Geometric and dosimetric evaluation of deep learning based auto-segmentation for clinical target volume on breast cancer.

BACKGROUND: Recently, target auto-segmentation techniques based on deep learning (DL) have shown promising results. However, inaccurate target delineation will directly affect the treatment planning dose distribution and the effect of subsequent radiotherapy work. Evaluation based on geometric metrics alone may not be sufficient for target delineation accuracy assessment. The purpose of this paper is to validate the performance of automatic segmentation with dosimetric metrics and try to construct new evaluation geometric metrics to comprehensively understand the dose-response relationship from the perspective of clinical application. MATERIALS AND METHODS: A DL-based target segmentation model was developed by using 186 manual delineation modified radical mastectomy breast cancer cases. The resulting DL model were used to generate alternative target contours in a new set of 48 patients. The Auto-plan was reoptimized to ensure the same optimized parameters as the reference Manual-plan. To assess the dosimetric impact of target auto-segmentation, not only common geometric metrics but also new spatial parameters with distance and relative volume ( R V ${R}_V$ ) to target were used. Correlations were performed using Spearman's correlation between segmentation evaluation metrics and dosimetric changes. RESULTS: Only strong (|R2 | > 0.6, p < 0.01) or moderate (|R2 | > 0.4, p < 0.01) Pearson correlation was established between the traditional geometric metric and three dosimetric evaluation indices to target (conformity index, homogeneity index, and mean dose). For organs at risk (OARs), inferior or no significant relationship was found between geometric parameters and dosimetric differences. Furthermore, we found that OARs dose distribution was affected by boundary error of target segmentation instead of distance and R V ${R}_V$ to target. CONCLUSIONS: Current geometric metrics could reflect a certain degree of dose effect of target variation. To find target contour variations that do lead to OARs dosimetry changes, clinically oriented metrics that more accurately reflect how segmentation quality affects dosimetry should be constructed.

The impact of training sample size on deep learning-based organ auto-segmentation for head-and-neck patients.

To investigate the impact of training sample size on the performance of deep learning-based organ auto-segmentation for head-and-neck cancer patients, a total of 1160 patients with head-and-neck cancer who received radiotherapy were enrolled in this study. Patient planning CT images and regions of interest (ROIs) delineation, including the brainstem, spinal cord, eyes, lenses, optic nerves, temporal lobes, parotids, larynx and body, were collected. An evaluation dataset with 200 patients were randomly selected and combined with Dice similarity index to evaluate the model performances. Eleven training datasets with different sample sizes were randomly selected from the remaining 960 patients to form auto-segmentation models. All models used the same data augmentation methods, network structures and training hyperparameters. A performance estimation model of the training sample size based on the inverse power law function was established. Different performance change patterns were found for different organs. Six organs had the best performance with 800 training samples and others achieved their best performance with 600 training samples or 400 samples. The benefit of increasing the size of the training dataset gradually decreased. Compared to the best performance, optic nerves and lenses reached 95% of their best effect at 200, and the other organs reached 95% of their best effect at 40. For the fitting effect of the inverse power law function, the fitted root mean square errors of all ROIs were less than 0.03 (left eye: 0.024, others: <0.01), and theRsquare of all ROIs except for the body was greater than 0.5. The sample size has a significant impact on the performance of deep learning-based auto-segmentation. The relationship between sample size and performance depends on the inherent characteristics of the organ. In some cases, relatively small samples can achieve satisfactory performance.

Clinical Implementation of Automated Treatment Planning for Rectum Intensity-Modulated Radiotherapy Using Voxel-Based Dose Prediction and Post-Optimization Strategies.

