Researchers are training AI to detect breast cancer

"Its potential use in diagnosing cancers early is phenomenal”

Researchers at Cardiff University have developed an artificial intelligence (AI) system with the potential to assist doctors in detecting breast cancer earlier and more accurately. The system is specifically designed to support radiologists in their decision-making process, aiming to enhance medical diagnostics.

Hantao Liu, from Cardiff University, emphasised the importance of creating AI systems that radiologists can trust and work with effectively. To achieve this, the researchers collaborated with radiologists from NHS hospitals in Wales and England. They recorded the eye movements of a radiologist while examining medical images and fed this data into the AI system to learn the radiologists' visual patterns.

Once trained, the AI system can identify regions on a scan that radiologists are more likely to focus on when making a diagnosis. "With all of the challenges facing the NHS, it is important that we look to data science and AI for possible solutions," Mr Liu said.

Cancer charity Tenovus expressed its enthusiasm for the research and recognised the tremendous potential of AI in diagnosing cancers early. Chief Executive Judi Rhys noted that AI is not meant to replace radiologists but rather to enhance their responsiveness. “Its [AI's] potential use in diagnosing cancers early is phenomenal," she said. By highlighting worrisome scans, the system could accelerate the process for people who are more likely to have cancer.

“It is important that we look to data science and AI for possible solutions.”

The accuracy of the AI technology has been demonstrated through various studies. Moreover, its continuous operation without the need for breaks allows it to process a vast number of scans, offering vital support to radiologists, especially considering the 30% shortage of radiology consultants across the UK.

Judi Rhys emphasized the significance of such AI tools in assisting and supporting radiologists in their work, which ultimately benefits patients. With AI's potential to aid in early cancer detection, it could play a pivotal role in improving healthcare outcomes. While AI may have had a negative reputation, its application in the field of medicine shows tremendous promise and opens up new possibilities for advancing medical diagnostics and patient care.