Finding Dementia Early Using Non-Invasive MRI

New research reveals that a non-invasive MRI technique could help doctors detect dementia before irreversible brain damage occurs, offering hope for earlier intervention and improved patient outcomes.

Shaheen Haidrus, MD, presented the results of a study that explored using arterial spin labeling (ASL) MRI at 3 Tesla (3T) as a potential tool for quantifying cerebral blood flow (CBF). She and her colleagues aimed to detect early functional changes related to dementia before structural atrophy becomes evident.

“Dementia is frequently diagnosed after irreversible neuronal loss has already occurred,” said Dr. Haidrus, a third-year radiology resident in the Department of Radiodiagnosis at Subharti Medical College & Chhatrapati Shivaji Subharti Hospital in Meerut, India. “The need for accessible, radiation-free biomarkers in centers that lack PET or SPECT facilities served as motivation for our research.”

A total of 92 individuals were included in the study: 53 people diagnosed with dementia via the DSM-5, and 39 healthy controls matched by sex and age. Dr. Haidrus and fellow researchers assessed regional cerebral perfusion in these patients using 3T ASL-MRI. To do so, they developed a fully automated process that extracts quantitative perfusion metrics and integrates machine learning tools to aid in early detection of dementia. The system was designed to run from start to finish with very little manual work, making it efficient and easy to use.

An Effective Biomarker for Screening and Follow-up

Dr. Haidrus and colleagues found that people with dementia had much lower average cerebral blood flow (CBF) compared to healthy individuals. She noted the difference was highly significant. Specifically, the group with dementia’s average was about 43 milliliters per 100 grams of brain tissue each minute, while the healthy group’s average was about 54 milliliters.

The researchers also discovered that the group with dementia had lower values on normalized z-CBF values, further indicating reduced brain blood flow. Additionally, the study showed that higher overall brain blood flow was linked to better scores on a common memory and thinking text, the Mini-Mental State Examination, or MMSE.

Dr. Haidrus said patients with Alzheimer’s disease had pronounced hypoperfusion in the frontal opercular cortex, lingual gyrus and temporal lobes. Those with vascular dementia exhibited diffuse hypoperfusion.

“We observed unexpectedly prominent frontal hypoperfusion in vascular and mixed dementias, often preceding visible white matter changes,” Dr. Haidrus asserted, noting that the correlation between perfusion z-scores and MMSE scores was stronger than anticipated. She also emphasized ASL’s sensitivity to functional decline.

“Our AI classifier also performed best for Alzheimer’s patients and controls, highlighting the importance of balanced data for rarer subtypes,” Dr. Haidrus said.

“ASL-MRI is a repeatable, contrast-free and cost-effective biomarker for screening and follow-up,” Dr. Haidrus said. “It facilitates early diagnosis, subtype differentiation and longitudinal monitoring without radiation exposure, making it especially valuable for resource-limited settings and follow-ups.”

“I completed this research under the guidance of the Dr. Mahesh Kumar Mittal, professor & principal, Department of Radiodiagnosis; Dr. Vivek Kumar, professor & head, Department of Psychiatry, Dr. Akanksha Singh, associate professor, Department of Radiodiagnosis; and Dr. Vishal Vishnoi, assistant professor in the Department of General Medicine,” Dr. Haidrus said. “Our goal is to standardize ASL quantification on vendor MRI workstations so that quantitative perfusion can be incorporated into routine dementia protocols.”

Access the session, “Role of Cerebral Perfusion Using 3 Tesla Arterial Spin Labeling MRI in Dementia Patients,” (S1-STCE2-3) on demand at RSNA.org/MeetingCentral.