Is it possible to predict who will develop Alzheimer’s disease simply by looking at the writing patterns years before there are symptoms?
According to a new study by researchers at IBM, the answer is yes.
And, they and others say that Alzheimer’s is just the beginning. People with a wide variety of neurological diseases have distinct language patterns that, researchers suspect, may serve as early warning signs of their illnesses.
For the Alzheimer’s study, the researchers looked at a group of 80 men and women in their 80s – half had Alzheimer’s and the other did not. But seven and a half years earlier, everything was cognitively normal.
The men and women participated in the Framingham Heart Study, a long-term federal research effort that requires regular physical and cognitive testing. As part of that, they took a writing test before any of them developed Alzheimer’s, which asks participants to describe the drawing of a boy standing on an unstable stool and picking up a jar of cookies on a high shelf while a woman his back to him, he is oblivious to a sink overflowing.
The researchers examined the subjects’ use of words with an artificial intelligence program that looked for subtle differences in language. He identified a group of subjects who were more repetitive in the use of words at that time, when all were cognitively normal. These subjects also made mistakes, such as spelling wrong words or capitalizing them inappropriately, and used telegraphic language, that is, a language that has a simple grammatical structure and is missing subjects and words like “o”, “is” and ” is”.
The members of that group turned out to be the people who developed Alzheimer’s disease.
The AI program predicted, with 75 percent accuracy, who would have Alzheimer’s disease, according to results recently published in The Lancet EClinicalMedicine.
“We had no prior assumption that using words would show anything,” said Ajay Royyuru, vice president of health and life science research at IBM Thomas J. Watson Research Center in Yorktown Heights, NY, where AI analysis was made.
Alzheimer’s researchers were intrigued, saying that when there are ways to slow or stop the disease – a goal that has so far remained elusive – it will be important to have simple tests that can alert, right from the start, that without intervention a person will develop progressive disease. cerebral.
“What’s going on here is very smart,” said Dr. Jason Karlawish, an Alzheimer’s researcher at the University of Pennsylvania. “Given a high volume of spoken or written speech, can you detect a signal?”
For years, researchers have analyzed changes in speech and voice in people who show symptoms of neurological diseases – Alzheimer’s, ALS, Parkinson’s, frontotemporal dementia, bipolar disorder and schizophrenia, among others.
But, said Dr. Michael Weiner, who researches Alzheimer’s disease at the University of California, San Francisco, the IBM report breaks new ground.
“This is the first report I saw that it took people who are completely normal and predicted with some precision that they would have problems years later,” he said.
The hope is to extend Alzheimer’s work to find subtle changes in the use of language by people without obvious symptoms, but who will develop other neurological diseases.
Each neurological disease produces unique changes in speech, which are likely to occur long before the time of diagnosis, said Dr. Murray Grossman, professor of neurology at the University of Pennsylvania and director of the university’s frontotemporal dementia center.
He has studied speech in patients with a behavioral form of frontotemporal dementia, a disorder caused by the progressive loss of nerves in the frontal lobes of the brain. These patients exhibit apathy and declines in judgment, self-control and empathy that have proved difficult to objectively quantify.
The speech is different, Grossman said, because changes can be measured.
At the beginning of the disease course, changes in the speech rate of patients occur, with pauses distributed apparently at random. Word usage also changes – patients use less abstract words.
These changes are directly linked to changes in the frontotemporal parts of the brain, said Dr. Grossman. And they seem to be universal, not exclusive to English.
Dr. Adam Boxer, director of the neuroscience clinical research unit at the University of California, San Francisco, is also studying frontotemporal dementia. Your tool is a smartphone application. His subjects are healthy people who have inherited a genetic predisposition to develop the disease. Their method is to show participants an image and ask them to record a description of what they are seeing.
“We want to measure early changes, five to 10 years before they show symptoms,” he said.
“The good thing about smartphones,” added Dr. Boxer, “is that you can do all kinds of things.” Researchers can ask people to talk for a minute about something that happened that day, he said, or to repeat sounds like tatatatata.
Dr. Boxer said that he and others were focusing on speech because they wanted inexpensive, non-invasive tests.
Dr. Cheryl Corcoran, a psychiatrist at the Icahn School of Medicine at Mount Sinai in New York, hopes to use speech changes to predict which adolescents and young adults at high risk for schizophrenia may develop the disease.
Medicines to treat schizophrenia can help those who will develop the disease, but the challenge is to identify who the patients will be. A quarter of people with occasional symptoms saw them go and about a third never progressed to schizophrenia, although their occasional symptoms persisted.
Guillermo Cecchi, an IBM researcher who was also involved in the recent Alzheimer’s research, studied the speech in 34 of Dr. Corcoran’s patients, looking for “flight of ideas”, that is, the instances in which the patients were out of the way when speaking and developing ideas in different instructions ways. He also looked for “speech poverty”, meaning the use of simple syntactic structures and short sentences.
In addition, Dr. Cecchi and his colleagues studied another small group consisting of 96 patients in Los Angeles – 59 of whom had occasional delusions. The rest were healthy people and people with schizophrenia. He asked these guys to tell a story they just heard and looked for the same revealing speech patterns.
In both groups, the artificial intelligence program could predict, with 85 percent accuracy, which individuals developed schizophrenia three years later.
“Many small studies have found the same signs,” said Corcoran. At that point, she said, “we have not yet reached the point where we can tell people whether they are at risk or not.”
Dr. Cecchi is encouraged, although he knows that studies are still in their infancy.
“For us, it is a priority to do science correctly and at scale,” he said. “We should have a lot more samples. There are more than 60 million psychiatric interviews in the United States each year, but none of these interviews are using the tools we have. “