Deep learning-based accurate delineation of primary gross tumor volume of nasopharyngeal carcinoma on heterogeneous magnetic resonance imaging: A large-scale and multi-center study. (March 2023)

• Record Type: Journal Article
• Authors: Luo, Xiangde; Liao, Wenjun; He, Yuan; Tang, Fan; Wu, Mengwan; Shen, Yuanyuan; Huang, Hui; Song, Tao; Li, Kang; Zhang, Shichuan; Zhang, Shaoting; Wang, Guotai
• Is Part Of: Radiotherapy and oncology. Volume 180(2023)
• Publication Date: 2023-03
• Subjects: Nasopharyngeal carcinoma -- Automatic delineation -- Primary gross tumor volume -- Generalization -- Deep learning; Oncology -- Periodicals; Radiotherapy -- Periodicals; Tumors -- Periodicals; Medical Oncology -- Periodicals; Neoplasms -- radiotherapy -- Periodicals; Radiotherapy -- Periodicals; Radiothérapie -- Périodiques; Cancérologie -- Périodiques; Tumeurs -- Périodiques; Electronic journals; 616.9940642
• DOI: 10.1016/j.radonc.2023.109480
• Languages: English
• ISSNs: 0167-8140

Abstract

Highlights: An accurate and robust solution for the delineation of primary gross tumor volume (GTVp) of nasopharyngeal carcinoma (NPC) on heterogeneous magnetic resonance imaging (MRI). A new augmentation-invariant method can be a potential solution to boost the model's generalization ability and produce more reliable delineation results. The proposed method was evaluated on large-scale and multi-center datasets of images from 1057 patients collected from five hospitals. The results show that the proposed method can achieve reliable delineation of GTVp of NPC on heterogeneous MRI. Abstract: Background and purpose: The problem of obtaining accurate primary gross tumor volume (GTVp) segmentation for nasopharyngeal carcinoma (NPC) on heterogeneous magnetic resonance imaging (MRI) images with deep learning remains unsolved. Herein, we reported a new deep-learning method than can accurately delineate GTVp for NPC on multi-center MRI scans. Material and methods: We collected 1057 patients with MRI images from five hospitals and randomly selected 600 patients from three hospitals to constitute a mixed training cohort for model development. The resting patients were used as internal (n = 259) and external (n = 198) testing cohorts for model evaluation. An augmentation-invariant strategy was proposed to delineate GTVp from multi-center MRI images, which encouraged networks to produce similar predictions for inputs with different augmentations to learn invariant anatomical structure features. The Dice similarity coefficient (DSC), 95 % Hausdorff distance (HD95), average surface distance (ASD), and relative absolute volume difference (RAVD) were used to measure segmentation performance. Results: The model-generated predictions had a high overlap ratio with the ground truth. For the internal testing cohorts, the average DSC, HD95, ASD, and RAVD were 0.88, 4.99 mm, 1.03 mm, and 0.13, respectively. For external testing cohorts, the average DSC, HD95, ASD, and RAVD were 0.88, 3.97 mm, 0.97 mm, and 0.10, respectively. No significant differences were found in DSC, HD95, and ASD for patients with different T categories, MRI thickness, or in-plane spacings. Moreover, the proposed augmentation-invariant strategy outperformed the widely-used nnUNet, which uses conventional data augmentation approaches. Conclusion: Our proposed method showed a highly accurate GTVp segmentation for NPC on multi-center MRI images, suggesting that it has the potential to act as a generalized delineation solution for heterogeneous MRI images.