A team led by Carlos Cruchaga at Washington University developed an AI-based blood test to separate common causes of dementia from healthy brain aging. The classifier can identify when more than one disease process is present and often gives a clear diagnostic result.
To build the test, researchers selected 15 blood proteins that reflect brain pathology, including Alzheimer's markers and proteins linked to synapse and nerve damage and inflammation. The model was trained on blood protein data from more than 3,200 individuals and then verified on a separate group of 225 people who had brain examination after death. The outputs aligned with the pathological burden seen in tissue and with clinical presentation, and the test achieved an overall accuracy of 92.3% for single neurodegenerative diagnoses.
The model also helped in uncertain or evolving cases. For people with mild cognitive impairment or unclear diagnoses, predictions of Alzheimer's matched the amyloid plaque burden at autopsy. It identified Alzheimer-like changes in some people diagnosed with Parkinson's who later developed dementia. The test is not yet ready for clinical use; larger and prospective studies are needed to confirm results and assess clinical prediction and treatment guidance.
Difficult words
- classifier — a computer program that sorts data into categories
- pathology — study or presence of disease in tissue
- synapse — connection point between nerve cells
- inflammation — body response causing swelling and redness
- neurodegenerative — causing gradual loss of nerve cells
- accuracy — how often results are correct
- autopsy — medical examination of a body after death
- prospective — planned to collect data after study begins
Tip: hover, focus or tap highlighted words in the article to see quick definitions while you read or listen.
Discussion questions
- How could an AI blood test change diagnosis for people with mild cognitive impairment? Give one or two possible benefits.
- What concerns would you have about using an AI test like this to guide treatment decisions?
Related articles
2025 aid cuts threaten health and humanitarian services
Large reductions in international aid in 2025 disrupted health and humanitarian services in many low- and middle-income countries. The cuts began with a US suspension of aid and led to the closure of USAID and wider global impacts.
