Technology | 01.06.23
Digital Detection of Dementia: Using AI to Identify Undiagnosed Dementia
New research details an artificial intelligence (AI) tool designed to identify undiagnosed Alzheimer's disease and related dementias in primary care. The passive digital marker is a machine learning algorithm that uses natural language processing to collect unstructured and structured data from a patient's electronic health record (EHR). This could include references to memory issues, vascular concerns, or other symptoms potentially linked to dementia. "In this new study, we are evaluating the practical use of our tool when used alone and when used with an accompanying patient-reported outcomes survey," said Malaz Boustani with the Regenstrief Institute and Indiana University School of Medicine. Currently underway is a clinical trial in Indianapolis that is enrolling patients seen in primary care clinics at federally qualified health centers. Participants are mainly Black people living in urban areas. A second clinical trial starts soon at University of Miami primary care clinics, with participants expected to be predominantly Hispanic and rural residents. Patients in each trial will be assigned to either receive typical primary care; analysis with a specially designed, AI-driven passive digital marker; or assessment by the passive digital marker in combination with the patient-reported outcomes survey. Subjects will be followed for two years to determine how many new cases of Alzheimer's are identified and documented in EHRs.