Objective: Machine learning (ML) techniques have shown promise for enhancing prediction of clinical outcomes; however, its application to predicting binge eating has been scarcely explored. We applied ML techniques to predict binge eating onset (vs. continued absence) and persistence (vs. remission) over time.

Method: Data were used from a larger prospective study of 1106 participants who were assessed on a range of putative risk, maintaining, and protective factors at baseline and 8 months follow-up. Nine ML models for classification were developed and compared against a generalised linear model (GLM) for predicting onset (n = 334) and persistence (n = 623) outcomes using 39 self-reported baseline variables as predictors.

Results: All models performed poorly at predicting onset (AUC = 0.49-0.61) and persistence (AUC = 0.50-0.59) outcomes, with ML models demonstrating comparable performance to the GLM.

Conclusion: We suspect that poor ML performance may have been a result of the limited set of self-reported baseline predictors used to generate prediction models. Improved predictive accuracy and optimisation of ML models in future research may require consideration of a larger, more disparate set of predictors that also incorporate various data types, such as neuroimaging, physiological, or smartphone sensor data.

Download full-text PDF

Source
http://dx.doi.org/10.1002/erv.3154DOI Listing

Publication Analysis

Top Keywords

binge eating
12
machine learning
8
learning techniques
8
prospective study
8
predicting onset
8
self-reported baseline
8
models
5
application machine
4
techniques prediction
4
onset
4

Similar Publications

Want AI Summaries of new PubMed Abstracts delivered to your In-box?

Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!