Scientists from Nanjing University have developed an artificial intelligence (AI) system which can diagnose cancerous prostate samples as accurately as any pathologist.
Using machine learning, the system was trained to diagnose samples with an accuracy of 99.38%.
The software can also accurately classify the level of malignancy of the cancer, eliminating the variability that can be present in human diagnosis.
It is hoped that the technology can be used alongside pathologists to speed up the diagnosis process.
“This is not going to replace a human pathologist,” said Nanjing University research leader Hongqian Guo. “We still need an experienced pathologist to take responsibility for the final diagnosis.
“What it will do is help pathologists make better, faster diagnosis, as well as eliminating the day-to-day variation in judgement which can creep into human evaluations”.
The research team took 918 prostate whole-mount pathology section samples from 283 patients and ran them through the analysis system. These pathology images were subdivided into 40,000 smaller samples, with 30,000 used to ‘train’ the software and 10,000 used to test accuracy. Over time the software learned and improved diagnosis.
The results showed a near-100% accuracy when the accuracy of a human pathologist was used as a gold standard.
However, the system is yet to be tested with human patients.
Professor Rodolfo Montironi from the Polytechnic University of the Marche said of the technology: “This is interesting work which shows how artificial intelligence will increasingly step into clinical practice.
“This may be very useful in some areas where there is a lack of trained pathologists. Like all automation, this will lead to a lesser reliance on human expertise, but we need to ensure that the final decisions on treatment stay with a trained pathologist.”
Prostate cancer is the most common male cancer, with around 1.1 million diagnoses every year worldwide.
The artificial intelligence learning system was presented at the European Association of Urology congress in Copenhagen.