that serves as a triage signal. The agents fix 32 cases the classifier misclassifies while introducing 12 new errors, yielding a net improvement of 20 cases. These results are from a single training run without variance estimates and should be interpreted as preliminary evidence that uncertainty-gated routing to vision-language model agents can meaningfully improve diagnostic accuracy on the cases where automated classifiers are least reliable.