Intelligible Models for HealthCare: Predicting Pneumonia Risk and Hospital 30-day Readmission

By R. Caruana, Yin Lou, Johannes Gehrke, Paul Koch, Marc Sturm, Noémie Elhadad. 2015

1,942 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Medical & Health AI

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 Medical & Health AI

Citations in the corpus

Women also Snowboard: Overcoming Bias in Captioning Models (2018, background): “For example, [21] consider models which predict if pneumonia patients should be hospitalized.”
On Calibration of Modern Neural Networks (2017, background): “As a result, neural networks are now entrusted with making complex decisions in applications, such as object detection (Girshick,2015), speech recognition (Hannun et al.,2014), and medical diagnosis (Caruana et al., 2015). In these settings, neural networks are an essential component of larger decis”
Explainable Artificial Intelligence (XAI): Concepts, Taxonomies, Opportunities and Challenges toward Responsible AI (2019, neutral): “There are cases in which the lack of a proper understanding of the model might drive the user toward incorrect assumptions and fatal consequences [44,164] .”
Interpretable Machine Learning: Fundamental Principles and 10 Grand Challenges (2021, methodology): “Using a GAM, we might find counterintuitive patterns; e.g., as shown in Caruana et al. (2015), asthma patients fared better than non-asthma patients in a health outcomes study.”
A Survey on Explainable Artificial Intelligence (XAI): Toward Medical XAI (2019, background): “GDM (Generative Discriminative Models): ridge regression + least square [99] 7 7 GAM, GA2M (Generative Additive Model) [81], [101], [102] 7 7”
Fairwashing: the risk of rationalization (2019, background): “…use of machine learning models in high stakes decision-making systems ( e.g. , credit scoring [32], predictive justice [20] or medical diagnosis [9]) combined with proven risks of incorrect decisions – such as people being wrongly denied parole [35] – has considerably raised the public demand…”
Peeking Inside the Black-Box: A Survey on Explainable Artificial Intelligence (XAI) (2018, methodology): “[37] described an application of a learning method based on generalized additive models to the pneumonia problem.”
Deep k-Nearest Neighbors: Towards Confident, Interpretable and Robust Deep Learning (2018, background): “, the 5-th component f(x)[5] is P (y = 5 | θ, x).”
Interpreting Deep Classifier by Visual Distillation of Dark Knowledge (2018, background): “In contrast to deep neural networks, an alternative is to restrict the models to belong to a family that is inherently easy to interpret (Caruana et al., 2015; Doshi-Velez et al., 2014; Letham et al., 2015).”
A Survey of Methods for Explaining Black Box Models (2018, neutral): “A case study on health care showing the application of the GAM the refinement is presented in Reference [16].”
What do we need to build explainable AI systems for the medical domain? (2017, methodology): “A further example was presented by Caruana et al. (2015), where high-performance generalized additive models with pairwise interactions (GAMs) were applied to problems from the medical domain yielding intelligible models, which uncovered surprising patterns in the data that previously had prevented…”
Interpretability Beyond Feature Attribution: Quantitative Testing with Concept Activation Vectors (TCAV) (2017, background): “While option 1 offers simplicity as the explanation is embedded in the model (Kim et al., 2014; Doshi-Velez et al., 2015; Tibshirani, 1994; Zou et al., 2004; Ustun et al., 2013; Caruana et al., 2015), this option might be costly for users who already have a working high performance model.”
Distill-and-Compare: Auditing Black-Box Models Using Transparent Model Distillation (2017, methodology): “ a statistical test to determine if the audit data is missing key features used to train the black-box model. 5) An ancillary contribution of this paper is a new confidence interval estimate for iGAM [9, 24, 25], a type of transparent model, to compare two models of this class. 2 AUDIT APPROACH Our ”
Explainable Artificial Intelligence: Understanding, Visualizing and Interpreting Deep Learning Models (2017, background): “The authors of [7] show an example from this domain, where an AI system which was trained to predict the pneumonia risk of a person arrives at totally wrong conclusions.”
Interpretability via Model Extraction (2017, background): “Recent advances in machine learning have revolutionized our ability to use data to inform critical decisions, such as medical diagnosis [8, 19, 27], bail decisions for defendants [16, 17], and aircra collision avoidance systems [25].”
Methods for interpreting and understanding deep neural networks (2017, methodology): “Interpretability is especially important in applications such as medicine or self-driving cars, where the reliance of the model on the correct features must be guaranteed [15, 14].”
On the Safety of Machine Learning: Cyber-Physical Systems, Decision Sciences, and Data Products (2016, background): “Highly complex modeling techniques used today, including extreme gradient boostin g and deep neural networks, may pick up on those data vagaries in the learned models they produce to achieve high accuracy, but might fail due to an unknown shift in the data domain [26].”
RETAIN: An Interpretable Predictive Model for Healthcare using Reverse Time Attention Mechanism (2016, methodology): “There is a common belief that one has to trade accuracy for interpretability using one of three types of traditional models [6]: 1) identifying a set of rules (e.”