Diagnosing dementia is often hard because Alzheimer's, Parkinson's and related conditions can look alike and often happen together. Researchers report an AI-based blood test that may help distinguish these diseases from normal cognitive aging.
The team selected 15 proteins in blood that reflect brain pathology. The panel includes known markers of Alzheimer's and proteins tied to synapse damage and inflammation. Researchers trained the classifier on data from more than 3,200 people and found the test showed over 90% accuracy for single diagnoses.
The tool was checked against brain autopsy results and gave similar findings. It is not yet ready for clinical use and needs larger, more diverse studies. Possible uses include finding patients for trials and guiding follow-up care.
Difficult words
- dementia — A condition with loss of memory and thinking.
- classifier — A computer program that sorts data into categories.
- pathology — The study or signs of disease in organs.
- synapse — The connection point between two nerve cells.
- inflammation — The body's reaction to injury or infection.
- autopsy — A medical exam after death to find cause.
Tip: hover, focus or tap highlighted words in the article to see quick definitions while you read or listen.
Discussion questions
- Would you take this blood test if you had memory problems? Why or why not?
- How could this blood test help researchers or doctors?
- What do researchers say the test still needs before clinical use?
Related articles
LLMs change judgments when told who wrote a text
Researchers at the University of Zurich found that large language models change their evaluations of identical texts when given an author identity. The study tested four models and warns about hidden biases and the need for governance.
