Machine Learning Uses EHR Data to Predict Alzheimer's Risk

Scientists have developed machine learning algorithms that use structured and unstructured electronic health record data to predict Alzheimer's disease risk in individuals, reports Health IT Analytics . The researchers used a natural-language processing algorithm and a random forest model, trained on data from the Indiana Network for Patient Care. The models examined information in structured fields, like prescriptions and diagnoses, as well as unstructured fields like medical notes. The notes were the most valuable resource for helping to identify people at high risk for dementia. "The great thing about this method is that it's passive, and it provides similar accuracy to the more intrusive tests that are currently used," said Indiana University School of Medicine Professor Malaz Boustani. "This is a low cost, scalable solution that can provide substantial benefit to persons and their families by helping them prepare for the possibility of life with dementia and enabling them to take action."