New AI-Powered Blood Test Could Distinguish Between Multiple Forms of Dementia

Researchers have developed an experimental blood test capable of identifying four major neurodegenerative diseases—Alzheimer’s, Parkinson’s, frontotemporal dementia, and dementia with Lewy bodies—with 92.3 percent accuracy. By analyzing 15 specific proteins, the diagnostic tool can detect the presence of multiple overlapping pathologies in a single patient, according to a study published in the May issue of Alzheimer’s & Dementia. This development represents a shift toward precision diagnostics, as current FDA-approved blood tests are limited to detecting only Alzheimer’s disease.

How the GPND-AI Test Works

The test, formally titled the “generalizable protein-based neurodegenerative disease artificial intelligence classifier” (GPND-AI), uses an AI algorithm to identify patterns among 15 specific blood-based proteins. Washington University in St. Louis researchers developed the tool using medical records and samples from over 3,000 patients. The AI initially screened 123 proteins before narrowing the selection to the 15 most predictive markers, including p-tau217, a protein already established as a key biomarker for Alzheimer’s disease. By quantifying the ratio of these proteins, the test can estimate the biological contribution of different diseases within one individual, such as identifying a mix of Alzheimer’s and Parkinson’s pathology.

Why Distinguishing Between Dementias Matters

Dementia is not a singular condition, and clinical symptoms often overlap, making accurate diagnosis difficult in the early stages. Carlos Cruchaga, a human genomicist at Washington University in St. Louis, notes that these diseases are more biologically complex than previously understood. Different dementias require distinct clinical management strategies; therefore, identifying the specific combination of pathologies is essential for precision medicine. While current FDA-approved tests provide a binary “yes/no” for Alzheimer’s, the GPND-AI model aims to capture the biological reality of patients who may suffer from co-occurring neurodegenerative processes.

Limitations and Future Clinical Application

Despite the high accuracy reported in the study, medical experts emphasize that blood biomarkers are not a standalone diagnostic. Davide Cappon, a neuropsychologist at Tufts Medical Center, notes that blood tests cannot fully account for a patient’s daily functioning, the root cause of symptom onset, or external factors like vascular disease, depression, or sleep quality. Furthermore, while the team validated their model against brain tissue donated at autopsy at the Banner Sun Health Research Institute, larger and more diverse clinical trials are required before the test can be used in routine practice. The research team is currently engaging with pharmaceutical companies to initiate formal clinical trials and pursue FDA regulatory review.

Key Facts About Neurodegenerative Diagnostics

• Scope: The GPND-AI test targets four conditions: Alzheimer’s, Parkinson’s, frontotemporal dementia, and dementia with Lewy bodies.
• Accuracy: The study reports 92.3 percent accuracy in distinguishing between these diseases and identifying mixed-pathology cases.
• Methodology: The test relies on a 15-protein panel analyzed via an AI algorithm, moving away from single-disease diagnostic frameworks.
• Status: The test is currently experimental; researchers are preparing for future clinical trials to seek potential FDA approval.