Explainable AI in Biomedical Predictions — Hands-on

Build interpretable, auditable AI for biomedical use-cases. Learn local & global explanation methods, counterfactual analysis, uncertainty quantification, calibration, bias/fairness auditing, and rigorous reporting so your models can be trusted in research and translational settings.

Explainable AI in Biomedical Predictions

Help Desk · WhatsApp

Session Index

Session 1 — XAI Foundations & Global vs Local Session 2 — Feature Attributions & Saliency Session 3 — Counterfactuals, Uncertainty & Fairness Session 4 — Mini Capstone: XAI Audit & Report

Session 1

Fee: Rs 13,200

XAI Foundations & Global vs Local

Why explainability? risk, accountability, reproducibility

research & clinical contexts audit trails

Global vs local explanations & evaluation

global importances local attributions fidelity/stability tests

Toolchain

SHAP, LIME Captum / ELI5 Alibi / DiCE

Session 2

Fee: Rs 18,400

Feature Attributions & Saliency

Model-agnostic & model-specific methods

permutation importance partial dependence/ICE accumulated local effects

Deep nets interpretability

Integrated Gradients Grad-CAM/CAM DeepLIFT/SmoothGrad

Reliability checks for explanations

sanity checks stability/robustness faithfulness metrics

Session 3

Fee: Rs 23,600

Counterfactuals, Uncertainty & Fairness

Counterfactual explanations & recourse

DiCE actionability/feasibility cost constraints

Uncertainty & calibration

aleatoric/epistemic temperature scaling conformal prediction

Fairness & bias audits

group/individual fairness thresholding & parity mitigation strategies

Session 4

Fee: Rs 29,600

Mini Capstone: XAI Audit & Report

End-to-end audit on a real bio model (omics/clinical)

Theory + Practical

Governance artifacts & documentation

model cards datasheets risk register

Deliverables: notebook, XAI dashboard & PDF brief

attribution plots counterfactuals calibration curves

PDF

Explainable AI in Biomedical Predictions Training | SHAP/LIME, Counterfactuals, Uncertainty, Fairness

Explainable AI in Biomedical Predictions — Hands-on