Deep Learning Improved Clinical Target Volume Contouring Quality and Efficiency for Postoperative Radiation Therapy in Non-small Cell Lung Cancer.

Nan Bi,Kunpeng Xu,Jianrong Dai,Linfang Wu,Junjie Miao,Xinyuan Chen,Wenlong Xia,Zongmei Zhou,Luhua Wang,Tao Zhang,Jingbo Wang,Yexiong Li,Quanrong Fan

doi:10.3389/fonc.2019.01192

Abstract

Purpose: To investigate whether a deep learning-assisted contour (DLAC) could provide greater accuracy, inter-observer consistency, and efficiency compared with a manual contour (MC) of the clinical target volume (CTV) for non-small cell lung cancer (NSCLC) receiving postoperative radiotherapy (PORT).Materials and Methods: A deep dilated residual network was used to achieve the effective automatic contour of the CTV. Eleven junior physicians contoured CTVs on 19 patients by using both MC and DLAC methods independently. Compared with the ground truth, the accuracy of the contour was evaluated by using the Dice coefficient and mean distance to agreement (MDTA). The coefficient of variation (CV) and standard distance deviation (SDD) were rendered to measure the inter-observer variability or consistency. The time consumed for each of the two contouring methods was also compared.Results: A total of 418 CTV sets were generated. DLAC improved contour accuracy when compared with MC and was associated with a larger Dice coefficient (mean ± SD: 0.75 ± 0.06 vs. 0.72 ± 0.07, p < 0.001) and smaller MDTA (mean ± SD: 2.97 ± 0.91 mm vs. 3.07 ± 0.98 mm, p < 0.001). The DLAC was also associated with decreased inter-observer variability, with a smaller CV (mean ± SD: 0.129 ± 0.040 vs. 0.183 ± 0.043, p < 0.001) and SDD (mean ± SD: 0.47 ± 0.22 mm vs. 0.72 ± 0.41 mm, p < 0.001). In addition, a value of 35% of time saving was provided by the DLAC (median: 14.81 min vs. 9.59 min, p < 0.001).Conclusions: Compared with MC, the DLAC is a promising strategy to obtain superior accuracy, consistency, and efficiency for the PORT-CTV in NSCLC.

Highlights

Accurate delineation of the clinical target volume (CTV) is one of the most crucial aspects of treatment planning in radiotherapy
deep learning-assisted contour (DLAC) improved contour accuracy when compared with manual contour (MC) and was associated with a larger Dice coefficient and smaller mean distance to agreement (MDTA)
The ground truth (GT) contours were generated based on the contemporary re-delineation by the three senior radiation oncologists as per the protocol of a randomized phase III trial of a postoperative radiotherapy (PORT) study primarily investigated by our institute, rather than by using the previous contours administered in clinical practice

Summary

Introduction

Accurate delineation of the clinical target volume (CTV) is one of the most crucial aspects of treatment planning in radiotherapy The quality of this process depends on the expertise of the individual observer and is associated with quality outcomes in lung cancer care [1,2,3,4]. As a state of the art technique, deep learning has manifested superior performances by enabling self-taught discovery of informative representations and using hierarchical layers of learned abstraction to accomplish high-level tasks efficiently [10]. Another important advantage of deep learning methods is that they are suitable for various patient body sizes and shapes. Even if input images show huge differences in body size or shape, deep learning can increase knowledge from a sufficient number of examples and produce excellent segmentation results [10, 11]

Methods

Results

Discussion

Conclusion