A new research study aims to improve early detection of prostate cancer
"AI has exciting potential to enhance the processes of screening and treating patients, and this study will provide strong evidence on its performance." Dr Aarti Shah, consultant radiologist at Hampshire Hospitals Foundation Trust
Hampshire Hospitals NHS Foundation Trust (HHFT) is to launch a research project investigating the use of artificial intelligence (AI) in identifying signs of prostate cancer.
The retrospective study, named PAIR-1 (Prostate AI Research – 1), will collect anonymised data on 2,100 patients who have received a prostate cancer diagnosis at seven different centres. Using Pi (Prostate Intelligence) software from UK startup Lucida Medical, the study will analyse the patients’ MRI scans, using machine learning, image processing and radiogenomics. This is a technique for shedding light on underlying disease mechanisms by establishing links between genetic information and the features displayed on a radiology image, making it possible to predict the presence or absence of genetic mutations in a tumour.
Because performance of AI tools can vary between hospitals and different makes of scanner, one limitation of previous studies has been that images are usually taken from a single centre. This study will address that by testing the software on images taken from different NHS hospitals and all major MRI scanner manufacturers.
Interpreting MRI scans is time-consuming
Half of the patient data will be used to make sure that the software works across the full range of scanners and scanning methods in the different hospitals. This provides an opportunity to check that the AI software is correctly calibrated. The remaining data will be used to test how well the software works with unseen patients.
In the UK, prostate cancer is the most common cancer in men, and nearly 12,000 die from the disease every year. More than half of prostate cancer cases in England are detected late.
MRI scans are now widely used for prostate cancer diagnosis to reduce the use of biopsies. Interpretation of scans is time-consuming and requires expertise, however. If AI tools can be used both to speed up and to improve the accuracy of diagnosis, then it may help detect cases of prostate cancer earlier and save lives.
Dr Aarti Shah, consultant radiologist at HHFT and chief investigator for the PAIR-1 study, said:
“Reviewing prostate MRI requires experience and expertise to ensure that the right patients have a biopsy as well as to help target biopsies to maximise the chances of finding significant cancers. AI has exciting potential to enhance the processes of screening and treating patients, and this study will provide strong evidence on its performance.”
If the study is successful, the technology has the potential to help the 47,000 patients who develop prostate cancer every year in the UK.
FCC has recently reported on a WHO global report offering six guiding principles for AI design and use in health.