More Like this

1. An Explainable Artificial Intelligence Approach Using Graph Learning to Predict Intensive Care Unit Length of Stay

   https://doi.org/10.1287/isre.2023.0029  

   Guo, Tianjian; Bardhan, Indranil R; Ding, Ying; Zhang, Shichang (December 2024, Information Systems Research)

   We propose and test a novel graph learning-based explainable artificial intelligence (XAI) approach to address the challenge of developing explainable predictions of patient length of stay (LoS) in intensive care units (ICUs). Specifically, we address a notable gap in the literature on XAI methods that identify interactions between model input features to predict patient health outcomes. Our model intrinsically constructs a patient-level graph, which identifies the importance of feature interactions for prediction of health outcomes. It demonstrates state-of-the-art explanation capabilities based on identification of salient feature interactions compared with traditional XAI methods for prediction of LoS. We supplement our XAI approach with a small-scale user study, which demonstrates that our model can lead to greater user acceptance of artificial intelligence (AI) model-based decisions by contributing to greater interpretability of model predictions. Our model lays the foundation to develop interpretable, predictive tools that healthcare professionals can utilize to improve ICU resource allocation decisions and enhance the clinical relevance of AI systems in providing effective patient care. Although our primary research setting is the ICU, our graph learning model can be generalized to other healthcare contexts to accurately identify key feature interactions for prediction of other health outcomes, such as mortality, readmission risk, and hospitalizations. 

   more » « less
2. Clinical risk prediction using language models: benefits and considerations

   https://doi.org/10.1093/jamia/ocae030  

   Acharya, Angeela; Shrestha, Sulabh; Chen, Anyi; Conte, Joseph; Avramovic, Sanja; Sikdar, Siddhartha; Anastasopoulos, Antonios; Das, Sanmay (February 2024, Journal of the American Medical Informatics Association)

   Abstract ObjectiveThe use of electronic health records (EHRs) for clinical risk prediction is on the rise. However, in many practical settings, the limited availability of task-specific EHR data can restrict the application of standard machine learning pipelines. In this study, we investigate the potential of leveraging language models (LMs) as a means to incorporate supplementary domain knowledge for improving the performance of various EHR-based risk prediction tasks. MethodsWe propose two novel LM-based methods, namely “LLaMA2-EHR” and “Sent-e-Med.” Our focus is on utilizing the textual descriptions within structured EHRs to make risk predictions about future diagnoses. We conduct a comprehensive comparison with previous approaches across various data types and sizes. ResultsExperiments across 6 different methods and 3 separate risk prediction tasks reveal that employing LMs to represent structured EHRs, such as diagnostic histories, results in significant performance improvements when evaluated using standard metrics such as area under the receiver operating characteristic (ROC) curve and precision-recall (PR) curve. Additionally, they offer benefits such as few-shot learning, the ability to handle previously unseen medical concepts, and adaptability to various medical vocabularies. However, it is noteworthy that outcomes may exhibit sensitivity to a specific prompt. ConclusionLMs encompass extensive embedded knowledge, making them valuable for the analysis of EHRs in the context of risk prediction. Nevertheless, it is important to exercise caution in their application, as ongoing safety concerns related to LMs persist and require continuous consideration. 

   more » « less
3. Increasing value in the Veterans Affairs Healthcare System (VA) with precision health: a continuing landmark collaboration with the Department of Energy

   https://doi.org/10.1093/jamia/ocag062  

   Justice, Amy C; McMahon, Benjamin; Jacobson, Daniel A; Cho, Kelly; Kapadia, Anuj J; Aguayo, Samuel M; Gümüş, Zeynep H; Danciu, Ioana; Beckham, Jean C; Kimbrel, Nathan A; et al (April 2026, Journal of the American Medical Informatics Association)

   Abstract ObjectivePhase II of MVP-CHAMPION, a federal collaboration between the Veterans Affairs Healthcare System (VA) and the Department of Energy (DoE), leveraged large-scale clinical, geo-spatial, and genetic data with state-of-the-art artificial intelligence (AI), and high-performance computing (HPC) to improve value in healthcare. Materials and MethodsEight clinical priority projects for which AI was a critical missing capability were initiated to address: lung cancer screening (MVP 061), suicide risk screening (MVP 062), cardiovascular risk in obstructive sleep apnea (MVP 063), checkpoint inhibitor toxicity (MVP 064), heart failure (MVP 065), renal complications in diabetes (MVP 066), post COVID-19 sequelae (MVP 067), and antipsychotic medication toxicity (MVP 068). ResultsBuilding on a strong regulatory and administrative foundation, we developed multimorbidity-aware analytic frameworks, reusable computational tools, and analytic pipelines. These greatly facilitated identification of novel risk factors including genetic variants and specification of more discriminating prediction models. Novel genetic risk factors are informing development and repurposing of medications and discriminating prediction models promise to improve healthcare value. DiscussionThe research foundation developed in Phase I and extended in Phase II of MVP CHAMPION has supported an unprecedented federal collaboration and yielded significant scientific advances. Our clinical findings are poised for near-term application, while advances in machine learning and high-performance computing may accelerate the broader adoption of artificial intelligence in healthcare. ConclusionThis maturing VA-DoE federal collaboration is poised to transform the future of Veterans’ healthcare and the broader national landscape of precision health. 

