How Artificial Intelligence is Assisting Radiologists and Saving Lives
In a groundbreaking trial at Aberdeen Royal Infirmary, artificial intelligence (AI) is proving to be a game-changer in the field of radiology. The trial aims to determine whether AI can assist radiologists in reviewing thousands of mammograms, potentially revolutionizing breast cancer screenings. The impact has already been seen in the case of June, a participant in the trial, whose early-stage breast cancer was detected thanks to the AI system. Let’s delve into this remarkable development that has the potential to save countless lives.
Mammograms, low level X-rays used for breast cancer screenings, play a crucial role in detecting changes that are too small to see or feel. According to the NHS, these screenings save around 1,300 lives each year in the UK. However, as the number of women attending routine breast screenings increases, the number of radiologists available to review the results is shrinking. This is where AI steps in as a potential solution.
AI, which enables computers to perform tasks that typically require human intelligence, has already made its mark across various industries. In healthcare, its practical applications are becoming evident. Rather than replacing practitioners, AI is seen as a powerful tool that can speed up the process of drug discovery and disease identification. Aberdeen’s AI trial in collaboration with the University of Aberdeen, private industry, and Kheiron Medical Technologies showcases this potential.
The trial utilizes an AI model called Mia, developed by Kheiron Medical Technologies in partnership with Microsoft. Each year, radiologists review around 5,000 mammograms, leading to approximately 250-300 patients being called back for further examination. The risk of missing cancers in such a volume of screenings is a concern. By implementing the AI model as an additional check at the end of mammogram scan reviews, the chances of detection can be significantly improved.
Dr. Gerald Lip, clinical director of the North East Scotland Breast Screening Programme, demonstrated how the AI software works using anonymized mammogram results. The AI identifies differences between two scans, enabling radiologists to focus on areas of concern. This emphasis on early detection is crucial in order to catch potential issues before they escalate.
The AI trial has already shown promising results, with June’s case serving as a prime example. June, a trial participant who has undergone previous surgeries, received a biopsy after the AI tool highlighted an area of concern. June expressed that using AI made the process feel less intrusive compared to relying solely on human judgment. The biopsy confirmed an early-stage cancer, providing reassurance that the technology can indeed aid in early detection.
Scotland’s breast screening program faces challenges due to staff shortages and an aging workforce. The introduction of AI could alleviate some of these concerns by covering a significant portion of the image reading burden. However, Dr. Lip emphasizes that the goal is not to replace human staff entirely but to find the best way to integrate AI into the screening process.
The potential of AI in healthcare extends beyond Aberdeen, with over 30 NHS trusts in the UK planning to implement similar technology. In England, trusts are already exploring how AI can enhance outcomes for breast cancer patients. The role of AI in assisting doctors and improving patient care is set to grow exponentially.
This AI trial in Aberdeen represents a significant step forward in breast cancer screenings. The technology has the potential to transform the way mammograms are reviewed and improve detection rates. As AI continues to evolve and be embraced by healthcare institutions worldwide, we can expect even more remarkable advancements that will save lives and enhance the effectiveness of medical practices.