Research Collaboration Aims to Improve Breast Cancer Diagnosis Using AI
Credit: Dr. Dwight Kaufman, National Cancer Institute
A new project to explore how artificial intelligence (AI) could improve breast screening could potentially lead to more accurate detection of cancers.
The project will see a consortium of leading breast cancer experts, clinicians and academics partner with leaders in artificial intelligence (AI) research to explore whether AI could help detect and diagnose breast cancers more effectively.
Led by Imperial College London, the consortium will be based at the Cancer Research UK Imperial Centre.
Clinicians and radiologists from the Centre will work on the research with technical partners at DeepMind Health and the AI health research team at Google, alongside the Cancer-Research UK funded OPTIMAM mammography database at the Royal Surrey County Hospital NHS Foundation Trust.
Breast cancer is the most common cancer in women worldwide, with 1.6 million women around the world diagnosed every year.
In the UK alone, over 150 patients are diagnosed with breast cancer every day, and despite medical advances, it still kills around 11,000 women in this country each year, and 500,000 people globally.
While early detection and treatment is shown to lead to better outcomes for women with cancer, accurately detecting and diagnosing breast cancer remains highly challenging. Mammograms (an X-ray of the breasts) are used by clinicians to identify cancers early, but breast screening is not perfect.
Thousands of cases are not picked up by mammograms each year, including 30% of cancers that develop between screenings, while false alarms and cases of overdiagnosis are also common.
The new collaboration aims to understand if machine learning tools can help doctors to address these challenges. Machine learning is a form of artificial intelligence in which computer algorithms can learn and improve without being explicitly told how.
Towards increased accuracy
The team believes it has the potential to increase the accuracy of breast screening interpretation, improving the ability to detect breast cancers on mammograms.
The project may also provide women with a better estimate of their risk of breast cancer, which could help guide them in taking actions to prevent it.
As part of this project, machine learning technology from DeepMind Health and the AI health research team at Google will be applied to historic ‘de-identified’ mammograms from around 7,500 women provided by the Cancer Research UK-funded OPTIMAM database at the Royal Surrey County Hospital NHS Foundation Trust.
These digital images have been stripped of any information which could be used to identify patients and have been available to research groups around the world for a number of years.
Through this research, the team hope to explore whether it is possible to train computer algorithms to analyse these images, to spot signs of cancerous tissue and alert radiologists more accurately than current techniques allow.
It is hoped that further international research partners will join the project over the next 12-months so that this research, if successful, could eventually lead to technology that will help clinicians around the world to make more accurate diagnoses, leading to earlier detection and intervention for patients.
Professor Ara Darzi, Director of the Cancer Research UK Imperial Centre, said: “This partnership marks an exciting exploration of the potential for artificial intelligence in healthcare."
He added: "Ultimately, we want this kind of technology to benefit patients and it may be a number of years until this kind of approach is used, but if these initial trials prove successful, AI could make screening services for cancer far more efficient and improve outcomes.”
Dr Iain Foulkes, Cancer Research UK’s Executive Director of Research and Innovation, said: “Harnessing the power of artificial intelligence could enable us to address some of the biggest challenges in breast cancer research, including improving the accuracy of detection.
"Too many cancers are detected at a late stage when they are more difficult to treat. This is why Cancer Research UK is building capacity, forging new partnerships, and supporting a community for early detection research so that more people might survive their disease.”
Dominic King, Clinical Lead, DeepMind Health, said: “We set up DeepMind because we wanted to use artificial intelligence to help solve some of society's biggest challenges – challenges exactly like breast cancer. For that reason, we’re incredibly excited to be part of this collaboration, working alongside leading researchers to try and tackle this problem, and bring real benefits for cancer patients across the world, and the clinicians who treat them.”
Professor Anne Mackie, Director of Programmes for the UK National Screening Committee at Public Health England, commented: “Mammography is a key service for the NHS in the early detection of breast cancer. Exploring innovative new technologies such as machine learning has the potential to increase the efficiency of these services and make better use of the resources available to NHS screening programmes.”
Dr Nicola Strickland, President of the Royal College of Radiologists and Consultant Radiologist at ICHNT, said: “Image-based population screening to pick up cancer early, when it is still curable, is becoming more and more important. The use of machine learning in medical imaging is still in its infancy, but this exciting research approach could help boost our diagnostic capacity and potentially increase our diagnostic accuracy, as well as speed up the detection of early cancers, which would be of immense benefit to patients.”
This article has been republished from materials provided by Imperial College London. Note: material may have been edited for length and content. For further information, please contact the cited source.
MIT researchers have developed a cryptographic system that could help neural networks identify promising drug candidates in massive pharmacological datasets, while keeping the data private. Secure computation done at such a massive scale could enable broad pooling of sensitive pharmacological data for predictive drug discovery.