RISED: A Pre-Deployment Evaluation Framework for High-Stakes AI Decision-Support Systems, with Application to Healthcare
Not on the map. Why?
Abstract (arXiv)
Clinical decision-support systems are expert systems whose recommendations clinicians act on directly, yet they are usually cleared on one aggregate accuracy number from a held-out test set. That number says nothing about input reliability under encoding shifts, subgroup gaps, threshold sensitivity, or operational feasibility. We present RISED, a pre-deployment evaluation framework operationalising five dimensions (Reliability, Inclusivity, Sensitivity, Equity, Deployability) through BCa bootstrap 95% confidence intervals, literature-grounded thresholds, and Holm-Bonferroni-corrected PASS / FAIL / INCONCLUSIVE verdicts; Equity is a proxy-dependence diagnostic rather than a gating test. Applied to seven cohorts spanning 35 years (n from 303 to 99,492), RISED surfaces failures invisible to AUROC: on Diabetes 130, Reliability passes by three orders of magnitude (PSS = 0.0004) while Inclusivity (AUC parity gap = 0.262) and Sensitivity (max threshold-flip rate 49.1%) fail decisively; both NHIS cohorts reproduce this. NHANES 2021-2023, with a complete feature profile, achieves INCONCLUSIVE verdicts; BRFSS 2024 produces the suite's most severe Sensitivity failure (max threshold-flip rate 64.2%) after instrument rotation removed hypertension and cholesterol. The pattern recurs on credit- and income-prediction cohorts, confirming domain-agnosticity; a multi-model check shows the failures are data-driven, not model-specific. RISED ships as an open-source Python package complementing TRIPOD+AI, FUTURE-AI, and Fairlearn with the structured numerical evidence those standards require but do not prescribe.
Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Builds on 24 of its 126 references that are in the map. First posted on arXiv 13 May 2026.
Builds on
24 of 126 references.
References · 24
- 2017A Unified Approach to Interpreting Model Predictionscited by 6,488
- 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,326
- 2016XGBoost: A Scalable Tree Boosting Systemcited by 5,120
- 2017On Calibration of Modern Neural Networkscited by 4,860
- 2016Equality of Opportunity in Supervised Learningcited by 2,000
- 2019A Survey on Bias and Fairness in Machine Learningcited by 1,911
- 2018Datasheets for Datasetscited by 1,371
- 2018Model Cards for Model Reportingcited by 1,068
- 2020Beyond Accuracy: Behavioral Testing of NLP models with CheckListcited by 979
- 2016Fair prediction with disparate impact: A study of bias in recidivism prediction instrumentscited by 671
- 2020Towards Faithfully Interpretable NLP Systems: How should we define and evaluate faithfulness?cited by 590
- 2021Retiring Adult: New Datasets for Fair Machine Learningcited by 443
12 more
- 2017On Fairness and Calibrationcited by 406
- 2020Closing the AI Accountability Gap: Defining an End-to-End Framework for Internal Algorithmic Auditingcited by 380
- 2018A comparative study of fairness-enhancing interventions in machine learningcited by 208
- 2019Mitigating Bias in Algorithmic Hiring: Evaluating Claims and Practicescited by 170
- 2020Challenges in Deploying Machine Learning: a Survey of Case Studiescited by 151
- 2018Aequitas: A Bias and Fairness Audit Toolkitcited by 145
- 2019What Clinicians Want: Contextualizing Explainable Machine Learning for Clinical End Usecited by 134
- 2023Fairlearn: Assessing and Improving Fairness of AI Systemscited by 74
- 2023FUTURE-AI: International consensus guideline for trustworthy and deployable artificial intelligence in healthcarecited by 59
- 2020An Empirical Characterization of Fair Machine Learning For Clinical Risk Predictioncited by 45
- 2019"The Human Body is a Black Box": Supporting Clinical Decision-Making with Deep Learningcited by 39
- 2020FairLens: Auditing Black-box Clinical Decision Support Systemscited by 23
Built on by
Citing papers per year: 2026: 1.
1 paper.