Global EditionASIA 中文双语Français
Home / China / Innovation

Scientists find way to predict dementia

By Zhou Wenting in Shanghai | China Daily | Updated: 2024-02-19 09:13
Share - WeChat

Research identifies blood biomarkers to detect disorder 15 years in advance

Using a massive databank and artificial intelligence, Shanghai scientists have discovered biomarkers in plasma that can predict dementia 15 years before symptoms begin.

The scientists said that their research results may play a major role in early intervention for healthy adults who are at high risk of developing the disorder.

A paper about the research conducted by teams from Fudan University's Institute of Science and Technology for Brain-inspired Intelligence and the university's Huashan Hospital was published in Nature Aging on Feb 13.

An editorial published in Nature on the same day said the researchers' work was "a step towards [the development of] a tool that scientists have been in search of for decades: blood tests that can detect Alzheimer's disease and other forms of dementia at a very early, presymptomatic stage".

It is often difficult to diagnose brain disorders, doctors said. Lumbar punctures are invasive, and examinations of brain images are expensive. In contrast, blood tests are convenient, non-invasive and cost-effective, said Cheng Wei, a co-author of the paper and a researcher with the institute.

Fudan University scientists said they are hopeful that blood testing to predict the likelihood of dementia can be applied in clinical settings within six months. The early detection of the disorder opens a door to early intervention, offering the potential to slow down or even halt its progression.

According to the World Health Organization, dementia affects over 55 million people worldwide, and that figure is expected to continue to rise. Dementia progresses slowly, from an asymptomatic stage to a fully expressed clinical syndrome, over the course of a decade or more.

"By the time patients begin showing cognitive behavioral problems, the disorder may have already progressed to the middle or late stages, and the best intervention time will have been missed," said Feng Jianfeng, a computational biologist at the Fudan University institute and another co-author of the paper.

The researchers employed the help of the massive United Kingdom Biobank cohort, which enrolled more than 52,600 healthy adults and had a median follow-up period of 14 years. Among them, 1,417 people were diagnosed with all-cause dementia, 691 with Alzheimer's disease and 285 with vascular dementia.

For each participant, 1,463 proteins in plasma associated with cardio metabolism, inflammation, neurology and oncology were tested, and researchers used survival association analysis and machine learning algorithms to perform modeling analysis.

They discovered significant associations of three proteins — GFAP, NEFL, and GDF15 — with the risk of those three types of dementia. They also found that the protein LTBP2 plays a role in the onset of the disorder.

These biomarkers, as well as conventional risk factors of age, gender, education level and genetics, were used to facilitate the high accuracy of the predictive model, exceeding 90 percent.

"Our study provides a great example of how AI can facilitate a research paradigm that fosters interdisciplinary collaboration," Feng said. "Employing machine learning, we extracted and optimized the combinations using a large-scale dataset and established a protein-based dementia prediction model with high accuracy."

The team will now focus on conducting data collection and cross-validation among populations at risk of dementia in China. It will tailor the dementia risk prediction model to fit the characteristics of the Chinese population by gathering relevant data.

Copyright 1995 - . All rights reserved. The content (including but not limited to text, photo, multimedia information, etc) published in this site belongs to China Daily Information Co (CDIC). Without written authorization from CDIC, such content shall not be republished or used in any form. Note: Browsers with 1024*768 or higher resolution are suggested for this site.
License for publishing multimedia online 0108263

Registration Number: 130349