Skip to main navigation Skip to search Skip to main content

Adaptive Conformal Prediction for Reliable and Explainable Medical Image Classification

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Deep learning models for medical imaging often exhibit overconfidence, creating safety risks in ambiguous diagnostic scenarios. While Conformal Prediction (CP) provides distribution-free statistical guarantees, standard methods like Regularized Adaptive Prediction Sets (RAPS) optimize for average efficiency, masking severe failures on difficult inputs. We propose an Adaptive Lambda Criterion for RAPS that minimizes the worst-case coverage violation across prediction set size strata. On OrganAMNIST (58,850 abdominal CT images, 11 classes), standard size-optimized RAPS converges to near-deterministic behavior with stratified undercoverage on uncertain samples, while our method achieves 95.72% global coverage with average set size 1.09 and ≥90% coverage across all strata. Cross-domain validation on PathMNIST (107,180 pathology images, 9 classes) confirms generalizability. Quantitative Grad-CAM analysis (ρ=-0.30, p<10-22) confirms that multi-label predictions correspond to focused attention on anatomically ambiguous regions.

Original languageEnglish
Title of host publicationAdvances and Trends in Artificial Intelligence. Theory and Applications - 39th International Conference on Industrial, Engineering and Other Applications of Applied Intelligent Systems, IEA/AIE 2026, Proceedings
EditorsHamido Fujita, Ali Selamat, Masitah Ghazali, Moonis Ali
PublisherSpringer Science and Business Media Deutschland GmbH
Pages70-81
Number of pages12
ISBN (Print)9789819228904
DOIs
Publication statusPublished - 2027
Event39th International Conference on Industrial, Engineering and Other Applications of Applied Intelligent Systems, IEA/AIE 2026 - Kuala Lumpur, Malaysia
Duration: 6 Jul 20268 Jul 2026

Publication series

NameLecture Notes in Computer Science
Volume16615 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference39th International Conference on Industrial, Engineering and Other Applications of Applied Intelligent Systems, IEA/AIE 2026
Country/TerritoryMalaysia
CityKuala Lumpur
Period6/07/268/07/26

Keywords

  • Conformal Prediction
  • Explainable AI
  • Medical Image Classification
  • Stratified Coverage
  • Uncertainty Quantification

Fingerprint

Dive into the research topics of 'Adaptive Conformal Prediction for Reliable and Explainable Medical Image Classification'. Together they form a unique fingerprint.

Cite this