1. a. Department of Medical Equipment|b. Department of Tumor Radiotherapy, Nanjing Hospital Affiliated to Nanjing Medical University (Nanjing First Hospital), Nanjing Jiangsu 210006, China|2. Department of Medical Physics, Nanjing Medical University,
Nanjing Jiangsu 210029, China|3. Nanjing Emergency Medical Center, Nanjing Jiangsu 210003, China
Abstract：Objective To evaluate the feasibility and accuracy of DeepViewer software in automatically delineating gross target volume after breast-conserving surgery for breast cancer patients. Methods Twenty patients with breast cancer after breastconserving
were selected for study, including 12 cases of left breast cancer and 8 cases of right breast cancer. The localized CT images of 20 patients were transferred from the Eclipse treatment planning system to the DeepViewer software, and the target areas
were automatically delineated by using DeepViewer software. Then transmitted the sketched structure files back to the Eclipse treatment planning system. The difference of volume, shape consistency, and sketching time of the target areas between automatically
delineating and manually delineating were compared, and statistical analysis was performed to evaluate the feasibility and accuracy of DeepViewer software in sketching the radiotherapy target area. Results Statistical analysis was performed on the data of 20
breast cancer patients. The percentage volume difference (△V%) of the heart structure between the automatic and manual sketching methods was (-21.43±15.29 )%, the similarity index (DSC) value was 0.68±0.11. The total position difference (ΔL) was
(1.76 ± 1.23) cm. The results of independent sample t test showed that there was no statistical difference between the two side groups of data (P>0.05). Conclusion Using DeepViewer software to sketch the target area after breast-conserving, the result is
basically satisfactory, which needs further optimization and improvement.
田书畅1a，时飞跃1b,2，赵紫婷1a，王敏1b，秦伟1b，吴倩倩1a，秦航1a，蒋红兵1a,3. DeepViewer软件自动勾画乳腺癌保乳术后肿瘤靶区的初步研究[J]. 中国医疗设备, 2020, 35(12): 29-32.
TIAN Shuchang1a，SHI Feiyue1b,2，ZHAO Ziting1a，WANG Min1b，QIN Wei1b，WU Qianqian1a，QIN Hang1a，JIANG Hongbing1a,3. The Initial Study on Automatic Delineation of Gross Target Volume after Breast-Conserving Surgery by DeepViewer Software. China Medical Devices, 2020, 35(12): 29-32.
