Ethical Machine Learning in Healthcare

By Irene Y. Chen, Emma Pierson, Sherri Rose, Shalmali Joshi, K. Ferryman, Marzyeh Ghassemi. Annual Review of Biomedical Data Science, 2020

577 citations recorded by the source corpus.

The use of machine learning (ML) in healthcare raises numerous ethical concerns, especially as models can amplify existing health inequities. Here, we outline ethical considerations for equitable ML in the advancement of healthcare. Specifically, we frame ethics of ML in healthcare through the lens of social justice. We describe ongoing efforts and outline challenges in a proposed pipeline of ethical ML in health, ranging from problem selection to postdeployment considerations. We close by summarizing recommendations to address these challenges.

Subfields: Artificial Intelligence, Health Informatics, Safety Research

View Paper: https://www.semanticscholar.org/paper/e8d330f11df9c69f38b78a7cc4b1333ebecf7c55

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Health Informatics

In Safety Research

Citations in the corpus

Coarse race data conceals disparities in clinical risk score performance (2023, methodology): “To quantify uncertainty in the performance of the machine learning methods, we report the 95% confidence interval across 1,000 random train-test splits, a widely used procedure (Chen et al., 2021a; Shan-mugam and Pierson, 2022).”
Selective Classification Can Magnify Disparities Across Groups (2020, background): “These group disparities are especially problematic in the same high-stakes areas where we might want to deploy selective classification, like medicine and criminal justice; there, poor performance on particular groups is already a significant issue (Chen et al., 2020; Hill, 2020).”
Algorithm fairness in artificial intelligence for medicine and healthcare (2023, background): “With the proliferation of AI algorithms in healthcare, there are growing ethical concerns regarding the disparate impact that the models may have on under-represented communities 1–8 .”
TRIPOD+AI statement: updated guidance for reporting clinical prediction models that use regression or machine learning methods (2024, methodology): “61 While TRIPOD 2015 is largely agnostic to the type of modelling approach, and much of its reporting recommendations apply equally to non-regression approaches, additional reporting considerations are needed for the growing class of machine learning methods.”
Towards Expert-Level Medical Question Answering with Large Language Models (2023, unknown): “The dataset was curated to draw on insights from literature on health equity in AI/ML and deﬁne a set of implicit and explicit adversarial queries that cover a range of patient experiences and health conditions [37–41].”
Fairness Violations and Mitigation under Covariate Shift (2019, background): “As a consequence, a growing body of work aims to develop algorithms embodying fairness principles specific to healthcare [9].”
Bias in artificial intelligence algorithms and recommendations for mitigation (2023, background): “countries such as Canada and France do not record race and ethnicity in their national health databases, making it difficult to account for less represented groups that may have different outcomes compared to the overall general population [22].”
WILDS: A Benchmark of in-the-Wild Distribution Shifts (2020, unknown): “…many of them are not due to distribution shifts, but to disparities in non-algorithmic factors (e.g., access to care or prevalence of comorbidities (Chen et al., 2020)) or to algorithmic problems unrelated to distribution shift (e.g., choice of a biased outcome variable (Obermeyer et al., 2019)).”