AI Shows Promise in Detecting Breast Cancer: Study Finds Increased Accuracy and Reduced Radiologist Workload
A recent study has revealed that artificial intelligence (AI) has shown promise in detecting breast cancer, offering increased accuracy and reducing the workload of radiologists. The interim results of the trial have been hailed as promising, but the authors have cautioned that further research is needed before AI can be used on a wider scale for breast cancer screening.
The shortage of radiologists in many countries has prompted the exploration of AI technology to analyze routine medical scans efficiently and accurately. Breast cancer, in particular, stands to benefit greatly from this advancement. According to the World Health Organization, over 2.3 million women were diagnosed with breast cancer in 2020, resulting in 685,000 deaths.
Regular screening plays a crucial role in identifying early signs of cancer. In Europe, women aged 50 to 69 are advised to undergo a mammogram every two years, with two radiologists analyzing the resulting scan. This approach ensures a thorough examination but can be time-consuming.
In Sweden, a study was conducted involving 80,000 women who had mammograms at four sites in southwest Sweden between April 2021 and July last year. The scans were divided for analysis, with one group assessed by an AI-supported system and the other group examined by two human radiologists as a control.
The AI algorithm read the scans and predicted the risk of cancer out of 10. Its predictions were subsequently verified by a radiologist. The study revealed that the AI-supported system detected 20 percent more cases of breast cancer, equating to an additional case for every thousand women screened.
Both the AI-supported system and the two human radiologists showed an equal rate of 1.5 percent for false positives, where a mammogram initially appears suspicious but is later cleared. However, the workload for radiologists was reduced by 44 percent in the AI group, as only one person was required to analyze the scans rather than the usual two.
Kristina Lang, a radiologist at Lund University in Sweden and the lead author of the study, emphasized the potential of AI in alleviating the excessive workload faced by radiologists. Nevertheless, she emphasized that the promising interim safety results alone were insufficient to confirm the readiness of AI for mammography screening.
The trial requires two more years before it can determine whether the use of AI leads to a decrease in interval cancers, which are detected between routine screenings. The researchers also cautioned about the possibility of over-diagnosing certain forms of early breast cancer called ductal carcinoma in situ.
Stephen Duffy, a professor of cancer screening at Queen Mary University of London, praised the study’s quality and underscored the importance of reducing the burden on radiologists’ time in various breast screening programs.
In conclusion, while AI has shown promise in detecting breast cancer with increased accuracy and reducing radiologist workload, further research is needed before its implementation on a wider scale. The potential benefits of AI technology in medical scans emphasize the importance of ongoing investigations into its effectiveness and safety.
