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.

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.

Acceptance procedure for the linear accelerator component of the 1.5 T MRI-linac.

PURPOSE: To develop and implement an acceptance procedure for the new Elekta Unity 1.5 T MRI-linac. METHODS: Tests were adopted and, where necessary adapted, from AAPM TG106 and TG142, IEC 60976 and NCS 9 and NCS 22 guidelines. Adaptations were necessary because of the atypical maximum field size (57.4 × 22 cm), FFF beam, the non-rotating collimator, the absence of a light field, the presence of the 1.5 T magnetic field, restricted access to equipment within the bore, fixed vertical and lateral table position, and the need for MR image to MV treatment alignment. The performance specifications were set for stereotactic body radiotherapy (SBRT). RESULTS: The new procedure was performed similarly to that of a conventional kilovoltage x-ray (kV) image guided radiation therapy (IGRT) linac. Results were acquired for the first Unity system. CONCLUSIONS: A comprehensive set of tests was developed, described and implemented for the MRI-linac. The MRI-linac met safety requirements for patients and operators. The system delivered radiation very accurately with, for example a gantry rotation locus of isocenter of radius 0.38 mm and an average MLC absolute positional error of 0.29 mm, consistent with use for SBRT. Specifications for clinical introduction were met.

Clinical value of (dedicated) 3 Tesla and 7 Tesla MRI for cT1 glottic carcinoma: A feasibility study.

OBJECTIVE: To assess the feasibility of the clinical use of 3 Tesla and 7 Tesla Magnetic Resonance Imaging for early (cT1) glottic carcinoma, including structural assessment of technical image quality and visibility of the tumor; and if feasible, to correlate MRI findings to routine diagnostics. METHODS: Prospective feasibility study. Twenty patients with primary clinical T1 glottic carcinoma underwent both routine clinical staging and CT. In addition, a 3 T and 7 T MRI protocol, developed for small laryngeal lesions, was performed in a 4-point immobilization mask, using dedicated surface coils. Afterwards, routine endoscopic direct suspension laryngoscopy under general anaesthesia was performed. RESULTS: Only 2 of 7 (29%) of 7 T MRI scans were rated as moderate to good technical image quality. After exclusion of three patients with only mild to moderate dysplasia at the time of MRI, 13 of 17 (76%) of 3 T MRIs were of adequate technical image quality. Tumor visualization was adequate in 8 of 13 (62%) of patients with invasive squamous cell carcinomas. With exclusion of the four MRIs with motion artefacts, the tumor and its boundaries could be adequately seen in 8 of 9 (89%) patients with squamous cell carcinoma versus only one in four (25%) of patients with carcinoma in situ lesions. CONCLUSIONS: 7 Tesla MRI was considered not feasible. 3 Tesla MRI, with adequate patient selection, namely clinical exclusion of patients with a history of claustrophobia and inclusion of only histologically proven invasive squamous cell carcinoma, can be feasible. Especially with further improvement of MR image quality. LEVEL OF EVIDENCE: 2B, prospective diagnostic study.

Systematic review of the diagnostic value of magnetic resonance imaging for early glottic carcinoma.

Objective: In early glottic cancer, accurate assessment of tumor extension, including depth infiltration, is of great importance for both staging, therapeutic approach and systematic comparison of data. Our goal was to assess the diagnostic value of MRI in pre-therapeutic staging of primary early stage (T1 and T2) glottic carcinoma. Study design: Systematic review of literature. Methods: We conducted a systematic search in Pubmed, Embase, and Scopus up to September 23, 2016. Included studies were selected and critically appraised for relevance and validity. Results: Seven out of 938 unique articles were selected, including 64 cases. MRI over-staged 6% and under-staged 13% of cT1 and cT2 tumors. However, available data is heterogeneous, very limited and mainly based on subanalysis of a small amount of patients. Reported MRI protocols appear to be suboptimal for small laryngeal lesions. Diagnostic value of MRI for subtle depth infiltration or laryngeal anatomical subsites (eg, laryngeal ventricle, vocal cord, etc.) could not be assessed. Conclusions: More studies are needed to assess the diagnostic value of MRI for small glottic tumors.