Identifying subtypes of type 2 diabetes mellitus with machine learning: development, internal validation, prognostic validation and medication burden in linked electronic health records in 420 448 individuals.

Mehrdad A MizaniAshkan DashtbanLaura PaseaQingjia ZengKamlesh KhuntiJonathan ValabhjiJil Billy MamzaHe GaoTamsin MorrisAmitava Banerjee

Published in: BMJ open diabetes research & care (2024)

In the largest study using ML to date in incident T2D, we identified four distinct subtypes, with potential future implications for etiology, therapeutics, and risk prediction.

Keyphrases

electronic health record
machine learning
adverse drug
clinical decision support
cardiovascular disease
small molecule
current status
risk factors
emergency department
big data