mlsanityData to 7 Oct 2026

Detecting Shortcut Learning for Fair Medical AI using Shortcut Testing

arXiv 2207.10384 ↗First posted 21 Jul 2022cs.LGcited by 52

Not on the map. Why?

Abstract (arXiv)

Machine learning (ML) holds great promise for improving healthcare, but it is critical to ensure that its use will not propagate or amplify health disparities. An important step is to characterize the (un)fairness of ML models - their tendency to perform differently across subgroups of the population - and to understand its underlying mechanisms. One potential driver of algorithmic unfairness, shortcut learning, arises when ML models base predictions on improper correlations in the training data. However, diagnosing this phenomenon is difficult, especially when sensitive attributes are causally linked with disease. Using multi-task learning, we propose the first method to assess and mitigate shortcut learning as a part of the fairness assessment of clinical ML systems, and demonstrate its application to clinical tasks in radiology and dermatology. Finally, our approach reveals instances when shortcutting is not responsible for unfairness, highlighting the need for a holistic approach to fairness mitigation in medical AI.

Key facts. Cited by 52 papers in the map (data through 7 Oct 2026). Builds on 31 of its 66 references that are in the map. First posted on arXiv 21 Jul 2022.

Builds on

31 of 66 references.

References · 31

  1. 2015Domain-Adversarial Training of Neural Networkscited by 3,023
  2. 2020Shortcut Learning in Deep Neural Networkscited by 1,928
  3. 2014Unsupervised Domain Adaptation by Backpropagationcited by 1,906
  4. 2019CheXpert: A Large Chest Radiograph Dataset with Uncertainty Labels and Expert Comparisoncited by 1,410
  5. 2017ChestX-ray8: Hospital-scale Chest X-ray Database and Benchmarks on Weakly-Supervised Classification and Localization of Common Thorax Diseasescited by 969
  6. 2020Language (Technology) is Power: A Critical Survey of "Bias" in NLPcited by 951
  7. 2019Big Transfer (BiT): General Visual Representation Learningcited by 766
  8. 2021Just Train Twice: Improving Group Robustness without Training Group Informationcited by 522
  9. 2020Underspecification Presents Challenges for Credibility in Modern Machine Learningcited by 488
  10. 2018Confounding variables can degrade generalization performance of radiological deep learning modelscited by 374
  11. 2021Reading Race: AI Recognises Patient's Racial Identity In Medical Imagescited by 155
  12. 2019A deep learning system for differential diagnosis of skin diseasescited by 118
19 more
  1. 2021Re-imagining Algorithmic Fairness in India and Beyondcited by 102
  2. 2020Estimating Example Difficulty Using Variance of Gradientscited by 96
  3. 2021Towards Interpreting and Mitigating Shortcut Learning Behavior of NLU Modelscited by 78
  4. 2018Achieving Fairness through Adversarial Learning: an Application to Recidivism Predictioncited by 77
  5. 2017Predicting Cardiovascular Risk Factors from Retinal Fundus Photographs using Deep Learningcited by 63
  6. 2021Causally motivated Shortcut Removal Using Auxiliary Labelscited by 55
  7. 2022Improving the Fairness of Chest X-ray Classifierscited by 52
  8. 2022Diagnosing failures of fairness transfer across distribution shift in real-world medical settingscited by 45
  9. 2020Automatic Shortcut Removal for Self-Supervised Representation Learningcited by 43
  10. 2018Blind Justice: Fairness with Encrypted Sensitive Attributescited by 41
  11. 2020Deep Learning Applied to Chest X-Rays: Exploiting and Preventing Shortcutscited by 29
  12. 2019This Thing Called Fairness: Disciplinary Confusion Realizing a Value in Technologycited by 24
  13. 2022Write It Like You See It: Detectable Differences in Clinical Notes By Race Lead To Differential Model Recommendationscited by 20
  14. 2021A Near-Optimal Algorithm for Debiasing Trained Machine Learning Modelscited by 19
  15. 2021Removing biased data to improve fairness and accuracycited by 17
  16. 2020Joint Optimization of AI Fairness and Utility: A Human-Centered Approachcited by 10
  17. 2021Does "AI" stand for augmenting inequality in the era of covid-19 healthcare?cited by 10
  18. 2021Algorithmic encoding of protected characteristics in image-based models for disease detectioncited by 7
  19. 2021Model-based metrics: Sample-efficient estimates of predictive model subpopulation performancecited by 6

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 2; 2023: 3; 2024: 15; 2025: 16; 2026: 16.

52 papers.

