In Denmark, the use of artificial intelligence (AI) by breast radiologists has significantly enhanced the performance of breast cancer screenings while also reducing the incidence of false-positive findings.
This advancement follows a previous study conducted 16 months ago at the same hospital, which demonstrated that AI could diagnose cancer in chest X-rays as accurately as board-certified radiologists, highlighting the issue of understaffed radiology departments.
When employed to triage likely normal screening results or to assist with decision support, AI can substantially decrease the workload of radiologists, Dr. Andreas Lauritzen, a researcher at Gentofte Hospital in Denmark and the lead author of the study, stated “Population-based screening with mammography reduces breast cancer mortality, but it places a substantial workload on radiologists who must read a large number of mammograms, the majority of which don’t warrant a recall of the patient.”
He further elaborated, “The reading workload is further compounded when screening programs employ double reading to improve cancer detection and decrease false-positive recalls.”
The study found that incorporating AI into the screening process led to the detection of significantly more breast cancers compared to traditional methods (0.82% versus 0.70%) and resulted in a lower false-positive rate (1.63% versus 2.39%).
Dr. Lauritzen explained, “In the AI-screened group, the recall rate decreased by 20.5%, the radiologists’ reading workload was lowered by 33.4%.”
Additionally, the positive predictive value of AI-assisted screening surpassed that of non-AI screening (33.5% versus 22.5%). In the AI group, a higher proportion of invasive cancers detected were 1 centimeter or less in size (44.93% versus 36.60%).
Dr. Lauritzen noted, “All screening performance indicators improved except for the node-negative rate which showed no evidence of change.”
The success of AI in improving screening performance and reducing radiologist workload was not isolated to Denmark. A Swedish study conducted at Lund University also reported a 20% improvement in the accurate diagnosis of breast cancers, with the AI system demonstrating even greater labor-saving capabilities than those observed in Dr. Lauritzen’s study.
The integration of AI in breast cancer screening represents a significant step forward in medical diagnostics. The reduction in false positives is particularly important as it minimizes unnecessary stress and additional procedures for patients, which can often be invasive and costly.
Furthermore, by lowering the workload of radiologists, AI allows these professionals to focus more on cases that require detailed analysis and human judgement, ultimately improving the overall efficiency and effectiveness of cancer screening programs.
An editorial accompaniment to the study praised the ability of AI to reduce human workload. It emphasized that, rather than suggesting AI could replace radiologists, the evidence indicated that AI should be developed as a labor-saving tool. This perspective underscores the complementary role that AI can play in the medical field, enhancing human capabilities rather than replacing them.
The implementation of AI in mammography is a part of a broader trend towards the integration of advanced technologies in healthcare. These technologies have the potential to revolutionize various aspects of medical practice, from diagnostics and treatment planning to patient management and administrative tasks. The key to successful integration lies in ensuring that these technologies are designed to support and augment the skills of healthcare professionals, leading to better patient outcomes and more efficient healthcare delivery systems.
The incorporation of AI into breast cancer screening programs in Denmark has proven to be a significant advancement in the field of radiology. By enhancing screening performance, reducing false positives, and alleviating the workload of radiologists, AI has demonstrated its potential as a valuable tool in the fight against breast cancer.
The positive results from both the Danish and Swedish studies highlight the broader applicability of AI in medical diagnostics and underscore the importance of continued research and development in this area. As AI technology continues to evolve, its role in healthcare is likely to expand, offering new opportunities to improve patient care and streamline medical processes.
Recommended For You
