Key Takeaways
- A multidisciplinary team at Stony Brook University, led by Dr. Sima Mofakham, received a $2.5 million NIMH grant to develop “SeeMe,” a brain‑behavior synchronization system for detecting hidden consciousness in brain‑injured patients.
- SeeMe combines electrophysiology, computer vision, hand‑sensor data, and AI to identify subtle, involuntary responses to spoken commands that standard bedside exams miss.
- The project addresses cognitive motor dissociation—a condition where up to 25 % of patients labeled “unresponsive” may retain covert awareness.
- Phase 1 will refine and validate SeeMe as an objective monitoring tool in 80 traumatic brain injury (TBI) patients.
- Phase 2 will transform SeeMe into a closed‑loop system that times vagus nerve stimulation to reinforce emerging voluntary movements and promote recovery of consciousness.
- Funding is split into an R61 phase (initial development) and a contingent R33 phase (closed‑loop testing), with support through April 2031.
- Co‑principal investigators include Dr. Chuck Mikell (Neurosurgery) and Dr. Petar M. Djurić (Electrical and Computer Engineering).
- Preliminary findings on SeeMe were published in Nature Communications Medicine (2025), establishing it as a promising breakthrough technology.
Overview of the SeeMe Initiative and NIMH Support
Researchers at Stony Brook University, under the leadership of Dr. Sima Mofakham, have secured a five‑year grant exceeding $2.5 million from the National Institute of Mental Health (NIMH) to advance a novel technology called SeeMe. This brain‑behavior synchronization system aims to uncover covert signs of consciousness in patients who appear unresponsive after severe brain injury. The award, designated R61MH138612, will fund the project through April 2031 and is structured in two phases: an initial R61 stage focused on development and validation, followed by a contingent R33 stage that will test a closed‑loop therapeutic application contingent on meeting early milestones.
The Problem of Cognitive Motor Dissociation
A critical challenge in neurology and critical care is cognitive motor dissociation (CMD), a phenomenon where patients cannot exhibit voluntary motor responses despite retaining conscious awareness. Estimates suggest that as many as one‑quarter of individuals diagnosed as “unresponsive” in hospitals may actually experience CMD, leaving them without appropriate communication or rehabilitative opportunities. Dr. Mofakham emphasizes that resolving this diagnostic blind spot is urgent, as misclassification can impede timely interventions that might foster recovery of consciousness and functional improvement.
Limitations of Current Consciousness Assessments
Standard bedside examinations rely heavily on observable behaviors such as eye tracking, limb movement, or verbal response to commands. These assessments are inherently subjective and can fluctuate with fatigue, medication effects, or spontaneous neurological variability, leading to false‑negative conclusions about a patient’s inner state. Because CMD often manifests as fleeting, sub‑threshold signals, conventional exams frequently overlook them, underscoring the need for an objective, sensitive, and continuous monitoring approach.
How SeeMe Detects Covert Consciousness
SeeMe integrates multiple data streams to amplify weak neural and muscular signatures that precede overt movement. The system employs high‑density electroencephalography (EEG) or intracranial electrophysiology to capture brain activity, computer vision algorithms to track minute facial or ocular movements, and wearable hand sensors to record sub‑threshold muscle activations. Artificial intelligence models fuse these modalities in real time, learning patterns that correlate with intentional command following even when no visible response occurs. By providing an automated, quantitative index of covert responsiveness, SeeMe offers clinicians a more reliable marker of consciousness than bedside observation alone.
Phase 1: Development and Validation of SeeMe as a Monitoring Tool
During the first phase of the NIMH‑funded project, the team will refine SeeMe’s signal processing pipelines and validate its ability to detect CMD in a cohort of 80 traumatic brain injury (TBI) patients. Participants will undergo standardized auditory command tasks while SeeMe records multimodal data. The researchers will compare SeeMe’s output against established neurobehavioral scales (e.g., the Coma Recovery Scale‑Revised) and, where possible, against functional neuroimaging benchmarks. Successful validation will demonstrate that SeeMe can reliably differentiate truly unresponsive states from those with covert awareness, establishing the system as a trustworthy monitoring adjunct in intensive care and neonatal neurology settings.
Phase 2: Closed‑Loop Stimulation Guided by SeeMe
Building on validated detection capabilities, the second phase will deploy SeeMe within a closed‑loop framework designed to actively promote recovery. The system will identify moments of emerging voluntary intent—detected as transient, task‑related neural bursts—and trigger precisely timed vagus nerve stimulation (VNS). VNS has shown promise in enhancing neuroplasticity and facilitating motor relearning after brain injury. By delivering stimulation only when the brain shows signs of engagement, SeeMe aims to reinforce the nascent neural pathways associated with command following, potentially accelerating the return of observable consciousness and functional movement.
Patient Cohort, Funding Structure, and Timeline
The validation study will enroll 80 TBI patients across multiple trauma centers, ensuring a diverse sample regarding injury severity, etiology, and demographic variables. The NIMH award allocates funds across two phases: the R61 phase supports the initial development, feasibility testing, and early validation; the R33 phase, which is contingent on achieving predefined R61 milestones, will finance the closed‑loop VNS experiments and longer‑term outcome assessments. Funding is secured through April 2031, providing a stable platform for iterative technology refinement, regulatory preparation, and eventual translation to clinical practice.
Collaborative Expertise and Prior Publications
The project benefits from interdisciplinary leadership. Dr. Chuck Mikell, a clinical associate professor and vice chair for neurosurgery, provides critical insight into patient selection, intraoperative monitoring, and therapeutic integration. Dr. Petar M. Djurić, a distinguished professor in electrical and computer engineering, contributes expertise in signal processing, machine learning, and system design. Earlier work describing SeeMe’s core concepts appeared in Nature Communications Medicine in 2025, where the authors reported preliminary proof‑of‑concept data demonstrating the system’s sensitivity to covert motor intentions in animal models and a small human pilot cohort. This publication laid the groundwork for the current NIH‑supported translational effort.
Potential Impact and Future Directions
If SeeMe fulfills its promise, it could transform the standard of care for disorders of consciousness by reducing misdiagnosis rates, enabling earlier prognostication, and guiding personalized rehabilitative strategies. Closed‑loop VNS guided by real‑time detection of intent may leave a lasting imprint on neural circuits, fostering sustained improvements in arousal, communication, and motor function. Beyond TBI, the technology could be adapted for other etiologies of hypoxic‑ischemic injury, stroke, or encephalopathy, broadening its applicability. Ultimately, SeeMe exemplifies how converging advances in neuroscience, engineering, and AI can yield concrete tools that uncover hidden awareness and actively assist the brain’s healing journey.