Accurate object localization facilitates automatic esophagus segmentation in deep learning.

BACKGROUND: Currently, automatic esophagus segmentation remains a challenging task due to its small size, low contrast, and large shape variation. We aimed to improve the performance of esophagus segmentation in deep learning by applying a strategy that involves locating the object first and then performing the segmentation task. METHODS: A total of 100 cases with thoracic computed tomography scans from two publicly available datasets were used in this study. A modified CenterNet, an object location network, was employed to locate the center of the esophagus for each slice. Subsequently, the 3D U-net and 2D U-net_coarse models were trained to segment the esophagus based on the predicted object center. A 2D U-net_fine model was trained based on the updated object center according to the 3D U-net model. The dice similarity coefficient and the 95% Hausdorff distance were used as quantitative evaluation indexes for the delineation performance. The characteristics of the automatically delineated esophageal contours by the 2D U-net and 3D U-net models were summarized. Additionally, the impact of the accuracy of object localization on the delineation performance was analyzed. Finally, the delineation performance in different segments of the esophagus was also summarized. RESULTS: The mean dice coefficient of the 3D U-net, 2D U-net_coarse, and 2D U-net_fine models were 0.77, 0.81, and 0.82, respectively. The 95% Hausdorff distance for the above models was 6.55, 3.57, and 3.76, respectively. Compared with the 2D U-net, the 3D U-net has a lower incidence of delineating wrong objects and a higher incidence of missing objects. After using the fine object center, the average dice coefficient was improved by 5.5% in the cases with a dice coefficient less than 0.75, while that value was only 0.3% in the cases with a dice coefficient greater than 0.75. The dice coefficients were lower for the esophagus between the orifice of the inferior and the pulmonary bifurcation compared with the other regions. CONCLUSION: The 3D U-net model tended to delineate fewer incorrect objects but also miss more objects. Two-stage strategy with accurate object location could enhance the robustness of the segmentation model and significantly improve the esophageal delineation performance, especially for cases with poor delineation results.

Study of peripheral dose from low-dose CT to adaptive radiotherapy of postoperative prostate cancer.

Objectives: The increasing use of computed tomography (CT) for adaptive radiotherapy (ART) has raised concerns about the peripheral radiation dose. This study investigates the feasibility of low-dose CT (LDCT) for postoperative prostate cancer ART to reduce the peripheral radiation dose, and evaluates the peripheral radiation dose of different imaging techniques and propose an image enhancement method based on deep learning for LDCT. Materials and methods: A linear accelerator integrated with a 16-slice fan-beam CT from UIH (United Imaging Healthcare, China) was utilized for prostate cancer ART. To reduce the tube current of CT for ART, LDCT was acquired. Peripheral doses of normal-dose CT (NDCT), LDCT, and mega-voltage computed tomography (MV-CT) were measured using a cylindrical Virtual Water™ phantom and an ion chamber. A deep learning model of LDCT for abdominal and pelvic-based cycle-consistent generative adversarial network was employed to enhance the image quality of LDCT. Six postoperative prostate cancer patients were selected to evaluate the feasibility of low-dose CT network restoration images (RCT) by the deep learning model for ART. The three aspects among NDCT, LDCT, and RCT were compared: the Hounsfield Unit (HU) of the tissue, the Dice Similarity Coefficient (DSC) criterion of target and organ, and dose calculation differences. Results: In terms of peripheral dose, the LDCT had a surface measurement point dose of approximately 1.85 mGy at the scanning field, while the doses of NDCT and MV-CT were higher at 22.85 mGy and 29.97 mGy, respectively. However, the image quality of LDCT was worse than NDCT. When compared to LDCT, the tissue HU value of RCT showed a significant improvement and was closer to that of NDCT. The DSC results for target CTV between RCT and NDCT were also impressive, reaching up to 94% for bladder and femoral heads, 98% for rectum, and 94% for the target organ. Additionally, the dose calculation differences for the ART plan based on LDCT and NDCT were all within 1%. Overall, these findings suggest that RCT can provide an effective alternative to NDCT and MV-CT with similar or better outcomes in HU values of tissue and organ damage. More testing is required before clinical application.

