Classification algorithms for the identification of structural injury in TBI using brain electrical activity

Point-of-Care Electroencephalography in Acute Neurological Care: A Narrative Review

Point-of-care electroencephalography (POC-EEG) systems are rapid-access, reduced-montage devices designed to address the limitations of conventional EEG (conv-EEG), enabling faster neurophysiological assessment in acute settings. This review evaluates their clinical impact, diagnostic performance, and feasibility in non-convulsive status epilepticus (NCSE), traumatic brain injury (TBI), stroke, and delirium. A comprehensive search of Medline, Scopus, and Embase identified 69 studies assessing 15 devices. In suspected NCSE, POC-EEG facilitates rapid seizure detection and prompt diagnosis, making it particularly effective in time-sensitive and resource-limited settings. Its after-hours availability and telemedicine integration ensure continuous coverage. AI-assisted tools enhance interpretability and accessibility, enabling use by non-experts. Despite variability in accuracy, it supports triaging, improving management, treatment decisions and outcomes while reducing hospital stays, transfers, and costs. In TBI, POC-EEG-derived quantitative EEG (qEEG) indices reliably detect structural lesions, support triage, and minimize unnecessary CT scans. They also help assess concussion severity and predict recovery. For strokes, POC-EEG aids triage by detecting large vessel occlusions (LVOs) with high feasibility in hospital and prehospital settings. In delirium, spectral analysis and AI-assisted models enhance diagnostic accuracy, broadening its clinical applications. Although POC-EEG is a promising screening tool, challenges remain in diagnostic variability, technical limitations, and AI optimization, requiring further research.

The Power of Public-Private Partnership in Medical Technology Innovation: Lessons From the Development of Fda-Cleared Medical Devices for Assessment of Concussion

Given the convergence of the long and challenging development path for medical devices with the need for diagnostic capabilities for mild traumatic brain injury (mTBI/concussion), the effective role of public–private partnership (PPP) can be demonstrated to yield Food and Drug Administration (FDA) clearances and innovative product introductions. An overview of the mTBI problem and landscape was performed. A detailed situation analysis of an example of a PPP yielding an innovative product was further demonstrated. The example of PPP has led to multiple FDA clearances and product introductions in the TBI diagnostic product category where there was an urgent military and public need. Important lessons included defining the primary public and military health objective for new product introduction, the importance of the government–academia–industry PPP triad with a “collaboration towards solutions” Quality-by-Design (QbD) mindset to assure clinical validity with regulatory compliance, the development of device comparators and integration of measurements into a robust, evidence-based statistical and FDA pathway, and the utility of top-down, flexible, practical action while operating within governmental guidelines and patient safety.

Phybrata Sensors and Machine Learning for Enhanced Neurophysiological Diagnosis and Treatment

Concussion injuries remain a significant public health challenge. A significant unmet clinical need remains for tools that allow related physiological impairments and longer-term health risks to be identified earlier, better quantified, and more easily monitored over time. We address this challenge by combining a head-mounted wearable inertial motion unit (IMU)-based physiological vibration acceleration (“phybrata”) sensor and several candidate machine learning (ML) models. The performance of this solution is assessed for both binary classification of concussion patients and multiclass predictions of specific concussion-related neurophysiological impairments. Results are compared with previously reported approaches to ML-based concussion diagnostics. Using phybrata data from a previously reported concussion study population, four different machine learning models (Support Vector Machine, Random Forest Classifier, Extreme Gradient Boost, and Convolutional Neural Network) are first investigated for binary classification of the test population as healthy vs. concussion (Use Case 1). Results are compared for two different data preprocessing pipelines, Time-Series Averaging (TSA) and Non-Time-Series Feature Extraction (NTS). Next, the three best-performing NTS models are compared in terms of their multiclass prediction performance for specific concussion-related impairments: vestibular, neurological, both (Use Case 2). For Use Case 1, the NTS model approach outperformed the TSA approach, with the two best algorithms achieving an F1 score of 0.94. For Use Case 2, the NTS Random Forest model achieved the best performance in the testing set, with an F1 score of 0.90, and identified a wider range of relevant phybrata signal features that contributed to impairment classification compared with manual feature inspection and statistical data analysis. The overall classification performance achieved in the present work exceeds previously reported approaches to ML-based concussion diagnostics using other data sources and ML models. This study also demonstrates the first combination of a wearable IMU-based sensor and ML model that enables both binary classification of concussion patients and multiclass predictions of specific concussion-related neurophysiological impairments.

