Issam A
- Research Program Mentor
MS candidate at University of North Carolina at Charlotte
Expertise
Medical AI | Digital Twins | Computer Vision
Bio
Hey! I’m Issam, AI student at UNC Charlotte, where I work with the AI4Health Center on digital twin research. I’m currently collaborating with the UNC School of Medicine and Duke Heart Center, with the goal of using AI to save lives through applied AI research. I’m planning to pursue a PhD to keep pushing this research forward. Outside academia, I’ve gained industry experience in both corporate tech and entrepreneurship. I previously founded a hardware startup, scaled it, and led it through a successful exit. On the side, I’m all about FPV drones, aviation, 3D printing, and restoring vintage Apple devices.Project ideas
Silent Seizure AI Detector
Project Description: Build a prototype that uses AI to detect non-visible seizures (also known as "silent" or "absence" seizures) from biometric data such as heart rate, EEG, or facial expressions. Skills Learned: Biomedical signal processing, time-series analysis, machine learning, healthcare data ethics. Information Gathering: Students will research existing biometric seizure datasets (e.g., CHB-MIT Scalp EEG Database), explore seizure symptoms, and review clinical case studies. Outcome: Prototype seizure detection model + short research summary explaining clinical relevance and detection accuracy.
AI Stethoscope Analyzer
Project Description: Train an AI model to detect abnormal heart or lung sounds (such as murmurs or crackles) using open-source stethoscope audio recordings. Skills Learned: Audio signal processing, classification models, medical acoustics, dataset labeling. Information Gathering: Students will explore open datasets like the PhysioNet/CinC Challenge databases and learn how to process and label audio recordings of bodily sounds. Outcome: Working audio classifier + research paper on its clinical applications.
Diabetic Retinopathy Detection
Project Description: Use deep learning to analyze retinal images and identify early signs of diabetic retinopathy, a leading cause of vision loss. Skills Learned: Computer vision (CNNs), medical imaging interpretation, data preprocessing, ethical AI. Information Gathering: Students will use public datasets such as the Kaggle EyePACS dataset and review medical literature to understand grading of retinopathy severity. Outcome: Trained CNN model + presentation on diagnostic accuracy and real-world impact.