AI and epilepsy diagnosis: a review of the challenges

At the Medical University of South Carolina, researchers are using machine learning to personalize repetitive transcranial magnetic stimulation (rTMS)—an FDA-approved treatment for smoking cessation. Their recent study, published in Brain Connectivity, revealed something unexpected: while most research has focused on the brain’s reward network, this study found that connectivity in the salience network was the strongest predictor of rTMS success. 📍 Impact: This breakthrough could lead to tailored, image-guided neuromodulation—improving effectiveness, reducing side effects, and offering new hope for treatment-resistant nicotine addiction. 💡 Insight: Combining fMRI and AI-driven analysis allows researchers to go beyond one-size-fits-all protocols. We’re entering an era where neural signatures can guide therapy, not just diagnosis. 🤔 Questions to consider: - Could this approach extend to other forms of substance use or behavioral addictions? - How do we build scalable biomarker pipelines to support personalized neuropsychiatric care? - What are the ethical implications of AI-driven brain modulation in clinical treatment? We often talk about "meeting patients where they are"—this is about meeting their brains where they are, with the help of AI. #Neurotech #DigitalHealth #SmokingCessation #AIinHealthcare #TMS #BrainStimulation #PersonalizedMedicine #MentalHealthInnovation #MachineLearning #AddictionTreatment #MUSC #PrecisionNeuroscience https://lnkd.in/g57bqpVx

From Crisis to Capability: AI Steps In Where Experts Are Scarce In 2020, the Association of British Neurologists raised the alarm: a severe shortage of EEG interpreters was hindering care for patients with epilepsy, sleep disorders, and brain injuries. These essential scans were accumulating, awaiting review by trained professionals. Patients experienced delays of weeks or months for results that could influence their treatment plans. Now, a team at UWE Bristol is turning the tide. Their project, MED-SHED, has developed AI models that interpret EEGs with accuracy matching or exceeding that of human experts, achieving over 90% accuracy in clinical validation. Supported by UKRI's Proof of Concept programme, the team is transitioning from laboratory to clinical settings, validating their models on real NHS data. What distinguishes this is the deployment approach. Instead of rushing to market, the team is considering a Community Interest Company structure to guarantee public benefit and transparency. This approach ensures the technology is developed with patient outcomes and equity embedded from the outset, not added later. This is more than just technical progress. It is a blueprint for trustworthy clinical AI that prioritises validation, responsible deployment, and accessible care. As Dr David Western puts it: "By focusing on validation and establishing a trustworthy route to market, this project will de-risk the clinical translation of our innovative AI technology." Practically, this means the AI will undergo thorough testing before reaching patients, lowering the chance of mistakes and increasing trust among clinicians who will use it. Faster diagnoses. Improved treatment choices. Greater access to healthcare. Read more: https://lnkd.in/eNt25dq8 #AIinHealthcare #EEG #ResponsibleAI #UKHealthTech #NeuroInnovation #DigitalHealth #NHSInnovation #HealthEquity #TrustworthyAI

🧠 Can AI match the expertise of a trained neurologist? A UWE Bristol research team led by Dr David Western has developed advanced AI models that interpret EEG brain activity with expert-level accuracy, helping tackle the national shortage of trained interpreters for neurological conditions. The project, MED-SHED, is one of just 48 across the UK funded through UKRI’s Proof of Concept programme - designed to help cutting-edge research translate into real-world impact. Now, the team is working to bring the technology into clinical settings, using NHS data and exploring social enterprise models that prioritise transparency and public benefit. 📍Potential impact? - Faster, more accurate diagnosis - Better treatment decisions - Wider access to brain monitoring tools in critical care 📖 Read more via UWE News: 🔗 https://lnkd.in/ehZdjHRG #AIinHealthcare #BrainHealth #Neurotechnology #EEG #UKRI #ProofOfConcept #UWEBristol #Innovation #ResearchWithPurpose

