The research, “Episodic-Memory Performance in Machine Learning Modeling for Predicting Cognitive Health Status Classification,” was published in the Journal of Alzheimer’s Disease.
Despite their increasing availability, memory assessment tools still have limited accuracy and reliability. MemTrax, which analyzes the ability to recognize images, is presented as a quick and simple way to track memory, but results regarding its efficacy are not conclusive.
A team from Florida Atlantic University (FAU), MemTrax, SIVOTEC Analytics, the brain training program HAPPYneuron, and Stanford Medicine hypothesized that artificial intelligence could provide a solution to manage Alzheimer’s. Their work used supervised machine learning — a subset of artificial intelligence — and predictive statistical modeling to evaluate the utility of MemTrax to assess cognitive impairment.
The scientists used a dataset including questions on memory, sleep quality, medications, and other medical conditions affecting thinking from a group of 18,395 adults who took the MemTrax online Continuous Recognition Tasks (M-CRT) test as part of the HAPPYneuron program. Of note, only 4,645 participants answered all four general health questions.
The test included five sets of five images of common scenes or objects, 25 of which were unique and 25 were repeats. Each participant was instructed to press the space bar of the computer to start viewing the images and to press it as quickly as possible whenever a repeated picture appeared.
The response time (up to three seconds) and the percentage of correct responses were then measured. Factors such as age, sex and recent alcohol consumption were taken into account.
The four health-related screening questions addressed memory problems, difficulty sleeping, medications taken and medical conditions that might affect thinking.
A general health status, called HealthQScore, was created based on the assumption that a greater number of “Yes” answers would mean a higher likelihood of having Alzheimer’s or some other form of cognitive impairment.
The findings supported the clinical utility of MemTrax to assess cognition. Its comparison to the well-established Montreal Cognitive Assessment Estimation of mild cognitive impairment further revealed MemTrax’s potential to assess short-term memory in a variety of disorders, including dementia.
“Our novel application of supervised machine learning and predictive modeling helps to demonstrate and validate cross-sectional utility of MemTrax in assessing early-stage cognitive impairment and general screening for [Alzheimer’s],” the researchers said.
“Seamless use and real-time interpretation will enhance case management and patient care through innovative technology and practical and readily usable integrated clinical applications that could be developed into a hand-held device and app,” Taghi Khoshgoftaar, PhD, a study co-author and a professor at FAU, said in a press release.
Michael F. Bergeron, PhD, the study’s senior author and senior vice president of development and applications at SIVOTEC, said: “Findings from our study provide an important step in advancing the approach for clinically managing a very complex condition like Alzheimer’s.”
“By analyzing a wide array of attributes across multiple domains … supervised machine learning and robust analytics can be integral, and in fact necessary, for health care providers to detect and anticipate further progression in this disease and myriad other aspects of cognitive impairment,” he said.
Stella Batalama, PhD, dean of FAU’s College of Engineering and Computer Science, noted the need for optimal tools to assess diseases affecting cognition, such as Alzheimer’s. “Results from this important study provide new insights and discovery that has set the stage for future impactful and significant research,” Batalama said.
We are sorry that this post was not useful for you!
Let us improve this post!
Tell us how we can improve this post?