Review of wearable technologies and machine learning methodologies for systematic detection of mild traumatic brain injuries

Mild traumatic brain injuries (mTBIs) are the most common type of brain injury. Timely diagnosis of mTBI is crucial in making 'go/no-go' decision in order to prevent repeated injury, avoid strenuous activities which may prolong recovery, and assure capabilities of high-level performance of the subject. If undiagnosed, mTBI may lead to various short- and long-term abnormalities, which include, but are not limited to impaired cognitive function, fatigue, depression, irritability, and headaches. Existing screening and diagnostic tools to detect acute andearly-stagemTBIs have insufficient sensitivity and specificity. This results in uncertainty in clinical decision-making regarding diagnosis and returning to activity or requiring further medical treatment. Therefore, it is important to identify relevant physiological biomarkers that can be integrated into a mutually complementary set and provide a combination of data modalities for improved on-site diagnostic sensitivity of mTBI. In recent years, the processing power, signal fidelity, and the number of recording channels and modalities of wearable healthcare devices have improved tremendously and generated an enormous amount of data. During the same period, there have been incredible advances in machine learning tools and data processing methodologies. These achievements are enabling clinicians and engineers to develop and implement multiparametric high-precision diagnostic tools for mTBI. In this review, we first assess clinical challenges in the diagnosis of acute mTBI, and then consider recording modalities and hardware implementation of various sensing technologies used to assess physiological biomarkers that may be related to mTBI. Finally, we discuss the state of the art in machine learning-based detection of mTBI and consider how a more diverse list of quantitative physiological biomarker features may improve current data-driven approaches in providing mTBI patients timely diagnosis and treatment.

Measures of resting state EEG rhythms for clinical trials in Alzheimer's disease: Recommendations of an expert panel

The Electrophysiology Professional Interest Area (EPIA) and Global Brain Consortium endorsed recommendations on candidate electroencephalography (EEG) measures for Alzheimer's disease (AD) clinical trials. The Panel reviewed the field literature. As most consistent findings, AD patients with mild cognitive impairment and dementia showed abnormalities in peak frequency, power, and "interrelatedness" at posterior alpha (8-12 Hz) and widespread delta (< 4 Hz) and theta (4-8 Hz) rhythms in relation to disease progression and interventions. The following consensus statements were subscribed: (1) Standardization of instructions to patients, resting state EEG (rsEEG) recording methods, and selection of artifact-free rsEEG periods are needed; (2) power density and "interrelatedness" rsEEG measures (e.g., directed transfer function, phase lag index, linear lagged connectivity, etc.) at delta, theta, and alpha frequency bands may be use for stratification of AD patients and monitoring of disease progression and intervention; and (3) international multisectoral initiatives are mandatory for regulatory purposes.

Evaluating Performance of EEG Data-Driven Machine Learning for Traumatic Brain Injury Classification

Objectives:

Big data analytics can potentially benefit the assessment and management of complex neurological conditions by extracting information that is difficult to identify manually. In this study, we evaluated the performance of commonly used supervised machine learning algorithms in the classification of patients with traumatic brain injury (TBI) history from those with stroke history and/or normal EEG.

Methods:

Support vector machine (SVM) and K-nearest neighbors (KNN) models were generated with a diverse feature set from Temple EEG Corpus for both two-class classification of patients with TBI history from normal subjects and three-class classification of TBI, stroke and normal subjects.

