Anatomical Prior-Driven Framework for Autonomous Robotic Cardiac Ultrasound Standard View Acquisition

2026-03-22

AI Summary

This study addresses the challenges in automated acquisition of standard echocardiographic views, which heavily relies on operator expertise and suffers from anatomically inconsistent segmentations in low-texture images, as well as heuristic- or black-box-based probe control strategies. To overcome these limitations, the authors propose an end-to-end framework integrating anatomical priors: a spatial relationship graph (SRG) module is embedded into the YOLOv11s segmentation network to enforce anatomical consistency, and—for the first time—a Gaussian prior over quantifiable anatomical features is incorporated into the state representation and reward function of a reinforcement learning agent for autonomous probe control. Experimental results demonstrate that SRG-YOLOv11s achieves an 11.3% improvement in mAP50 and a 6.8% gain in mIoU on the Special Case dataset, while the reinforcement learning agent attains standard view acquisition success rates of 92.5% in simulation and 86.7% in phantom experiments.

Technology Category

• Computer Vision: Segmentation
• Intelligent Robots: State Estimation
• Search and Optimization: Learning to Search

Application Category

• Search and Retrieval-Augmented AI: Agentic search
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Cardiac ultrasound diagnosis is critical for cardiovascular disease assessment, but acquiring standard views remains highly operator-dependent. Existing medical segmentation models often yield anatomically inconsistent results in images with poor textural differentiation between distinct feature classes, while autonomous probe adjustment methods either rely on simplistic heuristic rules or black-box learning. To address these issues, our study proposed an anatomical prior (AP)-driven framework integrating cardiac structure segmentation and autonomous probe adjustment for standard view acquisition. A YOLO-based multi-class segmentation model augmented by a spatial-relation graph (SRG) module is designed to embed AP into the feature pyramid. Quantifiable anatomical features of standard views are extracted. Their priors are fitted to Gaussian distributions to construct probabilistic APs. The probe adjustment process of robotic ultrasound scanning is formalized as a reinforcement learning (RL) problem, with the RL state built from real-time anatomical features and the reward reflecting the AP matching. Experiments validate the efficacy of the framework. The SRG-YOLOv11s improves mAP50 by 11.3% and mIoU by 6.8% on the Special Case dataset, while the RL agent achieves a 92.5% success rate in simulation and 86.7% in phantom experiments.

Problem

• cardiac ultrasound
• standard view acquisition
• anatomical inconsistency
• autonomous probe adjustment
• operator dependency

Innovation

• anatomical prior
• spatial-relation graph
• reinforcement learning
• autonomous robotic ultrasound
• standard view acquisition