Researchers at the Mayo Clinic have developed a new Artificial Intelligence (AI) tool that helps physicians identify nine types of dementia -attached brain activity patterns, including Alzheimer’s disease, using a single, widely available scan – a transformational advance in the initial, accurate diagnosis.

According to research published online on 27 June 2025, tools, stateviews, helped researchers identify dementia in 88% of cases. Nervous scienceMedical Journal of American Academy of Neurology. This enabled physicians to explain brain scans and with almost twice and three times more accuracy than standard workflows. Researchers trained and tested AI on more than 3,600 scans, including people without images and cognitive losses of dementia.

This innovation addresses a main challenge in dementia, identifying the disease quickly and properly, even when many conditions exist. As new remedies emerge, timely diagnostic diagnosis helps patients matching with the most appropriate care when it can have the greatest effect. This device can bring advanced clinical support to clinics with a lack of neurology expertise.

Growing toll of dementia

Dementia affects over 55 million people worldwide, about 10 million new cases every year. Alzheimer’s disease, the most common form, is now the fifth-virtuous cause of death globally. Diagnosis of dementia usually requires cognitive testing, blood draw, imaging, clinical interviews and specialist referrals. Even with comprehensive testing, separate conditions such as Alzheimer’s, Levy Body Dementia and Frontotemporal dementia are challenging, including highly experienced experts.

StateVur was developed under the direction of David Jones, MD, a Mayo Clinic Neurologist and Director of Mayo Clinic Neurology Artificial Intelligence Program.

Every patient who walks in my clinic carries a unique story of the shape of the complexity of the brain. That complexity attracted me to neurology and continued its commitment to clean the answers. StateViewer reflects that commitment – earlier understanding, more accurate treatment and, one day, changing the course of these diseases. ,

David Jones, MD, Mayo Clinic Neurologist

To bring that vision to life, Dr. Jones worked with Leyland Barnard, PhD, a data scientist, who leads AI engineering behind the stateview.

“As we were designing the state, we never lost the fact that each data point and brain scan had a difficult diagnosis and immediate question,” Say Barnard. “Seeing how this device can help doctors with real -time, accurate insights and guidance, highlights the ability to learning machine for clinical medicine.”

Conversion

The device analyzes a fluorodyglucose positron emission tomography (FDG-PET) scan, which shows how brain uses glucose for energy. The scan is then compared to a large database of the scan with the diagnosis of dementia, which identifies the pattern that matches the specific types of dementia, or combinations.

Alzheimer’s typically affects memory and processing regions, levy body dementia includes focus and movement bounded areas, and the frontotemporal dementia ultrasia region of the language and behavior. StateViewer displays these patterns through color-coded brain maps that highlight the major areas of brain activity, gives a visual explanation to all physicians, even without neurology training, even without neurology training and how AI see the diagnosis.

The Mayo Clinic researchers plan to expand the use of the equipment and will continue to evaluate its performance in a variety of clinical settings.