Comparative analysis of automatic segmentation of esophageal cancer using 3D Res-UNet on conventional and 40-keV virtual mono-energetic CT Images: a retrospective study

OBJECTIVES: To assess the performance of 3D Res-UNet for fully automated segmentation of esophageal cancer (EC) and compare the segmentation accuracy between conventional images (CI) and 40-keV virtual mono-energetic images (VMI(40 kev)). METHODS: Patients underwent spectral CT scanning and diagnosed of EC by operation or gastroscope biopsy in our hospital from 2019 to 2020 were analyzed retrospectively. All artery spectral base images were transferred to the dedicated workstation to generate VMI(40 kev) and CI. The segmentation model of EC was constructed by 3D Res-UNet neural network in VMI(40 kev) and CI, respectively. After optimization training, the Dice similarity coefficient (DSC), overlap (IOU), average symmetrical surface distance (ASSD) and 95% Hausdorff distance (HD_95) of EC at pixel level were tested and calculated in the test set. The paired rank sum test was used to compare the results of VMI(40 kev) and CI. RESULTS: A total of 160 patients were included in the analysis and randomly divided into the training dataset (104 patients), validation dataset (26 patients) and test dataset (30 patients). VMI(40 kev)as input data in the training dataset resulted in higher model performance in the test dataset in comparison with using CI as input data (DSC:0.875 vs 0.859, IOU: 0.777 vs 0.755, ASSD:0.911 vs 0.981, HD_95: 4.41 vs 6.23, all p-value <0.05). CONCLUSION: Fully automated segmentation of EC with 3D Res-UNet has high accuracy and clinically feasibility for both CI and VMI(40 kev). Compared with CI, VMI(40 kev) indicated slightly higher accuracy in this test dataset.