Meta Ray-Ban first-aid guidance: contactless heart rate (rPPG), facial droop detection, and Gemini voice agent with JRCALC 2022 clinical GraphRAG. 🏆 3rd place UnicornMafia To The Americas 2026.

This repository contains MATLAB code for remotely controlling a Vector Network Analyzer (VNA) to perform measurements on a head phantom equipped with an antenna helmet. The project aims to collect extensive data across multiple stroke configurations and phantom inclinations.

🧠 Stroke detection using CNN-based medical image analysis — deep learning for healthcare

Репозиторий реализует модели глубокого обучения для обнаружения ишемического инсульта на FLAIR МРТ-сканах путём анализа симметрии между полушариями мозга. Подход сочетает 2D/3D-CNN и топологический анализ данных для интерпретируемой диагностики.

NeuroStroke AI: An open-source deep learning system for automated brain stroke detection from CT scans. Features CNN with VGG19 transfer learning, achieving 92% accuracy. Includes interactive web interface built with Streamlit. Empowering healthcare with AI. 🧠💙 #HealthcareAI #DeepLearning

An interpretable facial asymmetry analysis pipeline using landmarks, deep learning, and temporal modeling to explore stroke-related facial patterns from images and videos.

🧠 Detect brain strokes in CT scans using deep learning. NeuroStroke AI enhances early diagnosis and improves patient care with advanced CNN technology.

Sleek UI with fine-tuned model using PyTorch based YOLOv7 application for stroke detection and monitoring old patients and targets at risk.

A stroke detection project developed using R.