{"repo":"Tisha-runwal/Personalized-Federated-Learning-for-Privacy--Preserving-and-Scalable-IoT-Driven-Smart-Healthcare","free":true,"listed":false,"github":"https://github.com/Tisha-runwal/Personalized-Federated-Learning-for-Privacy--Preserving-and-Scalable-IoT-Driven-Smart-Healthcare","clone":"git clone https://github.com/Tisha-runwal/Personalized-Federated-Learning-for-Privacy--Preserving-and-Scalable-IoT-Driven-Smart-Healthcare.git","description":"A Personalized Federated Learning (PFL-HCare) framework for IoT healthcare. Features MAML meta-learning, Differential Privacy (RDP), and gradient quantization for efficiency. Includes a React/FastAPI dashboard for real-time monitoring.","language":"Python","stars":266,"topics":["differential-privacy","fastapi","federated-learning","iot-healthcare","machine-learning","maml","meta-learning","privacy-preserving","pytorch","react"],"license":"MIT","category":"machine-learning","readme_excerpt":"  🏥 PFL-HCare Personalized Federated Learning for Privacy-Preserving and Scalable IoT-Driven Smart Healthcare TTEH LAB · School of Engineering, Dayananda Sagar University Bangalore – 562112, Karnataka, India     Prototype implementation of: \"Personalized Federated Learning for Privacy-Preserving and Scalable IoT-Driven Smart Healthcare\" ICICI-2025, IEEE Xplore · DOI: 10.1109/ICICI65870.2025.11069877   --- 🔭 Overview The exponential growth of IoT in healthcare has transformed patient monitoring and diagnosis, but traditional centralized machine learning methods present critical challenges in data privacy, scalability, and adaptability to diverse patient conditions. This work presents PFL-HCare , a Personalized Federated Learning framework for IoT-driven smart healthcare that enforces \"train locally, share globally\" through four integrated components: a MAML-based meta-learning personalizer for adaptive model customization, a Differential Privacy mechanism with RDP accounting for formal privacy guarantees, k-bit gradient quantization for communication-efficient updates, and adaptive client selection based on gradient norms for optimal convergence. All components are trained via Federated Learning — ensuring that sensitive patient data never leaves the edge device. The framework includes a real-time React + FastAPI dashboard for live visualization of convergence, privacy budget, communication overhead, and per-client metrics across five FL methods. Experime","default_branch":null,"files":null,"tree":[],"storefront":"/r/Tisha-runwal","claimed":false,"request_supported":{"post":"https://gitbuyer.com/r/Tisha-runwal/Personalized-Federated-Learning-for-Privacy--Preserving-and-Scalable-IoT-Driven-Smart-Healthcare/request-supported","requests":0},"note":"indexed from public GitHub; nothing is for sale on this page. Clone it from GitHub. Paid listings live at /search."}