A recent study evaluated the accuracy of breast ultrasound image analysis software in comparison to expert sonographers. Breast ultrasound is crucial for early breast cancer detection, especially in countries like China where breast cancer is prevalent. The study aimed to assess the agreement between the AI software and sonographers in analyzing features of breast nodules.
The research analyzed 493 breast ultrasound images from two hospitals in Shanghai. The software and sonographers assessed features such as echotexture, orientation, shape, margin, calcification, and posterior echo attenuation. The study found a high level of agreement between the software and sonographers in assessing these features, emphasizing the software’s accuracy.
While the proportion of agreement was over 80% for most features, there were some areas of moderate agreement, particularly in echo pattern, orientation, and posterior echo attenuation. The software significantly reduced analysis time compared to sonographers, highlighting its potential to enhance efficiency in healthcare settings.
AI technology has increasingly become a focal point in medical imaging research, offering promising solutions to reduce sonographers’ workload and improve healthcare delivery, especially in underserved areas with limited access to skilled professionals. The study suggests that AI software can play a vital role in automating feature analysis in breast ultrasound images, ultimately improving diagnostic accuracy and efficiency in breast cancer screening.
