Augmented Intelligence to Identify Patients With Advanced Heart Failure in an Integrated Health System

Cheema, Baljash; Mutharasan, R. Kannan; Sharma, A.; Jacobs, Maia; Powers, Kaleigh; Lehrer, Susan; Wehbe, Firas; Ronald, Jason; Pifer, Lindsay; Rich, Jonathan D.; Ghafourian, Kambiz; Tibrewala, Anjan; McCarthy, Patrick M.; Luo, Yuan; Pham, Duc Thinh; Wilcox, Jane E.; Ahmad, Faraz

doi:10.1016/j.jacadv.2022.100123

Cited by 8 publications

(14 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A total of 7346 patients were identified within the health system to have either stage C or stage D heart failure (HF), and 7500 patients were chosen to be the control group as a result of their normal echocardiography. Using a deep learning algorithm, they were able to classify each case as belonging to one of three categories (stage C, D, healthy) with an overall accuracy of 83% compared to 75% accuracy of physician assessment [ 14 ].…”

Section: Resultsmentioning

confidence: 99%

Diagnostic AI and Cardiac Diseases

Ozsahin

Ozgocmen²,

Balcioglu³

et al. 2022

Diagnostics

View full text Add to dashboard Cite

(1) Background: The purpose of this study is to review and highlight recent advances in diagnostic uses of artificial intelligence (AI) for cardiac diseases, in order to emphasize expected benefits to both patients and healthcare specialists; (2) Methods: We focused on four key search terms (Cardiac Disease, diagnosis, artificial intelligence, machine learning) across three different databases (Pubmed, European Heart Journal, Science Direct) between 2017–2022 in order to reach relatively more recent developments in the field. Our review was structured in order to clearly differentiate publications according to the disease they aim to diagnose (coronary artery disease, electrophysiological and structural heart diseases); (3) Results: Each study had different levels of success, where declared sensitivity, specificity, precision, accuracy, area under curve and F1 scores were reported for every article reviewed; (4) Conclusions: the number and quality of AI-assisted cardiac disease diagnosis publications will continue to increase through each year. We believe AI-based diagnosis should only be viewed as an additional tool assisting doctors’ own judgement, where the end goal is to provide better quality of healthcare and to make getting medical help more affordable and more accessible, for everyone, everywhere.

show abstract

Section: Resultsmentioning

confidence: 99%

Diagnostic AI and Cardiac Diseases

Ozsahin

Ozgocmen²,

Balcioglu³

et al. 2022

Diagnostics

View full text Add to dashboard Cite

show abstract

“…Recent machine learning models have overcome many of the limitations of these traditional models. These include an augmented intelligence-enabled workflow for identifying outpatients with Stage D HF warranting clinical review to determine need for referral to a HF cardiologist [ 19 ] and an ensemble deep learning model trained to predict all-cause death, listing for HT, or extracorporeal membrane oxygenation (ECMO)/VAD within 1-year [ 20 ].…”

Section: Discussionmentioning

confidence: 99%

“…Second, the model requires prospective validation using the EHR with subsequent clinician review of model recommendations. Such an approach, when implemented elsewhere, led to an increase in clinical referrals to HF cardiologists as well as an increase in advanced therapies evaluations [ 19 ]. Third, the current algorithm only uses the previous visit to predict whether the patient will subsequently require advanced therapies.…”

Section: Discussionmentioning

confidence: 99%

Predicting need for heart failure advanced therapies using an interpretable tropical geometry-based fuzzy neural network

Zhang,

Aaronson,

Gryak

et al. 2023

PLoS ONE

View full text Add to dashboard Cite

Background Timely referral for advanced therapies (i.e., heart transplantation, left ventricular assist device) is critical for ensuring optimal outcomes for heart failure patients. Using electronic health records, our goal was to use data from a single hospitalization to develop an interpretable clinical decision-making system for predicting the need for advanced therapies at the subsequent hospitalization. Methods Michigan Medicine heart failure patients from 2013–2021 with a left ventricular ejection fraction ≤ 35% and at least two heart failure hospitalizations within one year were used to train an interpretable machine learning model constructed using fuzzy logic and tropical geometry. Clinical knowledge was used to initialize the model. The performance and robustness of the model were evaluated with the mean and standard deviation of the area under the receiver operating curve (AUC), the area under the precision-recall curve (AUPRC), and the F1 score of the ensemble. We inferred membership functions from the model for continuous clinical variables, extracted decision rules, and then evaluated their relative importance. Results The model was trained and validated using data from 557 heart failure hospitalizations from 300 patients, of whom 193 received advanced therapies. The mean (standard deviation) of AUC, AUPRC, and F1 scores of the proposed model initialized with clinical knowledge was 0.747 (0.080), 0.642 (0.080), and 0.569 (0.067), respectively, showing superior predictive performance or increased interpretability over other machine learning methods. The model learned critical risk factors predicting the need for advanced therapies in the subsequent hospitalization. Furthermore, our model displayed transparent rule sets composed of these critical concepts to justify the prediction. Conclusion These results demonstrate the ability to successfully predict the need for advanced heart failure therapies by generating transparent and accessible clinical rules although further research is needed to prospectively validate the risk factors identified by the model.

show abstract

“…In case study 3, an AI-powered tool was embedded within the EHR to develop regular reports of patients at risk of advanced heart failure (Cheema et al, 2022). Labeling of datasets comprised a delicate balance between clinical expertise and the availability of human labor.…”

Section: Data Labeling With Clinical Expertisementioning

confidence: 99%

“…In this perspective, we examine three case studies, outlined in Figure 1, two from an author's previous work and one based on a recently-deployed system. These studies identify the labor performed by clinical teams to create and use AI technology (Jacobs et al, 2021a,b;Cheema et al, 2022). Using these findings and related literature, we describe four examples of how labor is needed in the creation of medical-AI that requires clinical expertise (summarized in Figure 2).…”

Section: Introductionmentioning

confidence: 99%

Invisible clinical labor driving the successful integration of AI in healthcare

Ulloa

Rothrock

Ahmad

et al. 2022

Front. Comput. Sci.

Self Cite

View full text Add to dashboard Cite

Artificial Intelligence and Machine Learning (AI/ML) tools are changing the landscape of healthcare decision-making. Vast amounts of data can lead to efficient triage and diagnosis of patients with the assistance of ML methodologies. However, more research has focused on the technological challenges of developing AI, rather than the system integration. As a result, clinical teams' role in developing and deploying these tools has been overlooked. We look to three case studies from our research to describe the often invisible work that clinical teams do in driving the successful integration of clinical AI tools. Namely, clinical teams support data labeling, identifying algorithmic errors and accounting for workflow exceptions, translating algorithmic output to clinical next steps in care, and developing team awareness of how the tool is used once deployed. We call for detailed and extensive documentation strategies (of clinical labor, workflows, and team structures) to ensure this labor is valued and to promote sharing of sociotechnical implementation strategies.

show abstract

Augmented Intelligence to Identify Patients With Advanced Heart Failure in an Integrated Health System

Cited by 8 publications

References 28 publications

Diagnostic AI and Cardiac Diseases

Diagnostic AI and Cardiac Diseases

Predicting need for heart failure advanced therapies using an interpretable tropical geometry-based fuzzy neural network

Invisible clinical labor driving the successful integration of AI in healthcare

Contact Info

Product

Resources

About