New research appearing in the peer-reviewed Journal of the American Geriatrics Society used electronic medical records (EMR) data to test the efficacy of machine learning algorithms in identifying patients at risk of developing neurodegenerative diseases, like Alzheimer’s.
According to researchers from Regenstrief Institute, Merck and Indiana University, screening tests for establishing dementia risk could be more efficient and cost-effective by using machine learning algorithms. Two methods were analyzed by the team: one of a natural language processing algorithm and the other from a random forest model.
From the two algorithms, certain data of patients such as diagnoses, prescriptions, and medical notes, were racked and utilized to predict the onset of dementia.
As of early-2020, many patients living with Alzheimer’s disease exhibit symptoms of cognitive decline for two to five years before receiving a diagnosis. Such algorithms as detailed in the study may expedite the early risk identification of dementia with precise accuracy as current procedures.
“We used data from the INPC that includes structured and unstructured (visit notes, progress notes, medication notes) EMR data,” the findings state. “Cases and controls were matched on age, race, and sex. The derivation sample consisted of 10 504 cases and 39 510 controls; the validation sample included 4500 cases and 16 952 controls.”
Based on the findings, researchers concluded the following: “EMR‐based data provide a targeted and scalable process for early identification of risk of ADRD as an alternative to traditional population screening.”
“This is a low cost, scalable solution that can provide substantial benefit to patients and their families by helping them prepare for the possibility of life with dementia and enabling them to take action,” Malaz Boustani, the study’s lead researcher, stated.
“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.”