Key Takeaways:
- New AI model predicts dementia up to 9 years early with 82% accuracy.
- Early detection could allow for timely interventions, potentially preventing dementia progression.
- Researchers emphasize the need for further validation across diverse populations.
What Happened?
Researchers at Queen Mary University of London developed a new machine learning model that predicts dementia with over 80% accuracy up to nine years before diagnosis. Analyzing 1,111 fMRI scans, the model identifies changes in the brain’s default mode network (DMN), particularly when a person is at rest.
The study, published in Nature Mental Health, included 81 individuals who later developed dementia and 1,030 matched controls. The model accurately predicted dementia within a two-year window before diagnosis.
Why It Matters?
This breakthrough offers a significant advancement in early dementia detection. Early diagnosis could lead to timely interventions, potentially delaying or preventing the onset of dementia.
Dr. Claire Sexton of the Alzheimer’s Association emphasized that Alzheimer’s is linked to decreased functional connectivity within the DMN, aligning with these findings. Early detection is crucial as it provides a chance for interventions before significant cognitive decline occurs.
What’s Next?
Researchers aim to validate the model across more diverse populations and different types of dementia, including frontotemporal and Lewy body dementias. Current limitations include the study’s predominantly white, healthier-than-average cohort.
Neurologists like Clifford Segil stress the need for consistent reproducibility of fMRI results. While no neuroprotective medications are available yet, early detection models like this one could guide future clinical trials, ensuring the right candidates receive new treatments as they emerge.
