Pavilion Publishing and Media Ltd
Blue Sky Offices Shoreham, 25 Cecil Pashley Way, Shoreham-by-Sea, West Sussex, BN43 5FF, UNITED KINGDOM
NICE has recommended five new artificial intelligence (AI) tools in draft guidance that could help doctors detect prostate cancer more accurately on MRI scans.
Demand for prostate MRI scans has increased in recent years, placing additional pressure on radiology services, with more than 58,000 new diagnoses of prostate cancer in England in 2023.
The consultation will review whether radiologists can use new AI technologies to help detect prostate cancer on MRI while further evidence is generated.
The AI technologies can identify and outline possible areas of cancer, classify risk and produce information to support radiologists when interpreting MRI scans, but NICE says further evidence is needed to understand how they perform in routine NHS practice.
Prostate cancer diagnosis
The aim is to free up time and capacity for clinicians to focus on supporting patients, as not all prostate cancer tumours need treatment. Some can instead be monitored safely for changes, as they are unlikely to cause symptoms. But MRI scans can play an important role in helping clinicians distinguish whether someone may need a biopsy or other tests to confirm if cancer that needs treatment is present.
Recommended technologies are:
- AI-Rad Companion Prostate MR
- DeepHealth Prostate
- ProstatID
- Prostate Intelligence (Pi)
- QP-Prostate.
The independent NICE committee concluded that the technologies could improve the accuracy and consistency of prostate MRI interpretation, particularly in settings without specialist expertise. The committee found that the technologies have potential, but the available evidence is limited, and findings were not always consistent across studies. Further NHS evidence is therefore needed on diagnostic accuracy, patient outcomes, NHS workflows, costs and performance across different groups of people.
AI technologies would support, not replace radiologists
Dr Adrian Hayter, GP and chief medical officer at NICE, said: “These tools could help radiologists identify clinically significant prostate cancer more accurately and consistently, while helping NHS services manage growing demand for scans.
“They would support, not replace, radiologists, who would still review the scan and the AI results before any decisions about patient care.”
Dr Anastasia Chalkidou, programme director for HealthTech at NICE, added: “These AI technologies have the potential to act like a second pair of eyes, helping radiologists review prostate MRI scans. They have the potential to make MRI reporting more accurate and could help identify people whose scans need priority review.
“Our committee concluded that the technologies can be used in the NHS while further evidence is collected on how they perform in routine practice and across different groups of patients.”
Two further technologies, hProstate and mdprostate, were not recommended for NHS funding because no clinical or cost-effectiveness evidence was available.