Automated ultrasound doppler angle estimation using deep learning
Researchers have developed a deep learning approach to automatically estimate the angle in Doppler ultrasound, a critical step for accurate blood velocity measurements. The method, tested on human carotid ultrasound images, achieved a mean absolute error between 3.9° and 9.4°, with the best model performing within acceptable clinical error thresholds. This technique has the potential to be integrated into commercial ultrasound scanners, improving diagnostic accuracy. AI