Identifying infected patients using semi-supervised and transfer learning.

Fereshteh S BashiriJohn R CaskeyAnoop MayampurathNicole DussaultJay DumanianSivasubramanium V BhavaniKyle A CareyEmily R GilbertChristopher J WinslowNirav S ShahDana P EdelsonMajid AfsharMatthew M Churpek

Published in: Journal of the American Medical Informatics Association : JAMIA (2022)

In a multicenter study of almost 3000 chart-reviewed patients, semi-supervised and transfer learning models showed similar performance for model discrimination as baseline XGBoost, while transfer learning improved calibration.

Keyphrases

end stage renal disease
machine learning
ejection fraction
newly diagnosed
chronic kidney disease
peritoneal dialysis
prognostic factors
electron transfer
low cost
patient reported