Evaluation of Deep Learning-Based Auto-Segmentation of Target Volume and Organs-at-Risk in Breast Cancer Patients

2020 ◽  
Vol 108 (3) ◽  
pp. e779
Author(s):  
S.Y. Chung ◽  
J.S. Chang ◽  
Y. Chang ◽  
B.S. Choi ◽  
J. Chun ◽  
...  
Keyword(s):  
Breast Cancer ◽  
At Risk ◽  
Deep Learning ◽  
Cancer Patients ◽  
Organs At Risk ◽  
Target Volume ◽  
Breast Cancer Patients

scholarly journals Clinical feasibility of deep learning-based auto-segmentation of target volumes and organs-at-risk in breast cancer patients after breast-conserving surgery

2021 ◽  
Vol 16 (1) ◽  
Author(s):  
Seung Yeun Chung ◽  
Jee Suk Chang ◽  
Min Seo Choi ◽  
Yongjin Chang ◽  
Byong Su Choi ◽  
...  
Keyword(s):  
Breast Cancer ◽  
At Risk ◽  
Deep Learning ◽  
Cancer Patients ◽  
Organs At Risk ◽  
Breast Conserving Surgery ◽  
Dice Similarity Coefficient ◽  
Breast Cancer Patients ◽  
Dose Volume Histograms ◽  
Radiation Oncologists

Abstract Background In breast cancer patients receiving radiotherapy (RT), accurate target delineation and reduction of radiation doses to the nearby normal organs is important. However, manual clinical target volume (CTV) and organs-at-risk (OARs) segmentation for treatment planning increases physicians’ workload and inter-physician variability considerably. In this study, we evaluated the potential benefits of deep learning-based auto-segmented contours by comparing them to manually delineated contours for breast cancer patients. Methods CTVs for bilateral breasts, regional lymph nodes, and OARs (including the heart, lungs, esophagus, spinal cord, and thyroid) were manually delineated on planning computed tomography scans of 111 breast cancer patients who received breast-conserving surgery. Subsequently, a two-stage convolutional neural network algorithm was used. Quantitative metrics, including the Dice similarity coefficient (DSC) and 95% Hausdorff distance, and qualitative scoring by two panels from 10 institutions were used for analysis. Inter-observer variability and delineation time were assessed; furthermore, dose-volume histograms and dosimetric parameters were also analyzed using another set of patient data. Results The correlation between the auto-segmented and manual contours was acceptable for OARs, with a mean DSC higher than 0.80 for all OARs. In addition, the CTVs showed favorable results, with mean DSCs higher than 0.70 for all breast and regional lymph node CTVs. Furthermore, qualitative subjective scoring showed that the results were acceptable for all CTVs and OARs, with a median score of at least 8 (possible range: 0–10) for (1) the differences between manual and auto-segmented contours and (2) the extent to which auto-segmentation would assist physicians in clinical practice. The differences in dosimetric parameters between the auto-segmented and manual contours were minimal. Conclusions The feasibility of deep learning-based auto-segmentation in breast RT planning was demonstrated. Although deep learning-based auto-segmentation cannot be a substitute for radiation oncologists, it is a useful tool with excellent potential in assisting radiation oncologists in the future. Trial registration Retrospectively registered.

scholarly journals OC-0164: Assessment of respiratory induced motion of target area and organs at risk in radiotherapy for breast cancer patients

2013 ◽  
Vol 106 ◽  
pp. S63
Author(s):  
A. Betgen ◽  
T. Alderliesten ◽  
P.H. Elkhuizen ◽  
C. van Vliet-Vroegindeweij ◽  
P. Remeijer
Keyword(s):  
Breast Cancer ◽  
At Risk ◽  
Cancer Patients ◽  
Organs At Risk ◽  
Target Area ◽  
Breast Cancer Patients ◽  
Induced Motion

The radiation doses to the heart and the left anterior descending coronary artery for various modes of radiation treatment of the breast cancer patients.

2017 ◽  
Vol 35 (15_suppl) ◽  
pp. e12092-e12092
Author(s):  
Elizaveta Maslyukova ◽  
Luiza Korytova ◽  
Anna Bondarenko ◽  
Razifa Zhabina ◽  
Oleg Korytov ◽  
...  
Keyword(s):  
Breast Cancer ◽  
At Risk ◽  
Cancer Patients ◽  
Supine Position ◽  
Radiation Treatment ◽  
Organs At Risk ◽  
Free Breathing ◽  
Breast Cancer Patients ◽  
Heart Volume ◽  
Treatment Plans

