AI Study Finds Robots Could Halve Radiologists’ Workload, Spot Breast Cancer
Robots equipped with artificial intelligence (AI) may have the potential to reduce radiologists’ workload by almost 50% and address the national shortage of skilled professionals, according to a study published in the Lancet. The research suggests that AI-supported mammography screenings, where machine learning takes the place of one radiologist, are as effective as having two human experts in identifying breast cancer. Currently, the National Health Service (NHS) requires double reading of such scans, but with a shortage of radiologists and vacancies in over 30% of positions, the implementation of machine learning technology could alleviate this burden.
The study, a randomized control trial involving more than 80,000 women from Sweden with an average age of 54, compared the performance of computer-aided detection to that of two human radiologists. The researchers found that AI-supported screenings detected cancer at a similar rate to the standard practice involving two radiologists. The AI system did not produce more false positives – incorrect diagnoses of abnormalities – compared to human experts. Furthermore, the use of AI reduced the workload of radiologists by 44%, resulting in 36,886 fewer screen readings.
While studies have previously explored the accuracy of AI in diagnosing breast cancer retrospectively, this study directly compared AI-supported screening to standard care. The researchers concluded that the combined expertise of radiologists and AI could potentially improve patient outcomes, especially in detecting interval cancers that are often missed during traditional screenings.
The potential benefits of AI in breast screening have not gone unnoticed by the NHS. The organization is already exploring AI’s capabilities in analyzing complex images rapidly and at scale, which could lead to faster diagnoses, early detection of cancers, and ultimately, saving lives. With a shortage of radiologists in the UK at 29%, implementing AI systems could alleviate the workforce issues faced by the healthcare system.
Medical professionals have expressed enthusiasm for the study’s findings. Dr. Katharine Halliday, President of the Royal College of Radiologists, believes that AI holds great promise in increasing efficiency, aiding decision-making, and identifying urgent cases. Professor Fiona Gilbert, a radiology expert from the University of Cambridge, highlights the considerable manpower savings that AI could bring to address workforce challenges in the UK.
While the study highlights the potential benefits of AI in breast cancer screening, it is important to exercise caution and evaluate the long-term implications on patient outcomes. More research is needed to determine how AI, combined with radiologists’ expertise, can improve the detection of interval cancers.
The study’s findings will aid in planning the testing and implementation of AI in the UK’s national breast screening program. With its potential to reduce workload and address the shortage of radiologists, AI represents a promising tool in the fight against breast cancer. As advancements in technology continue, the integration of AI systems into healthcare practices could revolutionize the field and improve patient care.
