Movember: What can AI do for prostate cancer?
Artificial Intelligence Centre for Value Based Healthcare
A consortium of academic, health and industry partners working to develop & deploy AI systems across the NHS.
This month, the UK is raising awareness and money for #Movember. This wonderful cause is in aid of men’s health, including; mental health, testicular cancer, and prostate cancer. You can find out more about the charity and their fund raising on their website.??
Prostate cancer is the most common male cancer. In the UK there are around 52,000 new cases each year- approximately 143 every day.??
The AI Centre ( Artificial Intelligence Centre for Value Based Healthcare ) is a cross-institutional consortium of leading AI, data science, research, and clinical experts. AI has a huge part to play in public health screening and treatment. Siemens Healthineers, in collaboration with the AI Centre, has identified a need for improved, objective criteria for determining which prostate cancer patients should remain on active surveillance, and which should switch to active treatment.??
Questions with Michela Antonelli, PhD, School of Biomedical Engineering & Imaging Sciences, King’s College London?
Dr Michela Antonelli is a senior research fellow at King’s College London. She is currently working on several projects involving prostate cancer. Here she speaks about the Siemens Healthineers – AI Centre collaborative AI prostate cancer project.?
Can you provide us with some context as to the need for this project???
I am currently working on three AI projects for prostate cancer – prostate cancer is one of the most prevalent cancers in men in the UK, and as such, there is a huge need to streamline diagnosis and treatment.??
This particular project that I am working on with support from the AI Centre is focused on active surveillance of prostate cancer. Active surveillance is a conservative management protocol for prostate cancer, offered to low- and intermediate-risk patients, which avoids long-term adverse effects on the patient’s quality of life. The idea is to find a way to structure and automate the steps clinicians should follow to enroll and monitor patients in active surveillance. We hope to support clinicians in the questions they have surrounding this pathway; when should screening be carried out? What sort of screening? How often? How long should these patients follow the protocol???
Currently, there are no clear protocols for clinicians to follow in the progression from patient being enrolled, to being flagged for an upstaging, to treatment.??
From a patient perspective, what we are primarily trying to avoid is unnecessary biopsies. This is an invasive procedure and can result in infections or other complications and an emotional burden on the patient. From a public health perspective, you can save costs by eliminating unnecessary examinations and treatments.??
What were the main challenges you faced???
One large hiccup in our project planning was COVID-19. Like many projects, we had to put pause on our research until a year ago. We have now hit the ground running and are keen to start development work as soon as possible.?
The project is in the data collection stage; we have collected data from over 100 patients with 3 different time points. One of the largest bottlenecks in this project is the data collection point. We require multidisciplinary input, particularly radiologists, to identify significant lesions on the imaging. We cannot wait to start AI development but to begin this we need to make sure we have a data set of high enough quality, and that takes time. ?
What is the long-term ambition for this project???
In my mind there are three ways that this AI project could contribute to the prostate cancer pathway:?
I touched on preventing unnecessary procedures earlier but reducing staffing needs is another massive benefit of this AI project. In breast cancer, two radiologists are needed to read any screening imaging. This is used as a form of quality control. Our AI project will act as a companion system to reduce the staffing need, and as such reduce public health costs.?
This project is in collaboration with Professor Vicky Goh, Dr Davide Prezzi, Dr Aishah Azam, Karen Welsh, Dr Amit Samani and Pritesh Mehta. ?
AI has enormous potential in improving diagnosis to treatment pathways, and public health screening. If you are interested in collaborating with the AI Centre on AI projects, please contact us at [email protected] or by using the form here.??
?