📚 Syllabus & Course Curriculum
AI & Machine Learning in Healthcare
Module-by-module breakdown of Integrating Machine Learning and Mathematical Computing for Advanced Applications in Medical Imaging, from foundations to a certified capstone project.
Formation
• Acquisition physics for CT, MRI and ultrasound and the artefacts each produces
• DICOM structure, spacing, orientation and the metadata errors that break pipelines
• Reconstruction as an inverse problem, with filtered back projection as the reference
Processing
• Intensity normalisation, bias field correction and resampling
• Filtering, denoising and the total variation and wavelet approaches
• Registration: rigid, affine and deformable, with the similarity metrics used
Segmentation
• Thresholding, region growing and level sets before the deep learning era
• U-Net and its variants, and why they still dominate medical segmentation
• Dice, Hausdorff distance and inter-observer variability as the accuracy ceiling
Learning
• Patch-based and 3D training under severe memory constraints
• Small datasets, augmentation and self-supervised pretraining
• Radiomics feature extraction and the reproducibility problem it has
Assessment
• Patient-level splits, because slice-level splitting leaks systematically
• Scanner and protocol shift between sites as the usual cause of failure
• Reader studies, regulatory expectations and reporting under CLAIM
Earn government-registered certification in Integrating Machine Learning and Mathematical Computing for Advanced Applications in Medical Imaging
e-Certificate and e-Marksheet issued on successful completion.