Quantitative study on dose distribution of Freiburg flap for keloid high-dose-rate brachytherapy based on MatriXX.

PURPOSE: To quantify the dose distribution effect of insufficient scattering conditions in keloid HDR brachytherapy with Freiburg fFlap (FF) applicator. MATERIALS AND METHODS: A phantom composed of FF applicator, MatriXX and solid water slices was designed and scanned for treatment planning. Bolus with different thicknesses were covered to offer different scatter conditions. Planar dose distributions were measured by MatriXX. The maximum value (Max), average value (Avg) and γ passing rate (3 mm/3%) were evaluated by the software MyQA Platform. RESULTS: The maximum and average doses measured by MatriXX were lower than the calculated values. The difference increased as field size decreased. The Max value, found at 0.86 cm level in the two tube case, was -20.0%, and the avg value was -11.9%. All the γ values were less than 95%. This difference gradually decreased with increasing bolus thickness and the γ values were significantly improved. CONCLUSION: MatriXX could be used for dose verification of HDR brachytherapy with an FF applicator. When the FF applicator was applied for keloid, insufficient scattering conditions would cause an insufficient target dose. This difference could be reduced by covering the bolus with different thicknesses on the applicator. The smaller the field, the thicker the bolus required.

Analysis of clinical and physical dosimetric factors that determine the outcome of severe acute radiation pneumonitis in lung cancer patients.

OBJECTIVE: We conducted a retrospective statistical analysis of clinical and physical dosimetric factors of lung cancer patients who had previously undergone lung and/or mediastinal radiotherapy and died of or survived severe acute radiation pneumonitis (SARP). Our study was the first to reveal the heterogeneity in clinical factors, physical dosimetric factors, and SARP onset time that determined the clinical outcomes of lung cancer patients who developed SARP. MATERIALS AND METHODS: The clinical characteristics, physical dosimetry factors, and SARP onset time of deceased and surviving patients were retrospectively analyzed. SPSS 20.0 was used for data analysis. Student's t-test was used for intergroup comparison, and a Mann-Whitney U test was used for data with skewed distribution. Qualitative data were represented using frequencies (%), and Fisher's exact test or χ2 test was used for intergroup comparison of nonparametric data. Binary logistic analysis was used for univariate and multivariate analyses. Differences with a P < 0.05 were considered statistically significant. RESULTS: Univariate analysis revealed that the potential predictors of SARP death were as follows: ipsilateral lung V5 and V30, contralateral lung V5, V10, and V30, total lung V5, V10, and V30, mean lung dose, mean heart dose, and maximum spinal cord dose. Multivariate analysis showed that ipsilateral lung V5 and total lung V5 were predictors that determined the final outcome of SARP patients. In addition, we analyzed the time from the completion of radiotherapy to SARP onset, and found significant difference between the two groups. CONCLUSIONS: There was no decisive correlation between clinical characteristics and SARP outcome (i.e., death or survival) in lung radiotherapy patients. Ipsilateral lung V5 and total lung V5 were independent predictors of death in SARP patients.

Establishment of a prediction model for severe acute radiation enteritis associated with cervical cancer radiotherapy.

