A meta study published this week in Nature.com examined the efficacy of diagnosing cognitive impairment such as Alzheimer’s Disease by applying natural language processing (NLP) technology to electronic health records (EHR). The review of 18 existing studies showed that using NLP to search patients’ EHRs showed promise in predicting certain kinds of cognitive decline sooner.
“The strong predictive performance across various modeling strategies highlights the feasibility of using NLP to surface subtle indicators of cognitive decline that may be missed in routine care, thus enabling earlier detection and intervention,” the report’s authors wrote.
According to the researchers, the 18 studies included in the review, representing over one million patients across diverse healthcare settings, demonstrate the potential of NLP to identify diagnostically relevant information from unstructured clinical text.
Despite the promise uncovered by the review, researchers identified several challenges and limitations across the studies including heterogeneity in reported metrics, limited details on error analysis, reliance on imperfect reference standards, such as diagnostic codes or brief cognitive screening tools, and the use of a single healthcare system in several of the studies.
While the researchers acknowledged these limitations, they contend the review is a systematic examination of the current state of the evidence, and the findings suggest that NLP could serve as a powerful tool for healthcare systems to improve the early detection and care of cognitive impairment.
