UniFLM: United Segmentation and Measurement on Fetal Limb Ultrasonic Image
UniFLM is a unified AI framework designed to automatically segment and precisely measure multiple fetal long bones from ultrasound images. It introduces novel modules like SASC for semantic alignment, Positive Sampling to filter noise, and Point Regression Mapping for accurate biometric measurement. This system overcomes data scarcity and improves clinical measurement accuracy.