Results:

For two-class classification, an accuracy of 0.94 was achieved in 10-fold cross validation (CV), and 0.76 in independent validation (IV). For three-class classification, 0.85 and 0.71 accuracy were reached in CV and IV respectively. Overall, linear discriminant analysis (LDA) feature selection and SVM models consistently performed well in both CV and IV and for both two-class and three-class classification. Compared to normal control, both TBI and stroke patients showed an overall reduction in coherence and relative PSD in delta frequency, and an increase in higher frequency (alpha, mu, beta and gamma) power. But stroke patients showed a greater degree of change and had additional global decrease in theta power.

Conclusions:

Our study suggests that EEG data-driven machine learning can be a useful tool for TBI classification.

Significance:

Our study provides preliminary evidence that EEG ML algorithm can potentially provide specificity to separate different neurological conditions.

Joint analysis of frontal theta synchrony and white matter following mild traumatic brain injury

Some of the most disabling aspects of mild traumatic brain injury (mTBI) include lingering deficits in executive functioning. It is known that mTBI can damage white matter tracts, but it remains unknown how this structural brain damage translates into cognitive deficits. This experiment utilized theta band phase synchrony to identify the dysfunctional neural operations that contribute to cognitive problems following mTBI. Sub-acute stage (< 2 weeks) mTBI patients (N = 52) and healthy matched controls (N = 32) completed a control-demanding task with concurrent EEG. Structural MRI was also collected. While there were no performance-specific behavioral differences between groups in the dot probe expectancy task, the degree of theta band phase synchrony immediately following injury predicted the degree of symptom recovery two months later. Although there were no differences in fractional anisotropy (FA) between groups, joint independent components analysis revealed that a smaller network of lower FA-valued voxels contributed to a diminished frontal theta phase synchrony network in the mTBI group. This finding suggests that frontal theta band markers of cognitive control are sensitive to sub-threshold structural aberrations following mTBI.

Classification of Non-Severe Traumatic Brain Injury from Resting-State EEG Signal Using LSTM Network with ECOC-SVM

Traumatic brain injury (TBI) is one of the common injuries when the human head receives an impact due to an accident or fall and is one of the most frequently submitted insurance claims. However, it is often always misused when individuals attempt an insurance fraud claim by providing false medical conditions. Therefore, there is a need for an instant brain condition classification system. This study presents a novel classification architecture that can classify non-severe TBI patients and healthy subjects employing resting-state electroencephalogram (EEG) as the input, solving the immobility issue of the computed tomography (CT) scan and magnetic resonance imaging (MRI). The proposed architecture makes use of long short term memory (LSTM) and error-correcting output coding support vector machine (ECOC-SVM) to perform multiclass classification. The pre-processed EEG time series are supplied to the network by each time step, where important information from the previous time step will be remembered by the LSTM cell. Activations from the LSTM cell is used to train an ECOC-SVM. The temporal advantages of the EEG were amplified and able to achieve a classification accuracy of 100%. The proposed method was compared to existing works in the literature, and it is shown that the proposed method is superior in terms of classification accuracy, sensitivity, specificity, and precision.

Pain phenotypes classified by machine learning using electroencephalography features

Pain is a multidimensional experience mediated by distributed neural networks in the brain. To study this phenomenon, EEGs were collected from 20 subjects with chronic lumbar radiculopathy, 20 age and gender matched healthy subjects, and 17 subjects with chronic lumbar pain scheduled to receive an implanted spinal cord stimulator. Analysis of power spectral density, coherence, and phase-amplitude coupling using conventional statistics showed that there were no significant differences between the radiculopathy and control groups after correcting for multiple comparisons. However, analysis of transient spectral events showed that there were differences between these two groups in terms of the number, power, and frequency-span of events in a low gamma band. Finally, we trained a binary support vector machine to classify radiculopathy versus healthy subjects, as well as a 3-way classifier for subjects in the 3 groups. Both classifiers performed significantly better than chance, indicating that EEG features contain relevant information pertaining to sensory states, and may be used to help distinguish between pain states when other clinical signs are inconclusive.

