New York: Researchers have developed a new artificial intelligence (AI) system to help radiologists improve their ability to diagnose prostate cancer.
The system, called FocalNet, helps identify and predict the aggressiveness of the disease by evaluating magnetic resonance imaging (MRI) scans, and does so with nearly the same level of accuracy as experienced radiologists.
FocalNet is an artificial neural network, which uses an algorithm that comprises more than a million trainable variables, according to researchers, including Ruiming Cao from University of California, Los Angeles.
For the study, published in the journal IEEE Transactions on Medical Imaging, the team trained the system by having it analyse MRI scans of 417 men with prostate cancer. Scans were fed into the system to help it learn to assess and classify tumours in a consistent way and then compare results to the actual pathology specimen.
The team compared the system’s results with readings by UCLA radiologists with more than 10 years of experience.
The findings showed the AI system was 80.5 per cent accurate in reading MRIs, while UCLA radiologists were 83.9 per cent accurate.
Radiologists use MRI to detect and assess the aggressiveness of malignant prostate tumours. But it typically takes practicing on thousands of scans to learn how to accurately determine whether a tumour is cancerous or benign and to accurately estimate the grade of cancer, the team said.
The research suggests an artificial intelligence system could save time and potentially provide diagnostic guidance to less-experienced radiologists.