AI Shows Improved Accuracy In Diagnosis in Pathology
Ronald M. Razmi, MD
October 25, 2022
In 2021, the FDA granted its first clearance for a cancer diagnosis AI program to Paige, a New York-based company launched in 2018 with data and digital pathology tech from Memorial Sloan Kettering Cancer Center. The product, Paige Prostate, analyzes slides of biopsied prostate tissue to spot the hallmarks of malignant cells. The software highlights areas of a standard prostate biopsy image that are most likely to contain cancer and flags them for further review by a trained professional. One of the benefits highlighted for this software is the increased efficiency of the diagnostic workload for the pathologists. Increasing the efficiency of diagnostic workloads is imperative because those workloads are growing fast.
There are projections of a 60% rise in the number of global cancer cases over the next two decades—while the number of available pathologists is not expected to keep pace. The use of AI-based software can help increase the output of pathologists in examining biopsy samples, which can ultimately save lives.
The approval was based on clinical study evaluating the program’s use by 16 pathologists. Each evaluated a mix of cancerous and benign tissue slides, both with and without Paige Prostate’s assistance. The samples were gathered from more than 150 institutions to test the AI’s ability to work in different hospitals and geographies.
While a final patient diagnosis may be based on multiple biopsies and other exam information, the AI program did help improve overall cancer detection by an average of 7.3%. The software also helped cut down the number of false-negative diagnoses by 70% and false positives by 24%.