University developed AI tool could help detect alzheimer’s and dementia
LONDON: There is yet another computer screening model supported by artificial intelligence (AI) that could better aid in detecting and diagnosing Alzheimer’s disease and other forms of dementia in the future.
That’s according to recent research by the University of Nebraska Medical Center (UNMC) which included a diagnostic computer model based on over 51,000 medical records from exams, test results, and magnetic resonance imaging scans.
The AI tool is among the latest designed to detect Alzheimer’s and other dementias earlier. Others include an AI-supported tool used in detecting dementia in Spanish-speaking older adults and detecting cognitive impairment at-large.
The main goal of the UNMC project was to develop a diagnostic tool that could help healthcare workers improve referrals to clinicians for those with dementia. The model was able to detect dementia in patients 26.25% more frequently than “neurologist-only evaluations” in a comparison of 100 randomly selected cases.
“It was exciting to be able to contribute to the training of this phenomenal tool by sharing our clinical wisdom,” said study author Olga Taraschenko, MD, PhD, in a news release.
The latest research in AI-supported dementia detection and diagnosis follows an earlier UNMC project that created a machine learning framework to distinguish between normal brain cognition and dementia.
Senior living providers in recent years have implemented various technology and AI-supported systems to improve resident care, specifically in higher-acute memory care settings.
The project’s latest AI algorithm is now able to accurately predict a dementia diagnosis even if certain aspects of supporting medical record information are missing by using information from other complete case records, Taraschenko said.
The current model could be integrated into electronic medical records in the future to assist physicians and medical providers in real time. However, more work remains to be done, with Taraschenko calling the datasets “imperfect” for use in “everyday clinical practice.”
By using test and clinical information, physicians could use the datasets to diagnose specific dementia types or make referrals to specialists.
While AI-supported detection models continue to advance, early detection in dementia diagnoses remains a key area of research. In April, the Food and Drug Administration (FDA) granted a “Breakthrough Device Designation” to Eli Lilly and Company for identifying the likelihood of amyloid pathology, a key protein found in plaque buildup in the brains of those with Alzheimer’s disease.
