UltraDP: Generalizable Carotid Ultrasound Scanning with Force-Aware Diffusion Policy

2025-11-19

AI Summary

Current robotic carotid ultrasound scanning suffers from poor generalizability and low data efficiency due to substantial anatomical variability among patients and complex human–robot interaction. To address these challenges, this work proposes an autonomous scanning method leveraging multimodal perception and diffusion-based policy learning. We design a dedicated guidance module that fuses real-time ultrasound images, wrist-mounted visual feedback, contact torque, and probe pose. A diffusion policy network models the high-dimensional action distribution, while a hybrid force–impedance controller ensures safe, stable, and compliant contact tracking. We construct a large-scale carotid scanning dataset comprising 460,000 sample pairs from 21 volunteers. Our approach achieves a 95% success rate in transverse-plane scanning on unseen subjects, significantly improving system robustness and data utilization efficiency.

Technology Category

• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Computer Vision: Diffusion Models for Vision
• Natural Language Processing: Safety and Robustness

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Security and Privacy: Data transparency and provenance

Abstract

Ultrasound scanning is a critical imaging technique for real-time, non-invasive diagnostics. However, variations in patient anatomy and complex human-in-the-loop interactions pose significant challenges for autonomous robotic scanning. Existing ultrasound scanning robots are commonly limited to relatively low generalization and inefficient data utilization. To overcome these limitations, we present UltraDP, a Diffusion-Policy-based method that receives multi-sensory inputs (ultrasound images, wrist camera images, contact wrench, and probe pose) and generates actions that are fit for multi-modal action distributions in autonomous ultrasound scanning of carotid artery. We propose a specialized guidance module to enable the policy to output actions that center the artery in ultrasound images. To ensure stable contact and safe interaction between the robot and the human subject, a hybrid force-impedance controller is utilized to drive the robot to track such trajectories. Also, we have built a large-scale training dataset for carotid scanning comprising 210 scans with 460k sample pairs from 21 volunteers of both genders. By exploring our guidance module and DP's strong generalization ability, UltraDP achieves a 95% success rate in transverse scanning on previously unseen subjects, demonstrating its effectiveness.

Problem

• Addresses autonomous ultrasound scanning challenges from anatomical variations
• Overcomes limitations in generalization and data utilization for robotic scanning
• Ensures stable contact and safe human-robot interaction during carotid scanning

Innovation

• Diffusion-Policy method for multi-sensory autonomous scanning
• Specialized guidance module centers artery in ultrasound images
• Hybrid force-impedance controller ensures stable contact safety