2 min

Tags in this article

, , ,

Big Blue has teamed up with Pfizer to create a program that uses small language samples to make the prediction.

IBM announced last week that its researchers, working together with a team from Pfizer, had developed an AI model that could correctly predict Alzheimer’s in patients with 71 percent accuracy, using machine-learning algorithms analyzing human communication.

Using a short, non-invasive language sample from a standard verbal cognitive test given to a patient, the new AI model produces valuable predictive feedback significantly better than clinical-scale predictions (59 percent) based on other available biomedical data.

The research breakthrough was published on October 22 in The Lancet eClinicalMedicine. According to IBM, this achievement advances the quest to predict Alzheimer’s in several major ways.

A new way to predict Alzheimer’s

The study is one of the first to use AI to predict outcomes in healthy individuals. This differentiates it from the vast majority of other Alzheimer’s prediction research. Most such research has focused on individuals already starting to show signs of cognitive decline. Such research also studies those with risk factors such as family history or genetics.

In addition, IBM’s work also focuses on using AI to assess the risk of Alzheimer’s disease in the general population. This is in contrast to research that focuses solely on high-risk groups or those with a genetic history or predisposition to the disease.

One of IBM’s goals is to be able to train the model using expanded datasets. That would include data that might become newly available to the healthcare research community.

Such data would reflect wider geographical, socioeconomical and racial diversity. As a result, IBM researchers’ algorithms could serve as a potential clinical asset for health professionals. It could help them gauge a holistic view of an individual’s health and risk factors.

The study is part of a larger effort by IBM Research

IBM says that this Alzheimers study is part of a larger platform the company is building. The goal of this new platform to better understand neurological health and chronic illnesses. It achieves this through biomarkers and signals in speech and language.

IBM Research recently announced a related initiative with the National Institutes of Health. The project analyzes speech to to uncover risk factors for schizophrenia and psychosis

The Alzheimer’s research will be another part of this platform, which is now moving into prospective studies.