Mitigating the Consequences of Subconcussive Head Injuries

Subconcussive head injury represents a pathophysiology that spans the expertise of both clinical neurology and biomechanical engineering. From both viewpoints, the terms injury and damage, presented without qualifiers, are synonymously taken to mean a tissue alteration that may be recoverable. For clinicians, concussion is evolving from a purely clinical diagnosis to one that requires objective measurement, to be achieved by biomedical engineers. Subconcussive injury is defined as subclinical pathophysiology in which underlying cellular- or tissue-level damage (here, to the brain) is not severe enough to present readily observable symptoms. Our concern is not whether an individual has a (clinically diagnosed) concussion, but rather, how much accumulative damage an individual can tolerate before they will experience long-term deficit(s) in neurological health. This concern leads us to look for the history of damage-inducing events, while evaluating multiple approaches for avoiding injury through reduction or prevention of the associated mechanically induced damage.

ERPs predict symptomatic distress and recovery in sub-acute mild traumatic brain injury

Mild traumatic brain injury (mTBI) can affect high-level executive functioning long after somatic symptoms resolve. We tested if simple EEG responses within an oddball paradigm could capture variance relevant to this clinical problem. The P3a and P3b components reflect bottom-up and top-down processes driving engagement with exogenous stimuli. Since these features are related to primitive decision abilities, abnormal amplitudes following mTBI may account for problems in the ability to exert executive control. Sub-acute (<2 weeks) mTBI participants (N = 38) and healthy controls (N = 24) were assessed at an initial session as well as a two-month follow-up (sessions 1 and 2). We contrasted the initial assessment to a comparison group of participants with chronic symptomatology following brain injury (N = 23). There were no group differences in P3a or P3b amplitudes. Yet in the sub-acute mTBI group, higher symptomatology on the Frontal Systems Behavior scale (FrSBe), a questionnaire validated as measuring symptomatic distress related to frontal lobe injury, correlated with lower P3a in session 1. This relationship was replicated in session 2. These findings were distinct from chronic TBI participants, who instead expressed a relationship between increased FrSBe symptoms and a lower P3b component. In the sub-acute group, P3b amplitudes in the first session correlated with the degree of symptom change between sessions 1 and 2, above and beyond demographic predictors. Controls did not show any relationship between FrSBe symptoms and P3a or P3b. These findings identify symptom-specific alterations in neural systems that vary along the time course of post-concussive symptomatology.

The Distribution of Transplanted Umbilical Cord Mesenchymal Stem Cells in Large Blood Vessel of Experimental Design With Traumatic Brain Injury

The authors aim to track the distribution of human umbilical cord mesenchymal stem cells (MSCs) in large blood vessel of traumatic brain injury -rats through immunohistochemical method and small animal imaging system. After green fluorescent protein (GFP) gene was transfected into 293T cell, virus was packaged and MSCs were transfected. Mesenchymal stem cells containing GFP were transplanted into brain ventricle of rats when the infection rate reaches 95%. The immunohistochemical and small animal imaging system was used to detect the distribution of MSCs in large blood vessels of rats. Mesenchymal stem cells could be observed in large vessels with positive GFP expression 10 days after transplantation, while control groups (normal group and traumatic brain injury group) have negative GFP expression. The vascular endothelial growth factor in transplantation group was higher than that in control groups. The in vivo imaging showed obvious distribution of MSCs in the blood vessels of rats, while no MSCs could be seen in control groups. The intravascular migration and homing of MSCs could be seen in rats received MSCs transplantation, and new angiogenesis could be seen in MSCs-transplanted blood vessels.