𝐀𝐥𝐳𝐡𝐞𝐢𝐦𝐞𝐫’𝐬 𝐝𝐢𝐬𝐞𝐚𝐬𝐞 𝐜𝐥𝐚𝐬𝐬𝐢𝐟𝐢𝐜𝐚𝐭𝐢𝐨𝐧 𝐮𝐬𝐢𝐧𝐠 𝐚 𝐡𝐲𝐛𝐫𝐢𝐝 𝐝𝐞𝐞𝐩 𝐥𝐞𝐚𝐫𝐧𝐢𝐧𝐠 𝐚𝐩𝐩𝐫𝐨𝐚𝐜𝐡 𝐰𝐢𝐭𝐡 𝐦𝐮𝐥𝐭𝐢-𝐥𝐚𝐲𝐞𝐫 𝐔-𝐧𝐞𝐭 𝐬𝐞𝐠𝐦𝐞𝐧𝐭𝐚𝐭𝐢𝐨𝐧 𝐚𝐧𝐝 𝐗𝐀𝐈 𝐝𝐫𝐢𝐯𝐞𝐧 𝐚𝐧𝐚𝐥𝐲𝐬𝐢𝐬 𝐀𝐛𝐬𝐭𝐫𝐚𝐜𝐭: Alzheimer’s disease (AD) is a neurodegenerative illness causing a significant decrease in cognitive function, and early, accurate diagnosis is of great therapeutic and diagnostic value. Currently, there is promising potential for applying various types of artificial intelligence techniques, such as enhanced models of deep learning, for classifying Alzheimer’s disease. Therefore, this study proposes an Outline of deep learning to classify Alzheimer’s disease with segmentation using the Multi-Layer U-Net and a hybrid classification approach combining multi-scale EfficientNet with SVM. The proposed methodology consists of a four-phase process: (1) Whole brain segmentation, (2) Gray matter segmentation using multi-layer U-Net segmentation, (3) Feature extraction using Multi-Scale Efficient Net with SVM for classification, and (4) XAI (explainable AI) techniques by integrating Saliency Map Quantitative Analysis for increased clinical trustworthiness and model interpretability. It is found that the experiment results provide promising classification performance for three classes – Alzheimer’s Disease (AD), Mild Cognitive Impairment (MCI) and Cognitively Normal (CN) with an overall accuracy of 97.78% ± 0.54%, precision of 97.18% ± 1.14% (AD), 97.78% ± 0.29% (CN) and 97.03% ± 1.10% (MCI), recall of 97.90% ± 0.77% (AD), 97.49% ± 1.34% (CN) and 97.25% ± 0.99% (MCI), and F1 score of 97.74% ± 0.63% (AD), 97.78% ± 0.79% (CN), and 97.54% ± 0.69%(MCI). The results obtained underscore the elegance of the proposed approach in correctly classifying Alzheimer’s disease stages. Future work will evaluate the model on publicly accessible Alzheimer’s disease MRI datasets and incorporate advanced XAI techniques for increased interpretability and diagnostic reliability. The work focuses on Human Health. 𝐀𝐮𝐭𝐡𝐨𝐫𝐬: Muhammad Zubair, Arfan Jaffar, Dr. Sadaf Hussain, @sheeraz Akra https://lnkd.in/duwBcpTu

Did you miss our session with BrainSightAI on how they're redefining the way clinicians investigate neurological and psychiatric conditions? Laina Emmanuel, Co-founder and CEO of BrainSightAI, revealed the powerful connection between technology and brain science, showing exactly how they're using AI to improve the diagnosis, prognosis, and treatment of complex disorders. Learn what drives real patient outcomes. Watch the session recording to catch up on all the insights: https://lnkd.in/gSPYp2g6 Applied AI Association (AAIA) Marv Su Maury Rogow Anand V. Suyambu Selvi #AAIA #AAIAIndiaChapter #AIForAll #ContinuousLearning

Recent advances in neuroscience, supported by AI, have enabled researchers to identify the precise moment when the brain commits to a decision. By analyzing neural activity in the frontal cortex of rats, scientists observed a distinct shift from processing external information to a committed choice, marking the end of deliberation. This discovery provides a new biomarker for decision-making and offers valuable insights into attention, impulsivity, and cognitive disorders. The findings may inform future approaches to diagnosing and treating conditions such as ADHD and schizophrenia, and could inspire the development of more adaptive AI decision-making systems.

Mouse brain map: Researchers have used an AI model to map a mouse brain in unprecedented detail, revealing previously-unknown subregions. The map will allow scientists to pinpoint more precisely which parts of the brain govern specific functions, behaviors, and diseases. The AI model, CellTransformer, was built by UC San Francisco and the Seattle-based Allen Institute using the same underlying framework as tools like ChatGPT. However, instead of evaluating relationships between words, CellTransformer analyzes relationships between neighboring cells. Drawing solely on large datasets, the model identified 1300 brain regions and subregions, including uncharted areas that may provide insight into movement disorders, memory loss, and other neurological conditions. One researcher compared the advance to moving from a map that only shows continents & countries to one that reveals states & cities. https://lnkd.in/gQJxD773