Qynapse, a technology company using AI to enhance the quantification and predictive capabilities of imaging to improve outcomes for neurological diseases, including Alzheimer's disease (AD), presented new data at the Alzheimer's Association International Conference (AAIC) for QyScore^® and QyPredict^®.

QyScore^® is an FDA-cleared and CE-marked proprietary neuroimaging software platform that enables fully automated, high-resolution segmentation of brain structures from MRI, supporting clinical and research applications in neurodegeneration. QyScore^® demonstrated higher segmentation accuracy and greater reliability over other established products, including FreeSurfer, FSL, and ANTs, across key regions of the brain associated with neurological diseases.

QyPredict^®, is a predictive research-use-only platform that uses QyScore^® outputs to identify patients more likely to experience clinical decline from baseline. QyPredict^® was shown to accurately identify patients with preclinical AD and mild cognitive impairment (MCI) who were more likely to experience cognitive decline over 24 months. Specifically, the model was able to filter out 73.7% of predicted stable participants from the modeled placebo group, resulting in a twofold increase in observed cognitive decline in the control group (p < .001). These findings build upon earlier results showing predictive value at 12 months in a combined MCI and AD population. 

Dr. James E Galvin, Professor of Neurology at the University of Miami, notes, "AD treatments will be most effective when delivered early, however this is also when it is most difficult for clinicians to predict rate of decline. Improving the ability to identify early which patients experience progressive cognitive decline due to AD is essential for evaluating disease-modifying medications and for determining which patients will benefit from treatment. While more work is needed to validate QyPredict^®, this preliminary data is very promising." 

According to Olivier Courrèges, CEO of Qynapse, "These promising results suggest that imaging biomarkers and predictive modeling can play a crucial role in improving patient selection and powering clinical trials to detect meaningful treatment effects. We are very encouraged by these new and expanded findings and look forward to further validating our offerings through collaborations with drug developers to bring products to market that enhance outcomes for patients, healthcare providers, and payers."