AI tool identifies nine dementia types from a single scan with 88% accuracy
Mayo Clinic’s StateViewer doubles diagnostic speed, triples clinician accuracy
Breakthrough supports earlier, more precise dementia care in everyday clinics
The earlier someone is diagnosed with dementia, the sooner treatment can begin. So it is a hopeful development that a team of researchers at the Mayo Clinic has developed a new artificial intelligence tool that could revolutionize the way dementia is diagnosed, offering a major leap forward in speed and accuracy.
The tool is called StateViewer. It’s an AI-based system capable of identifying nine different forms of dementia from a single brain scan, achieving an impressive 88% diagnostic accuracy. The findings, recently published in Neurology, the journal of the American Academy of Neurology, highlight the potential of this tool to change clinical practice and improve patient outcomes.
Revolutionizing dementia detection
StateViewer analyzes fluorodeoxyglucose positron emission tomography (FDG-PET) scans, which measure how the brain uses glucose, a vital marker for brain activity. The AI compares each scan against a massive database of confirmed dementia cases, allowing it to differentiate among various conditions such as Alzheimer’s disease, Lewy body dementia, and frontotemporal dementia.
In clinical trials, physicians using StateViewer completed assessments in half the time compared to conventional methods and with up to three times the accuracy.
The researchers said StateViewer enables clinicians to diagnose complex cases more quickly and with greater confidence.
While traditionally, accurate dementia diagnosis has required referral to major medical centers with specialized teams, StateViewer is designed for broad accessibility. Its user-friendly interface translates brain patterns into color-coded maps, simplifying interpretation even for non-specialists.
Making new treatments more effective
This democratization of diagnostic expertise could be especially impactful as new dementia treatments become available — treatments that are most effective when administered early. According to the Mayo team, delayed diagnoses often result from the complexity of testing and limited access to neurological expertise. StateViewer aims to remove these barriers.
A significant strength of the system is its ability to handle overlapping or atypical dementia presentations, a frequent challenge in real-world clinical settings. By converting subtle brain activity changes into intuitive visuals, researchers said StateViewer effectively reduces diagnostic ambiguity.
The researchers stess that their work is ongoing. Plans are underway to expand clinical trials and evaluate the tool’s performance in a wider variety of healthcare environments, including community hospitals and international clinics.