Improving the fairness and robustness of deep learning classification models through bias mitigation and post hoc calibration - Takeaways - MDSpire
Coming Soon: Introducing MDSpire News. Learn more
Conexiant’s news site is now MDSpire News. Learn more

Enhancing Fairness and Robustness in Deep Learning Classifiers with Bias Mitigation and Post Hoc Calibration

  • By

  • Majid Ramedani

  • Jaya C. Terli

  • Ole H. Martensen

  • Tobias Strauß

  • Katharina Buerger

  • Peter Dechent

  • Emrah Düzel

  • Klaus Fliessbach

  • Maria Gemenetzi

  • Wenzel Glanz

  • Julian Hellmann-Regen

  • Stefan Hetzer

  • Enise I. Incesoy

  • Daniel Janowitz

  • Marie T. Kronmüller

  • Christoph Laske

  • Falk Lüsebrink

  • Robert Perneczky

  • Oliver Peters

  • Lukas Preis

  • Josef Priller

  • Boris-Stephan Rauchmann

  • Ayda Rostamzadeh

  • Anja Schneider

  • Björn H. Schott

  • Sebastian Sodenkamp

  • Annika Spottke

  • Eike J. Spruth

  • Sophia Stöcklein

  • Jens Wiltfang

  • Frank Jessen

  • Stefan J. Teipel

  • Martin Dyrba

  • October 17, 2026

Share

  • 1

    The study examined age-related confounding and probability calibration as two challenges to trustworthiness in MRI-based Alzheimer disease classification.

  • 2

    Models used whole-brain, voxel-wise gray matter probability maps from T1-weighted MRI scans in the ADNI2/GO, AIBL, and DELCODE cohorts.

  • 3

    Age was the sole confounding variable explicitly addressed, reflecting concern that classifiers could rely on normal aging patterns that overlap with Alzheimer disease-related atrophy.

  • 4

    Three classification models shared a common 3-dimensional convolutional neural network feature extractor to enable comparison of bias-mitigation strategies.

  • 5

    The study evaluated four post hoc probability calibration methods, comprising two parametric and two nonparametric approaches.

Original Source(s)

Improving the fairness and robustness of deep learning classification models through bias mitigation and post hoc calibration

  • by Majid Ramedani, Jaya C. Terli, Ole H. Martensen, Tobias Strauß, Katharina Buerger, Peter Dechent, Emrah Düzel, Klaus Fliessbach, Maria Gemenetzi, Wenzel Glanz, Julian Hellmann-Regen, Stefan Hetzer, Enise I. Incesoy, Daniel Janowitz, Marie T. Kronmüller, Christoph Laske, Falk Lüsebrink, Robert Perneczky, Oliver Peters, Lukas Preis, Josef Priller, Boris-Stephan Rauchmann, Ayda Rostamzadeh, Anja Schneider, Björn H. Schott, Sebastian Sodenkamp, Annika Spottke, Eike J. Spruth, Sophia Stöcklein, Jens Wiltfang, Frank Jessen, Stefan J. Teipel, Martin Dyrba

  • October 17, 2026

Related Content

Letter to Editor External Clinical Validation of the RECOVER Research Index: A Response to Goldman and Martin

Acute myelopathy following chimeric antigen receptor (CAR) T cell therapy: an unusual diagnostic challenge

Decreasing Trends in Birth Prevalence and Severity of Cerebral Palsy in Sweden During 2000–2019: A Population-Based Registry Study