IBM Study Uses Writing Tests to Predict Onset of Neurological Diseases



Researchers at IBM are using writing tests to analyze language and speech patterns in order to predict Alzheimer’s diagnoses, the New York Times reported in an article published February 1.

Using an artificial intelligence (A.I.) program, IBM researchers conducted a study with 80 participants — half with Alzheimer’s dementia and the remaining participants without Alzheimer’s. Participants took a writing test before any of them had developed Alzheimer’s. According to the article, the A.I. program predicted, with 75 percent accuracy, who would develop Alzheimer’s over time.

The IBM study shows signs of how this work can be extended to find subtle changes in language use by people with no obvious symptoms but who will later go on to develop other neurological diseases, such as FTD.

Dr. Murray Grossman, a professor of neurology at the University of Pennsylvania and the director of the Penn FTD Center, said in the article that “each neurological disease produces unique changes in speech, which probably occur long before the time of diagnosis.” Grossman, who is also a member of AFTD’s Medical Advisory Council, studies the speech of people diagnosed with behavioral variant FTD (bvFTD). He explained how it is difficult to “objectively quantify” a person with bvFTD, but speech is different because the changes can be measured.

Dr. Adam Boxer, director of the neurosciences clinical research unit at the University of California, San Francisco, also studies FTD and was featured in the New York Times article. Boxer uses a smartphone app to measure the possible symptoms of people who have inherited a genetic predisposition to develop the disease. Participants are shown a picture and use the app to record a description of what they see.

“We want to measure very early changes, five to 10 years before they have symptoms,” he said.

Boxer also mentioned that he and others were focusing on speech analysis tests because “they wanted tests that were noninvasive and inexpensive.”

Boxer was a former recipient of AFTD’s FTD Biomarkers Initiative award in 2018. Both Boxer and Grossman participated in the first-ever World FTD Marathon in October, a day-long series of interactive webinars from around the world focused on FTD-related content in conjunction with World FTD Awareness Week.

Read the full New York Times article here.

Stay Informed

Sign up now and stay on top of the latest with our newsletter, event alerts, and more…