Comparison of the use of a clinically implemented deep learning segmentation model with the simulated study setting for breast cancer patients receiving radiotherapy.

BACKGROUND: Deep learning (DL) models for auto-segmentation in radiotherapy have been extensively studied in retrospective and pilot settings. However, these studies might not reflect the clinical setting. This study compares the use of a clinically implemented in-house trained DL segmentation model for breast cancer to a previously performed pilot study to assess possible differences in performance or acceptability. MATERIAL AND METHODS: Sixty patients with whole breast radiotherapy, with or without an indication for locoregional radiotherapy were included. Structures were qualitatively scored by radiotherapy technologists and radiation oncologists. Quantitative evaluation was performed using dice-similarity coefficient (DSC), 95th percentile of Hausdorff Distance (95%HD) and surface DSC (sDSC), and time needed for generating, checking, and correcting structures was measured. RESULTS: Ninety-three percent of all contours in clinic were scored as clinically acceptable or usable as a starting point, comparable to 92% achieved in the pilot study. Compared to the pilot study, no significant changes in time reduction were achieved for organs at risks (OARs). For target volumes, significantly more time was needed compared to the pilot study for patients including lymph node levels 1-4, although time reduction was still 33% compared to manual segmentation. Almost all contours have better DSC and 95%HD than inter-observer variations. Only CTVn4 scored worse for both metrics, and the thyroid had a higher 95%HD value. INTERPRETATION: The use of the DL model in clinical practice is comparable to the pilot study, showing high acceptability rates and time reduction.

Knowledge-based versus deep learning based treatment planning for breast radiotherapy.

Background and Purpose: To improve radiotherapy (RT) planning efficiency and plan quality, knowledge-based planning (KBP) and deep learning (DL) solutions have been developed. We aimed to make a direct comparison of these models for breast cancer planning using the same training, validation, and testing sets. Materials and Methods: Two KBP models were trained and validated with 90 RT plans for left-sided breast cancer with 15 fractions of 2.6 Gy. The versions either used the full dataset (non-clean model) or a cleaned dataset (clean model), thus eliminating geometric and dosimetric outliers. Results were compared with a DL U-net model (previously trained and validated with the same 90 RT plans) and manually produced RT plans, for the same independent dataset of 15 patients. Clinically relevant dose volume histogram parameters were evaluated according to established consensus criteria. Results: Both KBP models underestimated the mean heart and lung dose equally 0.4 Gy (0.3-1.1 Gy) and 1.4 Gy (1.1-2.8 Gy) compared to the clinical plans 0.8 Gy (0.5-1.8 Gy) and 1.7 Gy (1.3-3.2 Gy) while in the final calculations the mean lung dose was higher 1.9-2.0 Gy (1.5-3.5 Gy) for both KPB models. The U-Net model resulted in a mean planning target volume dose of 40.7 Gy (40.4-41.3 Gy), slightly higher than the clinical plans 40.5 Gy (40.1-41.0 Gy). Conclusions: Only small differences were observed between the estimated and final dose calculation and the clinical results for both KPB models and the DL model. With a good set of breast plans, the data cleaning module is not needed and both KPB and DL models lead to clinically acceptable results.

Evaluation of a clinically introduced deep learning model for radiotherapy treatment planning of breast cancer.

Deep learning (DL) models are increasingly studied to automate the process of radiotherapy treatment planning. This study evaluates the clinical use of such a model for whole breast radiotherapy. Treatment plans were automatically generated, after which planners were allowed to manually adapt them. Plans were evaluated based on clinical goals and DVH parameters. Thirty-seven of 50plans did fulfill all clinical goals without adjustments. Thirteen of these 37 plans were still adjusted but did not improve mean heart or lung dose. These results leave room for improvement of both the DL model as well as education on clinically relevant adjustments.

Comparison of the output of a deep learning segmentation model for locoregional breast cancer radiotherapy trained on 2 different datasets.

