Early Prediction of Preeclampsia via Machine Learning

An early prediction model for preeclampsia was developed with the use of machine learning methods that analyzed all available clinical and laboratory data during routine prenatal visits. Using the elastic net algorithm, a subset of the most informative features from all variables can be automatically identified. The obtained prediction model for preeclampsia yielded an area under the curve of 0.79 (95% confidence interval, 0.75–0.83), sensitivity of 45.2%, and false-positive rate of 8.1%. The prediction model for early-onset preeclampsia achieved an area under the curve of 0.89 (95% confidence interval, 0.84–0.95), true-positive rate of 72.3%, and false-positive rate of 8.8%.