AI platform helps diagnose prostate cancer
The findings suggest AI systems can be trained to detect and grade cancer in prostate needle biopsy samples with an accuracy rate equal to that of prostate pathology experts.
A team of researchers in Sweden has discovered that an artificial intelligence platform is capable of accurately diagnosing prostate cancer in tissue samples, offering the potential to speed up diagnostics and reduce costs for healthcare services
The findings, published in The Lancet Oncology in December, suggest AI systems can be trained to detect and grade cancer in prostate needle biopsy samples with an accuracy rate equal to that of international prostate pathology experts.
Furthermore, the study noted that the use of AI technology could help reduce the workload of oncologists by reducing the assessment of benign biopsies and by automating the task of measuring cancer length in positive biopsy cores, as well as providing a second opinion.
“An AI system with expert-level grading performance might aid in standardizing grading, and provide pathology expertise in parts of the world where it does not exist,” the report noted.
The system was developed by the team behind Stockholm3 and OncoWatch, two projects supported by EIT Health, a network of top health innovators backed by the EU.
A team at Karolinksa Institutet launched Stockholm3, a blood-based prostate cancer diagnostic test, in 2017, which is currently used in clinical practice in Sweden, Norway, Finland and Denmark.
“Our AI tool has the potential to reduce the workload of uropathologists, allowing them to focus on the most difficult cases and at the same time act as a safety net to improve quality,” said Martin Eklund, associate professor at Karolinska’s Department of Medical Epidemiology and Biostatistics, in a statement.
The announcement marks the second time this month that AI-based technologies have proven to be of use to oncologists: Google’s AI platform has aided oncologists in breast cancer screenings.
A study published in Nature indicated Google’s the model was able to spot cancer in de-identified screening mammograms with fewer false positives and false negatives than experts, and was also able to more effectively screen for breast cancer using less information than human doctors.
As artificial intelligence and robotic process automation are more widely deployed in 2020, they are expected to help re-humanize medicine by allowing doctors to focus less on paperwork and administrative functions, and more on patient care.
A study of Brigham & Women’s Home Hospital program, recently published in Annals of Internal Medicine, employed AI-driven continuous monitoring combined with advanced physiology analytics and related clinical care as a substitute for usual hospital care.
The study found that the program–which included an investment in AI-driven predictive analytics as a key component–reduced costs, decreased healthcare use, and lowered readmissions while increasing physical activity compared with usual hospital care.
Meanwhile, pharmaceutical company Takeda recently announced plans to work with MIT’s School of Engineering to establish an education program focused on developing new machine learning approaches to drug development and other AI applications.