This project focuses on developing an image quality scoring model for robotic ultrasound.
An existing dataset of 15 participants with corresponding segmentation results will be used to define quality-scoring criteria and train a baseline model. Data from approximately 3-6 additional participants will be collected to serve as an independent test set, used to evaluate the model's generalisation across participants and imaging conditions.
The final goal is to support more intelligent robotic ultrasound scanning through automatic image quality assessment.
