Syllabus
Introduction to Deep Learning in Healthcare, covering diagnostic, prognostic, and personalized medicine applications with real-world case studies; Biomedical Data Types and Challenges, focusing on imaging, signals, text, and omics data along with issues like annotation, bias, and class imbalance; Tools and Frameworks for Implementation, introducing Python, PyTorch/TensorFlow, MONAI, and MedPy; CNNs for Medical Image Classification, including transfer learning and fine-tuning; Tumor Detection and Grading, with hands-on case studies using MRI/CT; Semantic Segmentation in Radiology, exploring U-Net and organ-specific tasks; 3D Medical Image Analysis, focusing on volumetric data and 3D architectures; Image Registration and Localization, covering alignment and landmark detection; Time-Series Signal Processing, centered on ECG, EEG, and wearable sensor data; RNNs and LSTMs for Biomedical Signals, including arrhythmia detection; Sleep Stage Classification using EEG, highlighting models like DeepSleepNet; Multimodal Biosignal Fusion, integrating multiple physiological signals for diagnosis; Clinical NLP and Predictive Modeling, including BERT and transformer-based approaches for EHR analysis; Explainable AI in Healthcare, introducing GradCAM, SHAP, and LIME for model interpretability; and Project-Based Learning, with hands-on demos such as retinal image analysis, biosignal-based stress/seizure detection, and a clinical triage assistant.