Imaging modalities: X-ray, CT, MRI, Ultrasound, PET

Image formation principles (conceptual)

Resolution, contrast, noise & artifacts

Clinical use cases: 1) TB detection (X-ray) 2) Stroke evaluation (CT) 3) Tumor imaging (MRI)

Radiology workflow: 1) Order → Scan → PACS → Reporting 2)PACS, RIS, HIS overview

Where AI helps vs where AI is risky

Medical imaging data formats: DICOM&NIfTI

DICOM metadata, tags & PHI

De-identification & patient privacy

Image preprocessing: 1) Normalization 2)Resampling 3) Windowing

Annotation standards & inter-observer variability

Dataset bias, leakage & quality risks

CNN fundamentals (medical context)

U-Net architecture (mandatory)

Image classification vs segmentation

Transfer learning

Class imbalance in medical datasets

Imaging AI pipeline design

Data → Model → Inference flow

Medical image transforms

Loss functions (Dice, Focal – conceptual)

Training & inference workflows

Evaluation metrics: 1) Dice, IoU 2) Sensitivity, specificity

Error analysis & failure modes

Explainability (Grad-CAM concepts)

Why accuracy alone is dangerous

AI as medical software

CDS vs diagnostic AI

Data privacy & consent

Bias, fairness & patient safety

Overview of regulatory pathways

TB detection (X-ray)

Brain tumor segmentation (MRI)

Stroke detection (CT)

Retinal disease detection