Real-Time Multi-Guidewire Endpoint Localization in Fluoroscopy Images

• Published in: IEEE transactions on medical imaging Vol. 40; no. 8; pp. 2002 - 2014
• Main Authors: Li, Rui-Qi, Xie, Xiao-Liang, Zhou, Xiao-Hu, Liu, Shi-Qi, Ni, Zhen-Liang, Zhou, Yan-Jie, Bian, Gui-Bin, Hou, Zeng-Guang
• Format: Journal Article
• Language: English
• Published: United States IEEE 01.08.2021; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Algorithms; Bus interconnections; Datasets; Detectors; Fluoroscopy; Guidewire endpoint; Image segmentation; Instruments; keypoint localization; Localization; Location awareness; Medical instruments; Microsurgery; Pixels; Post-production processing; Real time; Real-time systems; Retina; Surgery; surgical instrument; Surgical instruments; Task analysis
• ISSN: 0278-0062; 1558-254X
• DOI: 10.1109/TMI.2021.3069998

The real-time localization of the guidewire endpoints is a stepping stone to computer-assisted percutaneous coronary intervention (PCI). However, methods for multi-guidewire endpoint localization in fluoroscopy images are still scarce. In this paper, we introduce a framework for real-time multi-guidewire endpoint localization in fluoroscopy images. The framework consists of two stages, first detecting all guidewire instances in the fluoroscopy image, and then locating the endpoints of each single guidewire instance. In the first stage, a YOLOv3 detector is used for guidewire detection, and a post-processing algorithm is proposed to refine the guidewire detection results. In the second stage, a Segmentation Attention-hourglass (SA-hourglass) network is proposed to predict the endpoint locations of each single guidewire instance. The SA-hourglass network can be generalized to the keypoint localization of other surgical instruments. In our experiments, the SA-hourglass network is applied not only on a guidewire dataset but also on a retinal microsurgery dataset, reaching the mean pixel error (MPE) of 2.20 pixels on the guidewire dataset and the MPE of 5.30 pixels on the retinal microsurgery dataset, both achieving the state-of-the-art localization results. Besides, the inference rate of our framework is at least 20FPS, which meets the real-time requirement of fluoroscopy images (6-12FPS).