Search

Background: The goal was to investigate the feasibility of the registration generative adversarial network (RegGAN) model in image conversion for performing adaptive radiation therapy on the head and neck and its stability under different cone beam computed tomography (CBCT) models. Methods: A total of 100 CBCT and CT images of patients diagnosed with head and neck tumors were utilized for the training phase, whereas the testing phase involved 40 distinct patients obtained from four different linear accelerators. The RegGAN model was trained and tested to evaluate its performance. The generated synthetic CT (sCT) image quality was compared to that of planning CT (pCT) images by employing metrics such as the mean absolute error (MAE), peak signal-to-noise ratio (PSNR), and structural similarity index measure (SSIM). Moreover, the radiation therapy plan was uniformly applied to both the sCT and pCT images to analyze the planning target volume (PTV) dose statistics and calculate the dose difference rate, reinforcing the model's accuracy. Results: The generated sCT images had good image quality, and no significant differences were observed among the different CBCT modes. The conversion effect achieved for Synergy was the best, and the MAE decreased from 231.3 ± 55.48 to 45.63 ± 10.78; the PSNR increased from 19.40 ± 1.46 to 26.75 ± 1.32; the SSIM increased from 0.82 ± 0.02 to 0.85 ± 0.04. The quality improvement effect achieved for sCT image synthesis based on RegGAN was obvious, and no significant sCT synthesis differences were observed among different accelerators. Conclusion: The sCT images generated by the RegGAN model had high image quality, and the RegGAN model exhibited a strong generalization ability across different accelerators, enabling its outputs to be used as reference images for performing adaptive radiation therapy on the head and neck. [ABSTRACT FROM AUTHOR]

Background: Predicting the recurrence risk of bladder cancer is crucial for the individualized clinical treatment of patients with bladder cancer. Objective: To explore the radiomics based on multiphase CT images combined with clinical risk factors, and to further construct a radiomics-clinical model to predict the recurrence risk of bladder cancer within 2 years after surgery. Methods: Patients with bladder cancer who underwent surgical treatment at the First Affiliated Hospital of Soochow University from January 2016 to December 2019 were retrospectively included and followed up to record the disease recurrence. A total of 183 patients were included in the study, and they were randomly divided into training group and validation group in a ratio of 7: 3. The three basic models which are plain scan, corticomedullary phase, and nephrographic phase as well as two combination models, namely, corticomedullary phase + nephrographic phase and plain scan + corticomedullary phase + nephrographic phase, were built with the logistic regression algorithm, and we selected the model with higher performance and calculated the Rad-score (radiomics score) of each patient. The clinical risk factors and Rad-score were screened by Cox univariate and multivariate proportional hazard models in turn to obtain the independent risk factors, then the radiomics-clinical model was constructed, and their performance was evaluated. Results: Of the 183 patients included, 128 patients constituted the training group and 55 patients constituted the validation group. In terms of the radiomics-clinical model constructed by three independent risk factors—number of tumors, tumor grade, and Rad-score—the AUCs of the training group and validation group were 0.813 (95% CI 0.740–0.886) and 0.838 (95% CI 0.733–0.943), respectively. In the validation group, the diagnostic accuracy, sensitivity, and specificity were 0.727, 0.739, and 0.719, respectively. Conclusion: Combining with radiomics based on multiphase CT images and clinical risk factors, the radiomics-clinical model constructed to predict the recurrence risk of bladder cancer within 2 years after surgery had a good performance. [ABSTRACT FROM AUTHOR]