The Patient Repository for EEG Data + Computational Tools (PRED+CT)

Electroencephalographic (EEG) recordings are thought to reflect the network-wide operations of canonical neural computations, making them a uniquely insightful measure of brain function. As evidence of these virtues, numerous candidate biomarkers of different psychiatric and neurological diseases have been advanced. Presumably, we would only need to apply powerful machine-learning methods to validate these ideas and provide novel clinical tools. Yet, the reality of this advancement is more complex: the scale of data required for robust and reliable identification of a clinical biomarker transcends the ability of any single laboratory. To surmount this logistical hurdle, collective action and transparent methods are required. Here we introduce the Patient Repository of EEG Data + Computational Tools (PRED+CT: predictsite.com). The ultimate goal of this project is to host a multitude of available tasks, patient datasets, and analytic tools, facilitating large-scale data mining. We hope that successful completion of this aim will lead to the development of novel EEG biomarkers for differentiating populations of neurological and psychiatric disorders.

A Brain Electrical Activity Electroencephalographic-Based Biomarker of Functional Impairment in Traumatic Brain Injury: A Multi-Site Validation Trial

The potential clinical utility of a novel quantitative electroencephalographic (EEG)-based Brain Function Index (BFI) as a measure of the presence and severity of functional brain injury was studied as part of an independent prospective validation trial. The BFI was derived using quantitative EEG (QEEG) features associated with functional brain impairment reflecting current consensus on the physiology of concussive injury. Seven hundred and twenty adult patients (18-85 years of age) evaluated within 72 h of sustaining a closed head injury were enrolled at 11 U.S. emergency departments (EDs). Glasgow Coma Scale (GCS) score was 15 in 97%. Standard clinical evaluations were conducted and 5 to 10 min of EEG acquired from frontal locations. Clinical utility of the BFI was assessed for raw scores and percentile values. A multinomial logistic regression analysis demonstrated that the odds ratios (computed against controls) of the mild and moderate functionally impaired groups were significantly different from the odds ratio of the computed tomography (CT) postive (CT+, structural injury visible on CT) group (p = 0.0009 and p = 0.0026, respectively). However, no significant differences were observed between the odds ratios of the mild and moderately functionally impaired groups. Analysis of variance (ANOVA) demonstrated significant differences in BFI among normal (16.8%), mild TBI (mTBI)/concussed with mild or moderate functional impairment, (61.3%), and CT+ (21.9%) patients (p < 0.0001). Regression slopes of the odds ratios for likelihood of group membership suggest a relationship between the BFI and severity of impairment. Findings support the BFI as a quantitative marker of brain function impairment, which scaled with severity of functional impairment in mTBI patients. When integrated into the clinical assessment, the BFI has the potential to aid in early diagnosis and thereby potential to impact the sequelae of TBI by providing an objective marker that is available at the point of care, hand-held, non-invasive, and rapid to obtain.

Identification of hematomas in mild traumatic brain injury using an index of quantitative brain electrical activity

Rapid identification of traumatic intracranial hematomas following closed head injury represents a significant health care need because of the potentially life-threatening risk they present. This study demonstrates the clinical utility of an index of brain electrical activity used to identify intracranial hematomas in traumatic brain injury (TBI) presenting to the emergency department (ED). Brain electrical activity was recorded from a limited montage located on the forehead of 394 closed head injured patients who were referred for CT scans as part of their standard ED assessment. A total of 116 of these patients were found to be CT positive (CT+), of which 46 patients with traumatic intracranial hematomas (CT+) were identified for study. A total of 278 patients were found to be CT negative (CT-) and were used as controls. CT scans were subjected to quantitative measurements of volume of blood and distance of bleed from recording electrodes by blinded independent experts, implementing a validated method for hematoma measurement. Using an algorithm based on brain electrical activity developed on a large independent cohort of TBI patients and controls (TBI-Index), patients were classified as either positive or negative for structural brain injury. Sensitivity to hematomas was found to be 95.7% (95% CI = 85.2, 99.5), specificity was 43.9% (95% CI = 38.0, 49.9). There was no significant relationship between the TBI-Index and distance of the bleed from recording sites (F = 0.044, p = 0.833), or volume of blood measured F = 0.179, p = 0.674). Results of this study are a validation and extension of previously published retrospective findings in an independent population, and provide evidence that a TBI-Index for structural brain injury is a highly sensitive measure for the detection of potentially life-threatening traumatic intracranial hematomas, and could contribute to the rapid, quantitative evaluation and treatment of such patients.

