Interpretable Machine Learning in Healthcare

By Muhammad Aurangzeb Ahmad, Ankur Teredesai, Carly Eckert. 2018

690 citations recorded by the source corpus.

This tutorial extensively covers the definitions, nuances, challenges, and requirements for the design of interpretable and explainable machine learning models and systems in healthcare. We discuss many uses in which interpretable machine learning models are needed in healthcare and how they should be deployed. Additionally, we explore the landscape of recent advances to address the challenges model interpretability in healthcare and also describe how one would go about choosing the right interpretable machine learnig algorithm for a given problem in healthcare.

Subfields: Artificial Intelligence, Health Informatics

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

Citations in the corpus

Toward Interpretable-AI Policies Using Evolutionary Nonlinear Decision Trees for Discrete-Action Systems (2020, background): “The researchers now demand an explanation of why and how they work [10], [11].”
InterpretML: A Unified Framework for Machine Learning Interpretability (2019, background): “This can easily be observed in high-risk applications such as healthcare (Ahmad et al., 2018; Caruana et al., 2015), ﬁnance (Hajek, 2019; Chen et al., 2018) and judicial environments (Tan et al., 2018; Soundarajan and Clausen).”
Transformers in Medical Imaging: A Survey (2022, background): “These models have been extensively leveraged in medical domain for image segmentation [49], reconstruction [50], disease classification [51], enhancement [52], and anomaly detection [53] due to their interpretability with solid mathematical foundations and theoretical supports on the robustness, rec”
Interpretability of machine learning‐based prediction models in healthcare (2020, background): “Moreover, traits like fairness, privacy, reliability or robustness, causality and trust should be considered when developing ML based prediction models (Doshi-Velez & Kim, 2017; Ahmad et al. 2018).”
What Clinicians Want: Contextualizing Explainable Machine Learning for Clinical End Use (2019, background): “Ahmad et al. (2018) conduct the most extensive survey of explainable methods in clinical healthcare, and analyze them in the context of existing notions in general ML and how they relate to clinical healthcare, discussing ethical questions thereof.”
Explainable, trustworthy, and ethical machine learning for healthcare: A survey (2021, background): “This human-machine combination is a powerful tool to facilitate the patients and develop high-quality treatments [21].”
CHAOS Challenge - Combined (CT-MR) Healthy Abdominal Organ Segmentation (2020, background): “, 2018), and focusing on interpretable and explainable solutions (Samek, 2019; Ahmad et al., 2018)”
"The human body is a black box": supporting clinical decision-making with deep learning (2019, background): “ed to have substantial technical and quantitative expertise with which to engage with explainable machine learning and “to interrogate, understand, debug and even improve the machine learning system” [2]. A potential by-product of this engagement often left undiscussed is whether physicians want to ”
Privacy and artificial intelligence: challenges for protecting health information in a new era (2021, background): “Notably, in response to this problem, many researchers have been developing interpretable forms of AI that will be easier to integrate into medical care [24].”
Federated learning for medical image analysis: A survey (2023, background): “9) Model Interpretability: In clinical practice, one of the most significant hurdles in adopting machine learning and AI lies in the “black box” nature of certain machine learning systems, such as deep learning [182], [183].”
The role of explainability in creating trustworthy artificial intelligence for health care: a comprehensive survey of the terminology, design choices, and evaluation strategies (2020, methodology): “ng framework using Meijer G-function parametrization, minimizing the metamodeling loss via gradient descent. Instead of serving as explanation, global surrogate models can also replace the task model [7]. LIME (local interpretable model-agnostic explanations) and Anchors are examples of local surrog”
Explainable AI (XAI): A Systematic Meta-Survey of Current Challenges and Future Opportunities (2021, methodology): “The work in [26] has mentioned that the inclusion of end-users in the design of black-box AI models is essential, especially for specific domains, e.”