Machine learning to predict the incidence of retinopathy of prematurity

Retinopathy of Prematurity (ROP) is a disorder afflicting prematurely born infants. ROP can be positively diagnosed a few weeks after birth. The goal of this study is to build an automatic tool for prediction of the incidence of ROP from standard clinical factors recorded at birth for premature babies. The data presents various challenges including mixing of categorical and numeric attributes and noisy data. In this article we present an ensemble classifierhierarchical committee of random trees-that uses risk factors recorded at birth in order to predict the risk of developing ROP. We empirically demonstrate that our classifier outperforms other state of the art classification approaches. Copyright © 2008, Association for the Advancement of Artificial Intelligence (www.aaai.org). All rights reserved.