Transcranial vibro-acoustography can detect traumatic brain injury, in-vivo: Preliminary studies

Vibro-acoustography (VA) uses two or more beams of confocal ultrasound to generate local vibrations within their target tissue through induction of a time-dependent radiation force whose frequency equals that of the difference of the applied frequencies. While VA has proven effective for assaying the mechanical properties of clinically relevant tissue such as breast lesions and tissue calcifications, its application to brain remains unexplored. Here we investigate the ability of VA to detect acute and focal traumatic brain injury (TBI) in-vivo through the use of transcranially delivered high-frequency (2 MHz) diagnostic focused ultrasound to rat brain capable of generating measurable low-frequency (200-270 kHz) acoustic emissions from outside of the brain. We applied VA to acute sham-control and TBI model rats (sham N=6; TBI N=6) and observed that acoustic emissions, captured away from the site of TBI, had lower amplitudes for TBI as compared to sham-TBI animals. The sensitivity of VA to acute brain damage at frequencies currently transmittable across human skulls, as demonstrated in this preliminary study, supports the possibility that the VA methodology may one day serve as a technique for detecting TBI.

Identification of acute stroke using quantified brain electrical activity

OBJECTIVES

Acute stroke is a leading cause of brain injury and death and requires rapid and accurate diagnosis. Noncontrast head computed tomography (CT) is the first line for diagnosis in the emergency department (ED). Complicating rapid triage are presenting conditions that clinically mimic stroke. There is an extensive literature reporting clinical utility of brain electrical activity in early diagnosis and management of acute stroke. However, existing technologies do not lend themselves to easily acquired rapid evaluation. This investigation used an independently derived classifier algorithm for the identification of traumatic structural brain injury based on brain electrical activity recorded from a reduced frontal montage to explore the potential clinical utility of such an approach in acute stroke assessment.

METHODS

Adult patients (age 18 to 95 years) presenting with stroke-like and/or altered mental status symptoms were recruited from urban academic EDs as part of a large research study evaluating the clinical utility of quantitative brain electrical activity in acutely brain-injured patients. All patients from the parent study who had confirmed strokes, and a control group of stroke mimics (those with final ED diagnoses of migraine or syncope), were selected for this study. All stroke patients underwent head CT scans. Some patients with negative CTs had further imaging with magnetic resonance imaging (MRI). Ten minutes of electroencephalographic data were acquired on a hand-held device in development, from five frontal electrodes. Data analyses were done offline. A Structural Brain Injury Index (SBII) was derived using an independently developed binary discriminant classification algorithm whose input was specified features of brain electrical activity. The SBII was previously found to have high accuracy in the identification of traumatic brain-injured patients who were found to have brain injury on CT (CT+). This algorithm was applied to patients in this study and used to classify patients as CT+ or not CT+. Performance was assessed using sensitivity, specificity, and negative and positive predictive values (NPV, PPV).

RESULTS

Forty-eight stroke patients (31 ischemic and 17 hemorrhagic) and 135 stroke mimic controls were included. Within the ischemic population, approximately half were CT- but later confirmed for stroke with MRI (CT-/MRI+). Sensitivity to stroke was 91.7%, specificity 50.4% (to stroke mimic), NPV 94.4%, and PPV 39.6%. Eighty percent of the CT-/MRI+ ischemic strokes were correctly identified at the time of the CT- scan.

CONCLUSIONS

Despite a small population and the use of a classifier without the benefit of training on a stroke population, these data suggest that a rapidly acquired, easy-to-use system to assess brain electrical activity at the time of evaluation of acute stroke could be a valuable adjunct to current clinical practice.