Enhanced Uncertainty Estimation in Ultrasound Image Segmentation with MSU-Net
This work addresses the risk of segmentation errors in medical procedures for trauma and critical care patients, enabling safer autonomous systems for non-experts in resource-limited settings, though it is incremental as it builds on existing U-Net methods.
The paper tackles the problem of inaccurate vessel segmentation in ultrasound images for autonomous robotic needle insertion, introducing MSU-Net to improve uncertainty estimation and achieving an 18.1% enhancement over a baseline model.
Efficient intravascular access in trauma and critical care significantly impacts patient outcomes. However, the availability of skilled medical personnel in austere environments is often limited. Autonomous robotic ultrasound systems can aid in needle insertion for medication delivery and support non-experts in such tasks. Despite advances in autonomous needle insertion, inaccuracies in vessel segmentation predictions pose risks. Understanding the uncertainty of predictive models in ultrasound imaging is crucial for assessing their reliability. We introduce MSU-Net, a novel multistage approach for training an ensemble of U-Nets to yield accurate ultrasound image segmentation maps. We demonstrate substantial improvements, 18.1% over a single Monte Carlo U-Net, enhancing uncertainty evaluations, model transparency, and trustworthiness. By highlighting areas of model certainty, MSU-Net can guide safe needle insertions, empowering non-experts to accomplish such tasks.