Introduction: The development of deep learning (DL) models for auto-segmentation is increasing and more models become commercially available. Mostly, commercial models are trained on external data. To study the effect of using a model trained on external data, compared to the same model trained on in-house collected data, the performance of these two DL models was evaluated. Methods: The evaluation was performed using in-house collected data of 30 breast cancer patients. Quantitative analysis was performed using Dice similarity coefficient (DSC), surface DSC (sDSC) and 95th percentile of Hausdorff Distance (95% HD). These values were compared with previously reported inter-observer variations (IOV). Results: For a number of structures, statistically significant differences were found between the two models. For organs at risk, mean values for DSC ranged from 0.63 to 0.98 and 0.71 to 0.96 for the in-house and external model, respectively. For target volumes, mean DSC values of 0.57 to 0.94 and 0.33 to 0.92 were found. The difference of 95% HD values ranged 0.08 to 3.23 mm between the two models, except for CTVn4 with 9.95 mm. For the external model, both DSC and 95% HD are outside the range of IOV for CTVn4, whereas this is the case for the DSC found for the thyroid of the in-house model. Conclusions: Statistically significant differences were found between both models, which were mostly within published inter-observer variations, showing clinical usefulness of both models. Our findings could encourage discussion and revision of existing guidelines, to further decrease inter-observer, but also inter-institute variability.

Clinical evaluation of a deep learning segmentation model including manual adjustments afterwards for locally advanced breast cancer.

Introduction: Deep learning (DL) models are increasingly developed for auto-segmentation in radiotherapy. Qualitative analysis is of great importance for clinical implementation, next to quantitative. This study evaluates a DL segmentation model for left- and right-sided locally advanced breast cancer both quantitatively and qualitatively. Methods: For each side a DL model was trained, including primary breast CTV (CTVp), lymph node levels 1-4, heart, lungs, humeral head, thyroid and esophagus. For evaluation, both automatic segmentation, including correction of contours when needed, and manual delineation was performed and both processes were timed. Quantitative scoring with dice-similarity coefficient (DSC), 95% Hausdorff Distance (95%HD) and surface DSC (sDSC) was used to compare both the automatic (not-corrected) and corrected contours with the manual contours. Qualitative scoring was performed by five radiotherapy technologists and five radiation oncologists using a 3-point Likert scale. Results: Time reduction was achieved using auto-segmentation in 95% of the cases, including correction. The time reduction (mean ± std) was 42.4% ± 26.5% and 58.5% ± 19.1% for OARs and CTVs, respectively, corresponding to an absolute mean reduction (hh:mm:ss) of 00:08:51 and 00:25:38. Good quantitative results were achieved before correction, e.g. mean DSC for the right-sided CTVp was 0.92 ± 0.06, whereas correction statistically significantly improved this contour by only 0.02 ± 0.05, respectively. In 92% of the cases, auto-contours were scored as clinically acceptable, with or without corrections. Conclusions: A DL segmentation model was trained and was shown to be a time-efficient way to generate clinically acceptable contours for locally advanced breast cancer.

Clinical evaluation of two AI models for automated breast cancer plan generation.

BACKGROUND: Artificial intelligence (AI) shows great potential to streamline the treatment planning process. However, its clinical adoption is slow due to the limited number of clinical evaluation studies and because often, the translation of the predicted dose distribution to a deliverable plan is lacking. This study evaluates two different, deliverable AI plans in terms of their clinical acceptability based on quantitative parameters and qualitative evaluation by four radiation oncologists. METHODS: For 20 left-sided node-negative breast cancer patients, treated with a prescribed dose of 40.05 Gy, using tangential beam intensity modulated radiotherapy, two model-based treatment plans were evaluated against the corresponding manual plan. The two models used were an in-house developed U-net model and a vendor-developed contextual atlas regression forest model (cARF). Radiation oncologists evaluated the clinical acceptability of each blinded plan and ranked plans according to preference. Furthermore, a comparison with the manual plan was made based on dose volume histogram parameters, clinical evaluation criteria and preparation time. RESULTS: The U-net model resulted in a higher average and maximum dose to the PTV (median difference 0.37 Gy and 0.47 Gy respectively) and a slightly higher mean heart dose (MHD) (0.01 Gy). The cARF model led to higher average and maximum doses to the PTV (0.30 and 0.39 Gy respectively) and a slightly higher MHD (0.02 Gy) and mean lung dose (MLD, 0.04 Gy). The maximum MHD/MLD difference was ≤ 0.5 Gy for both AI plans. Regardless of these dose differences, 90-95% of the AI plans were considered clinically acceptable versus 90% of the manual plans. Preferences varied between the radiation oncologists. Plan preparation time was comparable between the U-net model and the manual plan (287 s vs 253 s) while the cARF model took longer (471 s). When only considering user interaction, plan generation time was 121 s for the cARF model and 137 s for the U-net model. CONCLUSIONS: Two AI models were used to generate deliverable plans for breast cancer patients, in a time-efficient manner, requiring minimal user interaction. Although the AI plans resulted in slightly higher doses overall, radiation oncologists considered 90-95% of the AI plans clinically acceptable.

