AI in Healthcare — 10 weekly modules, grounded in African healthcare contexts.
Define AI, ML, and deep learning. African healthcare landscape and challenges. AI success stories: Uganda, Kenya, Egypt.
Healthcare data types: structured, imaging, genomic. Sensitivity, specificity, AUC-ROC. Data quality, bias, and African data gaps.
ML paradigms and common algorithms. The ML workflow: training, validation, testing.
WHO, EU AI Act, OECD frameworks critiqued. Ubuntu philosophy and the CARE framework. Ethical dilemma debate: African healthcare scenarios.
Kenya Data Protection Act 2019 applied. Privacy-preserving techniques: federated learning. Data governance and community-based models.
How diagnostic AI works: pattern recognition and explainability. CDSS types and integration challenges. Case study: evaluating a malaria diagnostic AI at KNH.
CNNs, transfer learning, and image classification. Radiology, pathology, ophthalmology, obstetric ultrasound. Image interpretation lab: AI vs. expert radiologist.
AI for outbreak prediction, maternal and child health. NCD risk, supply chain, and workforce AI. Simulation: managing a cholera outbreak with AI tools.
AI platforms for clinical and research use. NLP in clinical documentation, ICD coding, and adverse event detection. Multilingual NLP challenges for Kiswahili and African languages.
Kenya National AI Strategy, KMPDC, and WHO ethics frameworks. Career positioning: research, clinical exposure, AI literacy. Capstone project presentations to an expert panel.