A spatio-temporal graph convolutional network for ultrasound echocardiographic landmark detection

• Published in: Medical image analysis 2024-10, Vol.97, p.103272, Article 103272
• Main Authors: Li, Honghe, Yang, Jinzhu, Xuan, Zhanfeng, Qu, Mingjun, Wang, Yonghuai, Feng, Chaolu
• Format: Article
• Language: English
• Subjects: Algorithms; Anatomic Landmarks; Deep learning; Echocardiography; Echocardiography - methods; Heart Ventricles - diagnostic imaging; Humans; Image Processing, Computer-Assisted - methods; Landmark detection; Neural Networks, Computer; Spatio-Temporal Analysis; Spatio-temporal dependencies
• ISSN: 1361-8415; 1361-8423; 1361-8423
• DOI: 10.1016/j.media.2024.103272

Landmark detection is a crucial task in medical image analysis, with applications across various fields. However, current methods struggle to accurately locate landmarks in medical images with blurred tissue boundaries due to low image quality. In particular, in echocardiography, sparse annotations make it challenging to predict landmarks with position stability and temporal consistency. In this paper, we propose a spatio-temporal graph convolutional network tailored for echocardiography landmark detection. We specifically sample landmark labels from the left ventricular endocardium and pre-calculate their correlations to establish structural priors. Our approach involves a graph convolutional neural network that learns the interrelationships among landmarks, significantly enhancing landmark accuracy within ambiguous tissue contexts. Additionally, we integrate gate recurrent units to grasp the temporal consistency of landmarks across consecutive images, augmenting the model’s resilience against unlabeled data. Through validation across three echocardiography datasets, our method demonstrates superior accuracy when contrasted with alternative landmark detection models. •Introducing spatial dependencies ensures the stability of landmark structures.•Introducing temporal dependencies maintains continuous temporal changes.•Varying landmark numbers and adjacent nodes affect prediction accuracy.•The proposed method is effective across multiple backbones.•Approximating target segmentation using landmarks achieves high precision.