April 20, 2022—PreciseDx announced today its artificial intelligence–enabled digital pathology technology can accurately diagnose Parkinson’s disease in patients prior to the severe onset of symptoms.
The PreciseDx study applied the company’s AI algorithms, Morphology Feature Array, to detect alpha-synuclein within peripheral nerves of salivary glands. Detection, along with quantitative morphology feature extraction, accurately identified Lewy-type synucleinopathy in early-stage Parkinson’s disease biopsy specimens. The training and validation of the algorithms were based on expert pathologist annotated biopsy images, the company said. The algorithms were validated using a separate set of confirmed biopsy specimens. Morphology Feature Array was able to detect Parkinson’s pathology in image patches from biopsy samples with 99 percent sensitivity and 99 percent specificity as compared to expert annotated ground truth.