📚 Syllabus & Course Curriculum
AI & Machine Learning in Healthcare
Module-by-module breakdown of AI in Medical Imaging and Diagnostics, from foundations to a certified capstone project.
Imaging Physics
• Acquisition physics of CT, MR, ultrasound and radiography, and the artefacts each produces
• DICOM structure, windowing, spacing and the metadata models silently depend on
• Reconstruction and dose settings as hidden covariates across sites
Preparation
• Annotation protocols and inter-reader variability as the real performance ceiling
• Weak labels from reports, and the noise that introduces
• Patient-level splitting to prevent the same patient appearing in train and test
Models
• U-Net style segmentation and nnU-Net as a strong default baseline
• Detection for lesions and nodules, and evaluation with FROC rather than accuracy
• Transfer learning from natural images and where it stops helping
Evaluation
• Sensitivity, specificity and the effect of disease prevalence on predictive value
• Reader studies, standalone versus assisted performance
• External validation and the scanner-generalisation failures that recur in the literature
Clinical Use
• Regulatory classification and the evidence expected for imaging AI
• PACS integration, worklist prioritisation and turnaround-time effects
• Monitoring after deployment as scanners, protocols and case mix change
Earn government-registered certification in AI in Medical Imaging and Diagnostics
e-Certificate and e-Marksheet issued on successful completion.