BACKGROUND: Cervical cancer is one of the most common gynecological malignant tumors. Radiation enteritis (RE) leads to radiotherapy intolerance or termination of radiotherapy, which negatively impacts the therapeutic effect and seriously affects the quality of life of patients. If the incidence of RE in patients can be predicted in advance, and targeted clinical preventive treatment can be carried out, the side effects of radiotherapy in cervical cancer patients can be significantly reduced. Furthermore, accurate prediction of RE is essential for the selection of individualized radiation dose and the optimization of the radiotherapy plan. AIM: To analyze the relationships between severe acute RE (SARE) of cervical cancer radiotherapy and clinical factors and dose-volume parameters retrospectively. METHODS: We included 50 cervical cancer patients who received volumetric modulated arc therapy (VMAT) from September 2017 to June 2018 in the Department of Radiotherapy at The First Affiliated Hospital Soochow University. Clinical and dose-volume histogram factors of patients were collected. Logistic regression analysis was used to evaluate the predictive value of each factor for SARE. A nomogram to predict SARE was developed (SARE scoring system ≥ 3 points) based on the multiple regression coefficients; validity was verified by an internal verification method. RESULTS: Gastrointestinal and hematological toxicity of cervical cancer VMAT gradually increased with radiotherapy and reached the peak at the end of radiotherapy. The main adverse reactions were diarrhea, abdominal pain, colitis, anal swelling, and blood in the stool. There was no significant difference in the incidence of gastrointestinal toxicity between the radical and postoperative adjuvant radiotherapy groups (P > 0.05). There were significant differences in the small intestine V20, V30, V40, and rectal V40 between adjuvant radiotherapy and radical radiotherapy after surgery (P < 0.05). Univariate and multivariate analyses revealed anal bulge rating (OR: 14.779, 95%CI: 1.281-170.547, P = 0.031) and disease activity index (DAI) score (OR: 53.928, 95%CI: 3.822-760.948, P = 0.003) as independent predictors of SARE. CONCLUSION: Anal bulge rating (> 0.500 grade) and DAI score (> 2.165 points) can predict SARE. The nomogram shows potential value in clinical practice.

Low-dose dynamic myocardial perfusion CT image reconstruction using pre-contrast normal-dose CT scan induced structure tensor total variation regularization.

Dynamic myocardial perfusion CT (DMP-CT) imaging provides quantitative functional information for diagnosis and risk stratification of coronary artery disease by calculating myocardial perfusion hemodynamic parameter (MPHP) maps. However, the level of radiation delivered by dynamic sequential scan protocol can be potentially high. The purpose of this work is to develop a pre-contrast normal-dose scan induced structure tensor total variation regularization based on the penalized weighted least-squares (PWLS) criteria to improve the image quality of DMP-CT with a low-mAs CT acquisition. For simplicity, the present approach was termed as 'PWLS-ndiSTV'. Specifically, the ndiSTV regularization takes into account the spatial-temporal structure information of DMP-CT data and further exploits the higher order derivatives of the objective images to enhance denoising performance. Subsequently, an effective optimization algorithm based on the split-Bregman approach was adopted to minimize the associative objective function. Evaluations with modified dynamic XCAT phantom and preclinical porcine datasets have demonstrated that the proposed PWLS-ndiSTV approach can achieve promising gains over other existing approaches in terms of noise-induced artifacts mitigation, edge details preservation, and accurate MPHP maps calculation.

[Design of Smart Care Tele-Monitoring System for Mother and Fetus].

OBJECTIVE: To study and design a maternal and fetal monitoring system based on the cloud computing and internet of things, which can monitor and take smart care of the mother and fetus in 24 h. METHODS: Using a new kind of wireless fetal monitoring detector and a mobile phone, thus the doctor can keep touch with hospital through internet. The mobile terminal was developed on the Android system, which accepted the data of fetal heart rate and uterine contraction transmitted from the wireless detector, exchange information with the server and display the monitoring data and the doctor's advice in real-time. RESULTS: The mobile phone displayed the fetal heart rate line and uterine contraction line in real-time, recorded the fetus' grow process. It implemented the real-time communication between the doctor and the user, through wireless communication technology. CONCLUSIONS: The system removes the constraint of traditional telephone cable for users, while the users can get remote monitoring from the medical institutions at home or in the nearest community at any time, providing health and safety guarantee for mother and fetus.