PURPOSE: This study aims to demonstrate the feasibility of clinical implementation of automated treatment planning (ATP) using voxel-based dose prediction and post-optimization strategies for rectal cancer on uRT (United Imaging Healthcare, Shanghai, China) treatment planning system. METHODS: A total of 180 previously treated rectal cancer cases were enrolled in this study, including 160 cases for training, 10 for validation and 10 for testing. Using CT image data, planning target volumes (PTVs) and contour delineation of the organs at risk (OARs) as input and three-dimensional (3D) dose distribution as output, a 3D-Uet DL model was developed. Based on the voxel-wise prediction dose distribution, intensity-modulated radiation therapy (IMRT) plans were then generated automatedly using post-optimization strategies, including a complex clinical dose target metrics homogeneity index (HI) and conformation index (CI). To evaluate the performance of the proposed ATP approach, the dose-volume histogram (DVH) parameters of OARs and PTV and the 3D dose distributions of the plan were compared with those of manual plans. RESULTS: By combining clinical post-optimization strategies, the automatically generated treatment plan can achieve better homogeneous PTV coverage and dose sparing for OARs except the mean dose for femoral-head compared with the use of the mean square error objective function alone. Compared with the manual plan, no statistically significant differences in HI, CI or global maximum dose were found. The manual plans perform slightly better than plans with post-optimization strategies in other dosimetric indexes, but these plans are still within clinical requirements. CONCLUSIONS: With the help of clinical post-optimization strategies, the proposed new ATP solution can generate IMRT plans that are within clinically acceptable levels and comparable to plans manually generated by dosimetrists.

A Preliminary Experience of Implementing Deep-Learning Based Auto-Segmentation in Head and Neck Cancer: A Study on Real-World Clinical Cases.

PURPOSE: While artificial intelligence has shown great promise in organs-at-risk (OARs) auto segmentation for head and neck cancer (HNC) radiotherapy, to reach the level of clinical acceptance of this technology in real-world routine practice is still a challenge. The purpose of this study was to validate a U-net-based full convolutional neural network (CNN) for the automatic delineation of OARs of HNC, focusing on clinical implementation and evaluation. METHODS: In the first phase, the CNN was trained on 364 clinical HNC patients' CT images with annotated contouring from routine clinical cases by different oncologists. The automated delineation accuracy was quantified using the Dice similarity coefficient (DSC) and 95% Hausdorff distance (HD). To assess efficiency, the time required to edit the auto-contours to a clinically acceptable standard was evaluated by a questionnaire. For subjective evaluation, expert oncologists (more than 10 years' experience) were randomly presented with automated delineations or manual contours of 15 OARs for 30 patient cases. In the second phase, the network was retrained with an additional 300 patients, which were generated by pre-trained CNN and edited by oncologists until to meet clinical acceptance. RESULTS: Based on DSC, the CNN performed best for the spinal cord, brainstem, temporal lobe, eyes, optic nerve, parotid glands and larynx (DSC >0.7). Higher conformity for the OARs delineation was achieved by retraining our architecture, largest DSC improvement on oral cavity (0.53 to 0.93). Compared with the manual delineation time, after using auto-contouring, this duration was significantly shortened from hours to minutes. In the subjective evaluation, two observes showed an apparent inclination on automatic OARs contouring, even for relatively low DSC values. Most of the automated OARs segmentation can reach the clinical acceptance level compared to manual delineations. CONCLUSIONS: After retraining, the CNN developed for OARs automated delineation in HNC was proved to be more robust, efficiency and consistency in clinical practice. Deep learning-based auto-segmentation shows great potential to alleviate the labor-intensive contouring of OAR for radiotherapy treatment planning.

An Artificial Intelligence-Based Full-Process Solution for Radiotherapy: A Proof of Concept Study on Rectal Cancer.

BACKGROUND AND PURPOSE: To develop an artificial intelligence-based full-process solution for rectal cancer radiotherapy. MATERIALS AND METHODS: A full-process solution that integrates autosegmentation and automatic treatment planning was developed under a single deep-learning framework. A convolutional neural network (CNN) was used to generate segmentations of the target and the organs at risk (OAR) as well as dose distribution. A script in Pinnacle that simulates the treatment planning process was used to execute plan optimization. A total of 172 rectal cancer patients were used for model training, and 18 patients were used for model validation. Another 40 rectal cancer patients were used for an end-to-end evaluation for both autosegmentation and treatment planning. The PTV and OAR segmentation was compared with manual segmentation. The planning results was evaluated by both objective and subjective assessment. RESULTS: The total time for full-process planning without contour modification was 7 min, and an additional 15 min may require for contour modification and re-optimization. The PTV DICE similarity coefficient was greater than 0.85 for all 40 patients in the evaluation dataset while the DICE indices of the OARs also indicated good performance. There were no significant differences between the auto plans and manual plans. The physician accepted 80% of the auto plans without any further operation. CONCLUSION: We developed a deep learning-based automatic solution for rectal cancer treatment that can improve the efficiency of treatment planning.