   more » « less
4. Perceived Penalties for Sharing Patient Beliefs with Health Care Providers

   https://doi.org/10.1177/0272989X241262241  

   Marsh, Jessecae K; Asan, Onur; Kleinberg, Samantha (August 2024, Medical Decision Making)

   BackgroundHealth care interactions may require patients to share with a physician information they believe but is incorrect. While a key piece of physicians’ work is educating their patients, people’s concerns of being seen as uninformed or incompetent by physicians may lead them to think that sharing incorrect health beliefs comes with a penalty. We tested people’s perceptions of patients who share incorrect information and how these perceptions vary by the reasonableness of the belief and its centrality to the patient’s disease. DesignWe recruited 399 United States Prolific.co workers (357 retained after exclusions), 200 Prolific.co workers who reported having diabetes (139 after exclusions), and 244 primary care physicians (207 after exclusions). Participants read vignettes describing patients with type 2 diabetes sharing health beliefs that were central or peripheral to the management of diabetes. Beliefs included true and incorrect statements that were reasonable or unreasonable to believe. Participants rated how a doctor would perceive the patient, the patient’s ability to manage their disease, and the patient’s trust in doctors. ResultsParticipants rated patients who shared more unreasonable beliefs more negatively. There was an extra penalty for incorrect statements central to the patient’s diabetes management (sample 1). These results replicated for participants with type 2 diabetes (sample 2) and physician participants (sample 3). ConclusionsParticipants believed that patients who share incorrect information with their physicians will be penalized for their honesty. Physicians need to be educated on patients’ concerns so they can help patients disclose what may be most important for education. HighlightsUnderstanding how people think they will be perceived in a health care setting can help us understand what they may be wary to share with their physicians. People think that patients who share incorrect beliefs will be viewed negatively. Helping patients share incorrect beliefs can improve care. 

   more » « less
5. Evaluating the Trustworthiness of Explainable Artificial Intelligence (XAI) Methods Applied to Regression Predictions of Arctic Sea Ice Motion

   https://doi.org/10.1175/AIES-D-24-0027.1  

   Hoffman, Lauren; Mazloff, Matthew R; Gille, Sarah T; Giglio, Donata; Heimbach, Patrick (January 2025, Artificial Intelligence for the Earth Systems)

   Abstract Recent advances in explainable artificial intelligence (XAI) methods show promise for understanding predictions made by machine learning (ML) models. XAI explains how the input features are relevant or important for the model predictions. We train linear regression (LR) and convolutional neural network (CNN) models to make 1-day predictions of sea ice velocity in the Arctic from inputs of present-day wind velocity and previous-day ice velocity and concentration. We apply XAI methods to the CNN and compare explanations to variance explained by LR. We confirm the feasibility of using a novel XAI method [i.e., global layerwise relevance propagation (LRP)] to understand ML model predictions of sea ice motion by comparing it to established techniques. We investigate a suite of linear, perturbation-based, and propagation-based XAI methods in both local and global forms. Outputs from different explainability methods are generally consistent in showing that wind speed is the input feature with the highest contribution to ML predictions of ice motion, and we discuss inconsistencies in the spatial variability of the explanations. Additionally, we show that the CNN relies on both linear and nonlinear relationships between the inputs and uses nonlocal information to make predictions. LRP shows that wind speed over land is highly relevant for predicting ice motion offshore. This provides a framework to show how knowledge of environmental variables (i.e., wind) on land could be useful for predicting other properties (i.e., sea ice velocity) elsewhere. Significance StatementExplainable artificial intelligence (XAI) is useful for understanding predictions made by machine learning models. Our research establishes trustability in a novel implementation of an explainable AI method known as layerwise relevance propagation for Earth science applications. To do this, we provide a comparative evaluation of a suite of explainable AI methods applied to machine learning models that make 1-day predictions of Arctic sea ice velocity. We use explainable AI outputs to understand how the input features are used by the machine learning to predict ice motion. Additionally, we show that a convolutional neural network uses nonlinear and nonlocal information in making its predictions. We take advantage of the nonlocality to investigate the extent to which knowledge of wind on land is useful for predicting sea ice velocity elsewhere. 

   more » « less