Development and evaluation of radiotherapy deep learning dose prediction models for breast cancer.

BACKGROUND AND PURPOSE: Treatment planning of radiotherapy is a time-consuming and planner dependent process that can be automated by dose prediction models. The purpose of this study was to evaluate the performance of two machine learning models for breast cancer radiotherapy before possible clinical implementation. MATERIALS AND METHODS: An in-house developed model, based on U-net architecture, and a contextual atlas regression forest (cARF) model integrated in the treatment planning software were trained. Obtained dose distributions were mimicked to create clinically deliverable plans. For training and validation, 90 patients were used, 15 patients were used for testing. Treatment plans were scored on predefined evaluation criteria and percent errors with respect to clinical dose were calculated for doses to planning target volume (PTV) and organs at risk (OARs). RESULTS: The U-net plans before mimicking met all criteria for all patients, both models failed one evaluation criterion in three patients after mimicking. No significant differences (p < 0.05) were found between clinical and predicted U-net plans before mimicking. Doses to OARs in plans of both models differed significantly from clinical plans, but no clinically relevant differences were found. After mimicking, both models had a mean percent error within 1.5% for the average dose to PTV and OARs. The mean errors for maximum doses were higher, within 6.6%. CONCLUSIONS: Differences between predicted doses to OARs of the models were small when compared to clinical plans, and not found to be clinically relevant. Both models show potential in automated treatment planning for breast cancer.

Reduction of heart and lung normal tissue complication probability using automatic beam angle optimization and more generic optimization objectives for breast radiotherapy.

During breast cancer radiotherapy, sparing of healthy tissue is desired. The effect of automatic beam angle optimization and generic dose fall-off objectives on dose and normal tissue complication probabilities was studied. In all patients, dose to lungs and heart showed a mean reduction of 0.4 Gy (range 0.1-1.3 Gy) and 0.2 Gy (range -0.2-0.7 Gy), respectively. These lower doses led to a statistically significant lower cumulative cardiac and lung cancer mortality risk. For smoking patients 40-45 years of age who continue to smoke, it would lead to a reduction from 3.2% ± 0.7% to 2.7% ± 0.6% (p < 0.001).

Artificial intelligence based treatment planning of radiotherapy for locally advanced breast cancer.

BACKGROUND AND PURPOSE: Treatment planning of radiotherapy for locally advanced breast cancer patients can be a time consuming process. Artificial intelligence based treatment planning could be used as a tool to speed up this process and maintain plan quality consistency. The purpose of this study was to create treatment plans for locally advanced breast cancer patients using a Convolutional Neural Network (CNN). MATERIALS AND METHODS: Data of 60 patients treated for left-sided breast cancer was used with a training, validation and test split of 36/12/12, respectively. The in-house built CNN model was a hierarchically densely connected U-net (HD U-net). The inputs for the HD U-net were 2D distance maps of the relevant regions of interest. Dose predictions, generated by the HD U-net, were used for a mimicking algorithm in order to create clinically deliverable plans. RESULTS: Dose predictions were generated by the HD U-net and mimicked using a commercial treatment planning system. The predicted plans fulfilling all clinical goals while showing small (≤0.5 Gy) statistically significant differences (p < 0.05) in the doses compared to the manual plans. The mimicked plans show statistically significant differences in the average doses for the heart and lung of ≤0.5 Gy and a reduced D2% of all PTVs. In total, ten of the twelve mimicked plans were clinically acceptable. CONCLUSIONS: We created a CNN model which can generate clinically acceptable plans for left-sided locally advanced breast cancer patients. This model shows great potential to speed up the treatment planning process while maintaining consistent plan quality.