Agent with Warm Start and Adaptive Dynamic Termination for Plane Localization in 3D Ultrasound
| Autor: | Dong Ni, Tianfu Wang, Ruobing Huang, Jikuan Qian, Huizhi Guo, Yi Xiong, Yuhao Huang, Xin Yang, Huanjia Luo, Haoran Dou, Wufeng Xue, Yuanji Zhang |
|---|---|
| Rok vydání: | 2021 |
| Předmět: |
FOS: Computer and information sciences
Computer Vision and Pattern Recognition (cs.CV) Computer Science - Computer Vision and Pattern Recognition Ultrasonography Prenatal Combinatorics Prenatal ultrasound Warm start Imaging Three-Dimensional Pregnancy Abdomen medicine FOS: Electrical engineering electronic engineering information engineering Humans 3D ultrasound Electrical and Electronic Engineering Ultrasonography Physics One shot Radiological and Ultrasound Technology medicine.diagnostic_test Plane (geometry) Image and Video Processing (eess.IV) Electrical Engineering and Systems Science - Image and Video Processing Computer Science Applications Transverse plane Clinical diagnosis Female Software Algorithms Fetal abdomen |
| ISSN: | 0278-0062 |
| DOI: | 10.48550/arxiv.2103.14502 |
| Popis: | Accurate standard plane (SP) localization is the fundamental step for prenatal ultrasound (US) diagnosis. Typically, dozens of US SPs are collected to determine the clinical diagnosis. 2D US has to perform scanning for each SP, which is time-consuming and operator-dependent. While 3D US containing multiple SPs in one shot has the inherent advantages of less user-dependency and more efficiency. Automatically locating SP in 3D US is very challenging due to the huge search space and large fetal posture variations. Our previous study proposed a deep reinforcement learning (RL) framework with an alignment module and active termination to localize SPs in 3D US automatically. However, termination of agent search in RL is important and affects the practical deployment. In this study, we enhance our previous RL framework with a newly designed adaptive dynamic termination to enable an early stop for the agent searching, saving at most 67% inference time, thus boosting the accuracy and efficiency of the RL framework at the same time. Besides, we validate the effectiveness and generalizability of our algorithm extensively on our in-house multi-organ datasets containing 433 fetal brain volumes, 519 fetal abdomen volumes, and 683 uterus volumes. Our approach achieves localization error of 2.52mm/10.26 degrees, 2.48mm/10.39 degrees, 2.02mm/10.48 degrees, 2.00mm/14.57 degrees, 2.61mm/9.71 degrees, 3.09mm/9.58 degrees, 1.49mm/7.54 degrees for the transcerebellar, transventricular, transthalamic planes in fetal brain, abdominal plane in fetal abdomen, and mid-sagittal, transverse and coronal planes in uterus, respectively. Experimental results show that our method is general and has the potential to improve the efficiency and standardization of US scanning. Comment: Accepted by IEEE Transactions on Medical Imaging (12 pages, 8 figures, 11 tabels) |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|