CL101FOUNDATIONCLINICAL

AI for Medical Analysis 1: Foundations and Applied Machine Learning

Applications, evaluation and limitations of machine learning in clinical data analysis.

Course Information

Course Information
ModeInstructor-led
TermSP1 · Jan–Feb
Duration8 weeks over 2 months · about 32 hours
Credits1.5 credits
Counts towardStandalone course
PrerequisitesBasic medical knowledge and computer literacy. No prior AI or programming; basic statistics helpful.
Who it is forB.Tech and M.Tech students, engineers, MBBS students from second year, interns, residents and practising clinicians.
FacultyDr. Chandra Sekhar Vorugunti
AssessmentWeekly quizzes 20% · two assignments 50% · mini-project presentation 30%. 60% to pass.

About the course

Clinicians and engineers learn together to frame clinical prediction problems, prepare data, evaluate predictions and build introductory tabular and imaging models. Every concept is anchored to a clinical case, and no prior programming is required.

Learning outcomes

  • Distinguish machine learning, deep learning and generative AI, and match each to medical problems
  • Prepare, split and label medical data, and recognise leakage, shortcut learning and biased sampling
  • Interpret sensitivity, specificity, predictive values, ROC, calibration and decision curves
  • Compare classical models with neural networks and fine-tune an imaging classifier
  • Present a clinical mini project with justified thresholds, error analysis and limitations

Course modules

8 modules

  • AI types

  • decision support

  • the medical AI lifecycle

Labs and projects: Weekly Colab notebooks on public data; two assignments and a mini project.

Books and readings: Course readings