e12092 Background: The comparison of the radiation load to the organs at risk for three modes of radiation treatment of the breast cancer patients. Methods: The research includes the dosimetric radiation treatment plans for the 20 breast cancer patients with the left-side localization. They all underwent a computed tomography (CT) scan in standard supine position in free-breathing (FB), supine position with Active Breathing Control (ABC) device in deep inspiratory breath hold, and prone position in free-breathing (PP). Three-dimensional treatment plans were made for all 3 CTs. The dose valuations for 3D-planning were carried out for three CT- series. For each mode of radiation, the doze-volume parameters of organs at risk were estimated. Results: For all cases the contoured heart volume varied from 477 см3 - 1056 см3, with medium volume 769 см3. The best marks such as V25heart, medium doses to the heart and LAD, were achieved with on ABC methods (V25heart 4,26%, Dmean heart 3,13Gy, DmeanLAD 13,8Gy) in comparison FB (V25heart 9,49%, Dmean heart 4,97Gy, DmeanLAD 19,55Gy) and PP (V25heart 12,8%, Dmean heart 9,06Gy, DmeanLAD 24,18 Gy) (V25heart P = 0.00153, Dmean heart: P =0,000; Dmean LAD: P = 0.00088), when both the breast and the axillary nodes were included in the volume. The advantage of the dosimetric indexes for FB and ABC did not change while axillary and supraclavicular nodes were added to the radiation volume ABC (V25heart 3,49%, Dmean heart 3,08Gy, DmeanLAD 13,88Gy) in comparison with FB methods (V25heart 7,91%, Dmean heart 4,99Gy, DmeanLAD 19,89Gy) (V25heart P = 0.00205, Dmean heart: P =0,004; Dmean LAD: P = 0.03). Conclusions: Radiation treatment in the position on the back with controlled delay of breath on inspiration height contributed to the statistically significant reduction of the heart volume exposed to more than 25 Gy (V25heart), mean dose to the heart and mean dose to LAD.

scholarly journals Helical tomotherapy with a complete-directional-complete block technique effectively reduces cardiac and lung dose for left-sided breast cancer

2020 ◽  
Vol 93 (1108) ◽  
pp. 20190792 ◽  
Author(s):  
Hsin-Pei Yeh ◽  
Yu-Chuen Huang ◽  
Li-Ying Wang ◽  
Pei-Wei Shueng ◽  
Hui-Ju Tien ◽  
...  
Keyword(s):  
Breast Cancer ◽  
Cancer Patients ◽  
Helical Tomotherapy ◽  
Organs At Risk ◽  
Intensity Modulated Radiation Therapy ◽  
Target Volume ◽  
Conformity Index ◽  
Breast Cancer Patients ◽  
Lung Dose ◽  
The Mean

Objectives: To evaluate the feasibility and optimal restricted angle of the complete-directional-complete block (CDCB) technique in helical tomotherapy (HT) by including regional nodal irradiation (RNI) with the internal mammary node (IMN) in left-sided breast cancer. Methods: Ten left-sided breast cancer patients treated with 50 Gy in 25 fractions were compared with five-field intensity-modulated radiation therapy (5F-IMRT) and six types of HT plans. In the HT plans, complete block (CB), organ-based directional block (OBDB) and CDCB with different restricted angles were used. Results: The conformity index (CI) between the CDCB0,10,15,20 and 5F-IMRT groups was similar. Compared to CB, OBDB and 5F-IMRT, CDCB20 resulted in a decreased ipsilateral mean lung dose. The low-dose region (V5) of the ipsilateral lung in OBDB (84.0%) was the highest among all techniques (p < 0.001). The mean dose of the heart in CB was significantly reduced (by 11.5–22.4%) compared with other techniques. The V30 of the heart in CDCB20 (1.9%) was significantly lower than that of CB, OBDB and 5F-IMRT. Compared to the mean dose of the left anterior descending (LAD) artery of 5F-IMRT (27.0 Gy), CDCB0, CDCB10, CDCB15, CDCB20 and OBDB reduced the mean dose effectively by 31.7%, 38.3%, 39.6%, 42.0 and 56.2%, respectively. Considering the parameters of the organs-at-risk (OARs), CDCB10,15,20 had higher expectative values than the other techniques (p = 0.01). Conclusions: HT with the CDCB technique is feasible for treating left-sided breast cancer patients. The CDCB10-20 techniques not only achieved similar planning target volume coverage, homogeneity and dose conformity but also allowed better sparing of the heart and bilateral lungs. Advances in knowledge: For left-sided breast cancer patients whose RNI field includes the IMN, heart avoidance is an important issue. The CDCB technique achieved good PTV coverage, homogeneity and dose conformity and allowed better sparing of the mean dose of the lung, the LAD artery, and the heart and reduced the V30 of the heart.

Sign in / Sign up

Export Citation Format

Share Document