Assessment of a Machine Learning Model Applied to Harmonized Electronic Health Record Data for the Prediction of Incident Atrial Fibrillation

Abstract

IMPORTANCE Atrial fibrillation (AF) is the most common sustained cardiac arrhythmia, and its early detection could lead to significant improvements in outcomes through the appropriate prescription of anticoagulation medication. Although a variety of methods exist for screening for AF, a targeted approach, which requires an efficient method for identifying patients at risk, would be preferred. OBJECTIVE To examine machine learning approaches applied to electronic health record data that have been harmonized to the Observational Medical Outcomes Partnership Common Data Model for identifying risk of AF. DESIGN, SETTING, AND PARTICIPANTS This diagnostic study used data from 2 252 219 individuals cared for in the UCHealth hospital system, which comprises 3 large hospitals in Colorado, from January 1, 2011, to October 1, 2018. Initial analysis was performed in December 2018; follow-up analysis was performed in July 2019. EXPOSURES All Observational Medical Outcomes Partnership Common Data Model-harmonized electronic health record features, including diagnoses, procedures, medications, age, and sex. MAIN OUTCOMES AND MEASURES Classification of incident AF in designated 6-month intervals, adjudicated retrospectively, based on area under the receiver operating characteristic curve and F1 statistic. RESULTS Of 2 252 219 individuals (1 225 533 [54.4%] women; mean [SD] age, 42.9 [22.3] years), 28 036 (1.2%) developed incident AF during a designated 6-month interval. The machine learning model that used the 200 most common electronic health record features, including age and sex, and random oversampling with a single-layer, fully connected neural network provided the optimal prediction of 6-month incident AF, with an area under the receiver operating characteristic curve of 0.800 and an F1 score of 0.110. This model performed only slightly better than a more basic logistic regression model composed of known clinical risk factors for AF, which had an area under the receiver operating characteristic curve of 0.794 and an F1 score of 0.079. CONCLUSIONS AND RELEVANCE Machine learning approaches to electronic health record data offer a promising method for improving risk prediction for incident AF, but more work is needed to show improvement beyond standard risk factors.

Application of Deep Learning Algorithm in Feature Mining and Rapid Identification of Colorectal Image

“…The process of the evolution of the distance regularization level set curve is the process of continuously reducing ε in equation 6and eventually approaching zero [52]. This process is expressed by differential equations, and the expression is shown in 7:…”

MINIMAR (MINimum Information for Medical AI Reporting): Developing reporting standards for artificial intelligence in health care

“…8 , 9 It is clear that the performance of any AI model broadly depends on its reliability and its ability to generalize to the setting and population in which it is applied, rather than its performance represented by the training and test data alone. 10 However, the characteristics of the data necessary to assess how these predictive models perform are not being adequately reported in the literature, 11 leaving uncertainty and doubt about the application in the broader healthcare setting. An empirical evaluation of 81 studies comparing AI models against clinicians showed major problems with lack of transparency, bias, and unjustified claims, likely because key details about the studies were often missing.…”

“…Indeed, our recent study showed an alarming lack of transparency of ML models developed in research studies. 11 Therefore, we need transparency in the reporting of the design, development, evaluation, and validation of AI models in health care to achieve and retain confidence and trust for all the stakeholders.…”

Development, validation, and proof-of-concept implementation of a two-year risk prediction model for undiagnosed atrial fibrillation using common electronic health data (UNAFIED)

“…However, these existing models required variables not typically available as structured electronic health record (EHR) fields (e.g., electrocardiogram [ECG] parameters). Newer attempts at EHR-based models have required extra steps of data harmonization [ 9 ] or use of long prediction horizons and require non-missing data or oft-unstructured data [ 10 ]. Other efforts use available data from the EHR, but make a prediction for 10-year risk using binned risk categories [ 11 ], or rely on claims databases and make predictions on simple cross-sectional association [ 12 ].…”