Deep learning-based segmentation for high-dose-rate brachytherapy in cervical cancer using 3D Prompt-ResUNet.

Journal: Physics in medicine and biology
PMID:

Abstract

To develop and evaluate a 3D Prompt-ResUNet module that utilized the prompt-based model combined with 3D nnUNet for rapid and consistent autosegmentation of high-risk clinical target volume (HRCTV) and organ at risk (OAR) in high-dose-rate brachytherapy for cervical cancer patients.We used 73 computed tomography scans and 62 magnetic resonance imaging scans from 135 (103 for training, 16 for validation, and 16 for testing) cervical cancer patients across two hospitals for HRCTV and OAR segmentation. A novel comparison of the deep learning neural networks 3D Prompt-ResUNet, nnUNet, and segment anything model-Med3D was applied for the segmentation. Evaluation was conducted in two parts: geometric and clinical assessments. Quantitative metrics included the Dice similarity coefficient (DSC), 95th percentile Hausdorff distance (HD95%), Jaccard index (JI), and Matthews correlation coefficient (MCC). Clinical evaluation involved interobserver comparison, 4-grade expert scoring, and a double-blinded Turing test.The Prompt-ResUNet model performed most similarly to experienced radiation oncologists, outperforming less experienced ones. During testing, the DSC, HD95% (mm), JI, and MCC value (mean ± SD) for HRCTV were 0.92 ± 0.03, 2.91 ± 0.69, 0.85 ± 0.04, and 0.92 ± 0.02, respectively. For the bladder, these values were 0.93 ± 0.05, 3.07 ± 1.05, 0.87 ± 0.08, and 0.93 ± 0.05, respectively. For the rectum, they were 0.87 ± 0.03, 3.54 ± 1.46, 0.78 ± 0.05, and 0.87 ± 0.03, respectively. For the sigmoid, they were 0.76 ± 0.11, 7.54 ± 5.54, 0.63 ± 0.14, and 0.78 ± 0.09, respectively. The Prompt-ResUNet achieved a clinical viability score of at least 2 in all evaluation cases (100%) for both HRCTV and bladder and exceeded the 30% positive rate benchmark for all evaluated structures in the Turing test.The Prompt-ResUNet architecture demonstrated high consistency with ground truth in autosegmentation of HRCTV and OARs, reducing interobserver variability and shortening treatment times.

Authors

  • Xian Xue
    Secondary Standard Dosimetry Laboratory, National Institute for Radiological Protection, Chinese Center for Disease Control and Prevention (CDC), Beijing, China.
  • Lining Sun
    School of Mechanical and Electronic Engineering, Soochow University, Suzhou, Jiangsu, China.
  • Dazhu Liang
    Digital Health China Technologies Co., LTD, Beijing, China.
  • Jingyang Zhu
    Department of radiation oncology, Zhongcheng Cancer center, Beijing 100160, People's Republic of China.
  • Lele Liu
    College of Science, University of Shanghai for Science and Technology, Shanghai, 200093, China. ahhylau@outlook.com.
  • Quanfu Sun
    Secondary Standard Dosimetry Laboratory, National Institute for Radiological Protection, Chinese Center for Disease Control and Prevention (CDC), Beijing, China.
  • Hefeng Liu
    Digital Health China Technologies Co., LTD, Beijing, China.
  • Jianwei Gao
    Digital Health China Technologies Co., LTD, Beijing, China.
  • Xiaosha Fu
    Biomedical Research Centre, Sheffield Hallam University, Sheffield S11WB, United Kingdom.
  • Jingjing Ding
    Department of General Practice, Tongren Hospital, Shanghai Jiao Tong University, School of Medicine, Shanghai, China.
  • Xiangkun Dai
    Department of Radiotherapy, First Medical Center of PLA General Hospital, BeiJing 100853, P.R.China.
  • Laiyuan Tao
    Digital Health China Technologies Co., LTD, Beijing, China.
  • Jinsheng Cheng
    Secondary Standard Dosimetry Laboratory, National Institute for Radiological Protection, Chinese Center for Disease Control and Prevention (CDC), Beijing, China.
  • Tengxiang Li
    Department of Nuclear Science and Engineering, Nanhua University, Hunan 421001, People's Republic of China.
  • Fugen Zhou