Deep Learning (DL) technique has shown great potential but still has limited success in online contouring for MR-guided adaptive radiotherapy (MRgART). This study proposed a patient-specific DL auto-segmentation (DLAS) strategy using the patient's previous images and contours to update the model and improve segmentation accuracy and efficiency for MRgART. A prototype model was trained for each patient using the first set of MRI and corresponding contours as inputs. The patient-specific model was updated after each fraction with all the available fractional MRIs/contours, and then used to predict the segmentation for the next fraction. During model training, a variant was fitted under consistency constraints, limiting the differences in the volume, length and centroid between the predictions for the latest MRI within a reasonable range. The model performance was evaluated for both organ-at-risks and tumors auto-segmentation for a total of 6 abdominal/pelvic cases (each with at least 8 sets of MRIs/contours) underwent MRgART through Dice Similarity Coefficient (DSC) and 95% Hausdorff Distance (HD95), and was compared with deformable image registration (DIR) and frozen DL model (no updating after pre-training). The contouring time was also recorded and analyzed. The proposed model achieved superior performance with higher mean DSC (0.90, 95 % CI: 0.88-0.95), as compared to DIR (0.63, 95 %CI: 0.59-0.68) and frozen DL models (0.74, 95 % CI: 0.71-0.79). As for tumors, the proposed method yielded a median DSC of 0.95, 95 % CI: 0.94-0.97, and a median HD95 of 1.63 mm, 95 % CI: 1.22 mm-2.06 mm. The contouring time was reduced significantly (p < 0.05) using the proposed method (73.4 ± 6.5 secs) compared to the manual process (12 ∼ 22 mins). The online ART time was reduced to 1650 ± 274 seconds with the proposed method, as compared to 3251.8 ± 447 seconds using the original workflow. The proposed patient-specific DLAS method can significantly improve the segmentation accuracy and efficiency for longitudinal MRIs, thereby facilitating the routine practice of MRgART.

Recent years have seen both a fresh knowledge of cancer and impressive advancements in its treatment. However, the clinical treatment paradigm of cancer is still difficult to implement in the twenty-first century due to the rise in its prevalence. Radiotherapy (RT) is a crucial component of cancer treatment that is helpful for almost all cancer types. The accuracy of RT dosage delivery is increasing as a result of the quick development of computer and imaging technology. The use of image-guided radiation (IGRT) has improved cancer outcomes and decreased toxicity. Online adaptive radiotherapy will be made possible by magnetic resonance imaging-guided radiotherapy (MRgRT) using a magnetic resonance linear accelerator (MR-Linac), which will enhance the visibility of malignancies. This review's objectives are to examine the benefits of MR-Linac as a treatment approach from the perspective of various cancer patients' prognoses and to suggest prospective development areas for additional study.

Background: This study was conducted to assess the efficacy and safety of magnetic resonance (MR)-guided hypofractionated radiotherapy in patients with unresectable hepatocellular carcinoma (HCC). Machine learning-based radiomics was utilized to predict responses in these patients. Methods: This retrospective study included 118 hCC patients who received MR-guided hypofractionated radiotherapy. The primary study endpoint was the objective response rate (ORR). Radiomics features were based on the gross tumor volume (GTV). K-means clustering was performed to differentiate cancer subtypes based on radiomics. Nine radiomics-utilizing machine learning models were built and validated internally through 5-fold cross-validation. Results: The ORR, median progression-free survival (mPFS), and median overall survival (mOS) were 54.4%, 21.7 months, and 40.7 months, respectively. No patient experienced Grade 3/4 adverse events. 1130 radiomics features were extracted from the GTV, of which 7 were included for further analysis. K-means clustering identified 2 subtypes based on the selected features. Subtype 1 had significantly higher response, longer mPFS, and longer mOS than Subtype 2. In both internal and external validations, the multi-layer perceptron (MLP) model demonstrated superior predictive performance for response, achieving a receiver operating characteristic-area under the curve (ROC-AUC) of 0.804 and 0.842, respectively. Conclusion: MR-guided radiotherapy was proven to be effective and safe for HCC. The machine learning radiomics model developed in this study could accurately predict the response of radiotherapy-treated inoperable HCC.

To evaluate and validate CT-based models using pre- and posttreatment deep learning radiomics features and hematological biomarkers for assessing esophageal squamous cell carcinoma (ESCC) pathological complete response (pCR) after neoadjuvant chemoradiotherapy (nCRT). This retrospective study recruited patients with biopsy-proven ESCC who underwent nCRT from two Chinese hospitals between May 2017 and May 2022, divided into a training set (hospital I, 111 cases), an internal validation set (hospital I, 47 cases), and an external validation set (hospital II, 33 cases). We used minimum redundancy maximum relevance (mRMR) and least absolute shrinkage and selection operator (LASSO) as feature selection methods and three classifiers as model construction methods. The assessment of models was performed using area under the receiver operating characteristic (ROC) curve (AUC) and decision curve analysis (DCA). A total 190 patients were included in our study (60.8 ± 7.08 years, 133 men), and seventy-seven of them (40.5 %) achieved pCR. The logistic regression (LR)-based combined model incorporating neutrophil to lymphocyte ratio, lymphocyte to monocyte ratio, albumin, and radscores performed well both in the internal and external validation sets with AUCs of 0.875 and 0.857 (95 % CI, 0.776-0.964; 0.731-0.984, P <0.05), respectively. DCA demonstrated that nomogram was useful for pCR prediction and produced clinical net benefits. The incorporation of radscores and hematological biomarkers into LR-based model improved pCR prediction after nCRT in ESCC. Enhanced pCR predictability may improve patients selection before surgery, providing clinical application value for the use of active surveillance.

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.

This study aimed to integrate artificial intelligence (AI)with heat shock protein 90 alpha (HSP90α)expression to improve patient selection and prognostic assessment in unresectable hepatocellular carcinoma (HCC)treated with transarterial chemoembolization (TACE). We retrospectively enrolled 2555 unresectable HCC patients treated between 2016 and 2021 at seven Chinese tertiary hospitals. Residual-based methods were used to define TACE benefit. Eight AI models revealed that HSP90α expression, Barcelona Clinic Liver Cancer (BCLC)stage, and tumor size were key predictive factors for TACE benefit. A nomogram based on these three variables achieved an area under the receiver operating characteristic curve (AUC)of 0.901 in the validation cohort. For overall survival (OS), we developed 101 machine learning models. The StepCox[forward] plus random survival forest model showed the best performance. Its C-indices were 0.84, 0.70, and 0.78 in the training, internal validation, and external validation sets, respectively. In the internal validation set, the time-dependent AUCs for 1-, 2-, and 3 year OS were 0.835, 0.821, and 0.776; in the external validation set, they were 0.854, 0.790, and 0.804. Integrating AI with HSP90α enables robust identification of TACE-benefit candidates and accurate prognostic stratification in unresectable HCC.