Sort:

By area:

  1. 2023Fast Diffusion-Based Counterfactuals for Shortcut Removal and Generationcited by 41
  2. 2022Addressing Fairness Issues in Deep Learning-Based Medical Image Analysis: A Systematic Reviewcited by 27
  3. 2024Shortcut Learning in Medical Image Segmentationcited by 16
  4. 2024Slicing Through Bias: Explaining Performance Gaps in Medical Image Analysis using Slice Discovery Methodscited by 14
  5. 2024BiasPruner: Debiased Continual Learning for Medical Image Classificationcited by 12
  6. 2024Mind the Graph When Balancing Data for Fairness or Robustnesscited by 9
  7. 2024A Systematic Review of Generalization Research in Medical Image Classificationcited by 8
  8. 2023Are demographically invariant models and representations in medical imaging fair?cited by 6
  9. 2024Improving Intervention Efficacy via Concept Realignment in Concept Bottleneck Modelscited by 6
  10. 2025Physics-informed Temporal Alignment for Auto-regressive PDE Foundation Modelscited by 5
  11. 2022Manifestations of Xenophobia in AI Systemscited by 4
  12. 2023Towards objective and systematic evaluation of bias in artificial intelligence for medical imagingcited by 4
38 more
  1. 2025Shortcut Learning Susceptibility in Vision Classifierscited by 4
  2. 2024Evaluating Model Bias Requires Characterizing its Mistakescited by 3
  3. 2025Gaze-Guided Learning: Avoiding Shortcut Bias in Visual Classificationcited by 3
  4. 2024Predicting risk of cardiovascular disease using retinal OCT imagingcited by 2
  5. 2024Fairness Evolution in Continual Learning for Medical Imagingcited by 2
  6. 2024Positive-Sum Fairness: Leveraging Demographic Attributes to Achieve Fair AI Outcomes Without Sacrificing Group Gainscited by 2
  7. 2025Ensuring Medical AI Safety: Interpretability-Driven Detection and Mitigation of Spurious Model Behavior and Associated Datacited by 2
  8. 2025Spurious Correlations and Beyond: Understanding and Mitigating Shortcut Learning in SDOH Extraction with Large Language Modelscited by 2
  9. 2026Feature salience - not task-informativeness - drives machine learning model explanationscited by 2
  10. 2026Learning beyond Site Bias for OOD Generalization in Brain Networkscited by 2
  11. 2024Common-Sense Bias Modeling for Classification Taskscited by 1
  12. 2024Clinical Domain Knowledge-Derived Template Improves Post Hoc AI Explanations in Pneumothorax Classificationcited by 1
  13. 2025Machine Learning Applications to Diffuse Reflectance Spectroscopy in Optical Diagnosis; A Systematic Reviewcited by 1
  14. 2025MOSCARD -- Causal Reasoning and De-confounding for Multimodal Opportunistic Screening of Cardiovascular Adverse Eventscited by 1
  15. 2025Correct-By-Construction: Certified Individual Fairness through Neural Network Trainingcited by 1
  16. 2025RoentMod: A Synthetic Chest X-Ray Modification Model to Identify and Correct Image Interpretation Model Shortcutscited by 1
  17. 2025Did Models Learn Sufficiently? Attribution-Guided Training via Subset-Selected Counterfactual Augmentationcited by 1
  18. 2026Clever Materials: When Models Identify Good Materials for the Wrong Reasonscited by 1
  19. 2026Prevalence calibration as shortcut mitigationcited by 1
  20. 2026Generating Chest X-Ray Counterfactuals by Specialising Foundation Image Modelscited by 1
  21. 2024Transfer Learning for T-Cell Response Predictioncited by 0
  22. 2024Interdisciplinary Expertise to Advance Equitable Explainable AIcited by 0
  23. 2024Integrating Deep Learning with Fundus and Optical Coherence Tomography for Cardiovascular Disease Predictioncited by 0
  24. 2025A Causal Framework for Aligning Image Quality Metrics and Deep Neural Network Robustnesscited by 0
  25. 2025Gradient-based Model Shortcut Detection for Time Series Classificationcited by 0
  26. 2025Beyond Ethics: How Inclusive Innovation Drives Economic Returns in Medical AIcited by 0
  27. 2025Mitigating representation bias caused by missing pixels in methane plume detectioncited by 0
  28. 2025Localising Shortcut Learning in Pixel Space via Ordinal Scoring Correlations for Attribution Representations (OSCAR)cited by 0
  29. 2025Fairness Evaluation of Risk Estimation Models for Lung Cancer Screeningcited by 0
  30. 2026Mitigating Shortcut Learning via Feature Disentanglement in Medical Imaging: A Benchmark Studycited by 0
  31. 2026Stable but Wrong: When Learning Stabilizes Away from the Truthcited by 0
  32. 2026Shortcut Learning in Glomerular AI: Adversarial Penalties Hurt, Entropy Helpscited by 0
  33. 2026Inductive Subgraphs as Shortcuts: Causal Disentanglement for Heterophilic Graph Learningcited by 0
  34. 2026Robust Mitigation of Age-Dependent Confounding Effects via Sample-Difficulty Decorrelationcited by 0
  35. 2026SHOVIR: A Benchmark for Evaluating Vision Shortcut Learning in Radiology Report Generationcited by 0
  36. 2026Adversarial LassoNet: Robust Feature Selection via Stability-Driven Sparse Learningcited by 0
  37. 2026ReMoDEx: A Local-to-Global Relevance-Based Model Decision Explainability Framework for large-Scale Image Datasetscited by 0
  38. 2026Why ML-based cough models do not generalize: a systematic cross-dataset evaluation for tuberculosis screeningcited by 0

50 most cited of 52 All 52 (JSON)