According to an estimate, almost 5 million people with older age living with Alzheimer’s. And this number is expected to be around 13.5 million in 2050. Alzheimer’s disease is an irreversible, progressive brain disorder that slowly destroys memory and thinking skills. So, to help stem this tsunami of pending illness, a team from MIT in collaboration with Lahey Hospital and Medical Center created a new AI tool. According to scientists, this new AI tool could make it easier to detect cognitive impairment by improving current cognitive testing.
Since from 50 years, doctors are asking patients to treat with a cognitive testing tool: the Clock Drawing Test (CDT). Although, it efficiently helps doctors to differentiate between normal aging and possible dementia.
In this two-part test, doctors usually ask patients to draw a clock face in the first step. The clock must show a specified time on a blank piece of paper. Whereas in the second part, the patient needs to copy a clock face that shows the same time. Although, this deceptively simple cognitive test look at different aspects of cognition. The first half aspects are more important for memory and language functions. At the other hand, the second half aspects are important for spatial reasoning and executive functions.
Dana Penney, the director of neuropsychology at Lahey hospital said, “Comparing performance through the two-part cognitive testing enables us to use the person as their own control. This, in turn, affords an opportunity to detect subtle problems in cognitive function before there are overt errors. One potential outcome of this would be the possibility of detecting Alzheimer’s sooner than it would be ordinarily, perhaps offering an opportunity for pharmaceutical companies to develop medications that target the earlier stages of the disease.”
Scientists’ uses a digitizing pen that captures time-stamped pen coördinates. The pen also analyzes the drawing (clock face) and the process used by the subject to draw it. After researching for almost 10 years, scientists administered this novel version of the test to some 4,000 subjects. After that, they used machine learning techniques to create prediction models. The models precisely detect the impairment than the previously used methods.
Souillard-Mandar said, “There was a problem with classifiers produced by machine learning. The classifiers were literally difficult to understand. Thus, we came the point that this complex and opaque algorithm was unlikely to be accepted by either doctors or their patients. So, we turned to develop the analysis more transparent.”
“The analysis can be split into two phases: understanding what the individual drew. For instance that a set of pen points represents a clock’s minute hand, and, based on that understanding, deriving clinician-interpretable metrics that assess cognitive function together with an overall test outcome. It’s that second phase where we have worked to make our models interpretable, so a clinician can easily understand how the algorithm reached its decision.”
Davis said, “This new cognitive test that we have developed saves time, boosts accuracy. Unlike a previous cognitive test, it is also easy to use and inexpensive.”
Now scientists are taking an effort to bring their test in the medical world.