Two major advances are making it easier for doctors to diagnose the correct type of dementia and catch it earlier: a newly approved brain imaging tracer that spots tau tangles with greater sensitivity, and artificial intelligence tools that can distinguish between multiple dementia types from routine scans and blood tests. These developments could help patients get the right treatment sooner, especially as new medications targeting different forms of dementia become available.

What Is the New Tau PET Imaging Tracer, and How Does It Work?

In August 2026, the FDA approved a new tau PET imaging tracer called MK6240, also known as Tauflarify, which detects tau tangles in the brain more effectively than the previously approved tracer Tauvid. Tau proteins normally stabilize the internal structures that keep neurons healthy and functional, but in Alzheimer's disease, they form tangles that kill off brain cells.

In a head-to-head comparison published in The Lancet, researchers tested both tracers on 682 individuals living in the US and Europe. The new tracer spotted tau tangles earlier and more often than Tauvid, even in people without cognitive symptoms. Most importantly, the study found that switching to the new tracer could help identify an extra 23 cases of Alzheimer's for every 100 people tested.

"The picture's the same, but the newer tracer brings the earliest changes into focus," said Neil Vasdev, scientific director of the Brain Health Imaging Center.

Neil Vasdev, Scientific Director, Brain Health Imaging Center

The new tracer was also more accurate at distinguishing between cognitive impairment caused by Alzheimer's, where beta-amyloid protein is also present in the brain, and impairment caused by other conditions. More than a third of study participants were placed at different disease stages depending on which tracer was used, with the newer tracer placing them further along in the disease process.

How Could Better Imaging Help With Treatment Decisions?

Currently, tau PET scans are rarely used to diagnose Alzheimer's in routine clinical practice. However, experts believe that more sensitive imaging could help doctors predict how quickly a person's disease will progress and identify who would benefit most from new medications. In clinical trials of Kisunla, one of two FDA-approved anti-amyloid medications, there was evidence that individuals with more tau in their brain benefited somewhat less than those with less tau.

Researchers are also optimistic that better tau imaging will become invaluable as new anti-tau treatments enter the market. Several medications targeting tau are in development, including Biogen's gene-silencing therapy diranersen and Eisai's etalanetug. These tools could help doctors track how well these medications lower tau levels in the brain.

"It really is how much and where the tau is that might help us predict for an individual person how fast they're likely to progress," said Dr. Gil Rabinovici, a neurologist at UCSF.

Dr. Gil Rabinovici, Neurologist, UCSF

What Role Could Artificial Intelligence Play in Dementia Diagnosis?

Beyond improved imaging tracers, artificial intelligence is emerging as a powerful tool to help doctors distinguish between different types of dementia, which is notoriously difficult because symptoms and brain changes overlap significantly. A new generation of AI-powered brain scans and blood tests is being developed to help neurologists make faster, more accurate diagnoses.

Researchers at the University of Florida developed an AI algorithm trained on routine MRI scans that measure how fluid moves in the brain. When tested on over 70 new images, the algorithm could distinguish between Alzheimer's and dementia with Lewy bodies (DLB), a condition often misdiagnosed as Alzheimer's, with 99 percent accuracy. The researchers confirmed these predictions by examining the post-mortem brains of 13 participants.

At Mayo Clinic, neurologist Dr. David T. Jones developed a tool called StateViewer that analyzes FDG-PET scans, which measure brain metabolism. In a study of over 3,600 older adults, the algorithm could pinpoint the correct type of dementia with 88 percent accuracy. Radiologists who received information from the algorithm were over three times more likely to make the correct diagnosis compared to those without it.

"It can provide supportive evidence across that entire continuum of degenerative and non-degenerative neurologic causes," said Dr. David T. Jones.

Dr. David T. Jones, Neurologist, Mayo Clinic

StateViewer is now helping differentiate between 14 different types of dementia at Mayo Clinic, including treatable forms like normal pressure hydrocephalus, a condition that can sometimes be reversed with surgery.

Can a Blood Test Detect Multiple Types of Dementia?

Researchers at Washington University in St. Louis used AI to analyze blood protein signatures from more than 3,200 people with various forms of dementia. They identified an ideal combination of 15 different proteins that could help distinguish between Alzheimer's, Parkinson's disease, dementia with Lewy bodies, frontotemporal dementia, and combinations of these diseases with 90 percent accuracy.

This approach recognizes an important reality: most people with dementia have a mix of protein aggregates in their brain, not just one type. Beta-amyloid and tau define Alzheimer's, but alpha-synuclein and TDP-43 are common across other dementias. Understanding which proteins are present helps determine what treatments and clinical trials a person may be eligible for.

"We cannot just think that these diseases are totally independent and isolated. We really need to understand what is happening in the brain to understand the patient's journey," said Carlos Cruchaga, director of the NeuroGenomics and Informatics Center at Washington University in St. Louis.

Carlos Cruchaga, Director, NeuroGenomics and Informatics Center, Washington University in St. Louis

The blood test is not yet FDA approved, and researchers estimate that approval could take over five years. However, if successful, the test could be used to screen people showing signs of cognitive impairment before sending them for imaging tests for confirmation.

How to Prepare for Dementia Screening and Diagnosis

• Know Your Family History: Understanding whether dementia runs in your family can help you and your doctor determine if early screening is appropriate, especially since these new diagnostic tools may become available for routine use in the coming years.
• Track Cognitive Changes: Keep notes on any memory problems, confusion, or difficulty with familiar tasks, and share these observations with your doctor during appointments to help with early detection.
• Ask About Biomarker Testing: If you or a loved one is experiencing cognitive concerns, ask your neurologist whether biomarker tests, such as blood tests or advanced imaging, might be appropriate to help identify the specific type of dementia and guide treatment decisions.
• Discuss Eligibility for Clinical Trials: As new treatments for different types of dementia emerge, ask your doctor whether you might qualify for clinical trials, which often require accurate diagnosis and biomarker information.

Why Accurate Diagnosis Matters Now More Than Ever

For decades, dementia diagnosis relied primarily on clinical presentation, cognitive testing, and brain imaging, making it difficult to distinguish between conditions with overlapping symptoms. The challenge is significant: about 30 percent of people over 75 years old have beta-amyloid plaques in their brain, but only 10 percent of those individuals actually develop Alzheimer's disease.

With two FDA-approved anti-amyloid medications now available (Leqembi and Kisunla) and more treatments targeting tau and other proteins in development, the ability to accurately identify which type of dementia a person has becomes increasingly important. Getting the diagnosis right means patients can receive the treatments most likely to help them, rather than medications that may not address their specific condition.

These advances in imaging and AI represent a shift toward precision medicine in dementia care. As these tools become more widely available and validated, they could help scale expert-level diagnosis to more patients and communities, potentially catching dementia earlier when treatments may be most effective.