Pioneering Study on Use of AI in Prostate Cancer Diagnosis to Be Expanded

Priscilla Lynch 

September 07, 2021

A groundbreaking trial of artificial intelligence (AI) in prostate cancer diagnosis is to be expanded to six new NHS trusts in England.

Funded as part of the £140 million NHSX AI in Health and Care Awards, the study will enable researchers to evaluate the effectiveness of the AI solution Galen Prostate in detecting and grading cancer in prostate biopsies using samples from 600 men over 14 months. Clinicians will compare the results of the AI analysis with current diagnosis methods, where biopsies are reviewed by pathologists.

The funding will be used for deploying and evaluating the AI technology, developed by health technology company Ibex Medical Analytics, with the potential for it to be adopted more widely across health services, cutting diagnosis times and freeing up valuable clinician time.

The trial represents the largest multisite deployment of AI in the UK. Nearly 100,000 men undergo a prostate biopsy every year in the UK—a number expected to double in the next 10 years.

Matthew Gould, NHSX CEO, said: “We are currently caught between having too few pathologists and rising demand for biopsies. This technology could help, and give thousands of men with prostate cancer faster, more accurate diagnoses.”

“It is a prime example of how AI can help clinicians improve care for patients as we recover from the pandemic.”

Prof Hashim Ahmed, Chair of Urology at Imperial College London and principal investigator, said: “We strongly believe that AI has the potential to enhance both quality and efficiency, which is of paramount importance as we focus on putting every patient on the path to recovery.”

“Ibex’s technology has demonstrated its robustness on several studies abroad and so we look forward to seeing its performance and utility first hand in the NHS.”

Meanwhile, a newly published systematic review in the  BMJ  on the accuracy of AI algorithms in breast cancer detection concluded that current evidence for AI does not yet allow judgement of its accuracy in breast cancer screening programmes, adding that it is unclear where on the clinical pathway AI might be of most benefit. The study authors said that AI systems are not sufficiently specific to replace a radiologist double reading in screening programmes and that promising results in smaller studies have not been replicated in larger studies to date.

This article originally appeared on Univadis, part of the Medscape Professional Network.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.