Japanese pharmaceutical company Eisai, in collaboration with Japan's Oita University, has developed what could be the world's first AI model to use data from wearable devices to predict Alzheimer's disease, the most common form of dementia. It was constructed.
Their research team set out to create a cost-effective and practical tool to proactively screen people suspected of developing the disease. Their findings were published in Alzheimer's Research & Therapy.
Investigation result
The study collected biological and lifestyle data from 122 people aged 65 and older with mild cognitive impairment or subjective memory impairment. Biological data such as physical activity, sleep, and heart rate were collected from wristband sensors worn for seven days every three months from 2015 to 2019. The following lifestyle data were also collected from the medical consultation: employment status, frequency of outings, means of transportation, number of days participating in community activities, and background (age, education, drinking history, medical history).
In addition, participants underwent annual amyloid PET (positron emission tomography) tests, which detect the accumulation of amyloid beta protein in the brain, a key biomarker for Alzheimer's disease.
The researchers developed a predictive model that combines three machine learning technologies: support vector machines, elastic nets, and logistic regression. Integrate all the data you collect. The goal is to determine how likely each participant is to test positive on an amyloid PET screen.
Evaluation of the AI model showed that it had a “good enough” ability to predict amyloid beta protein accumulation. Through this model, the study was able to identify 22 factors that contribute to the accumulation of amyloid beta protein: physical activity, sleep, heart rate, amount of conversation, age, years of education, presence of children, lifestyle, transportation. Means, number of people accompanying you to the hospital, frequency of contact, number of outings.
Why is it important?
Advances in Alzheimer's disease research have led to the discovery of amyloid beta protein accumulation in the brain as an important biomarker for Alzheimer's disease. This further led to the development of therapeutic drugs targeting it.
Still, screening tests are essential for incurable diseases such as Alzheimer's disease. Testing is currently possible through amyloid PET and cerebrospinal fluid (CSF) tests, but these are expensive, limited and invasive, according to Eisai and Oita University researchers.
AI is being applied to find ways to make testing for Alzheimer's disease more accessible to more people, especially as countries like Japan are aging at a faster pace. Some research has already applied AI to predict amyloid beta protein buildup, but so far only cognitive function tests, blood tests, and brain imaging tests have been used. The latest research by a team from Eisai and Oita University brings value to this growing field of research as it focuses on lifestyle and biological data.
Following their pioneering work, the research team believes that the AI model could be broadly applied to pre-screening people with Alzheimer's disease, particularly those who have little access to amyloid PET or CSF testing.
bigger trends
The Japanese research team was not the first to consider using wearable devices to track or predict Alzheimer's disease biomarkers. In 2020, British Alzheimer's Research Microsoft has announced a project funded by Microsoft founder Bill Gates to develop a wearable diagnostic device for early detection of neurodegenerative diseases.
In recent years, an increasing number of projects are leveraging AI technology to improve the detection of Alzheimer's disease. Researchers from Australia's CSIRO and Queensland University of Technology recently published the results of a study. AI-based benchmark for measuring brain atrophyis touted to be a world first. fujifilm We also developed an AI tool that was found to be able to accurately predict which patients will progress to Alzheimer's disease within two years. lotte healthcare The South Korean company recently announced a partnership with iMediSync to explore developing new AI-driven healthcare services, including screening for neuropsychiatric disorders.
Meanwhile, Eisai continues its efforts in Alzheimer's disease research. Recently, we launched a new digital health business. Theoria, with a particular focus on dementia. Launching in April, Theoria will unveil its flagship solution, a risk prediction algorithm for early detection of mild cognitive impairment.