Abstract Book

S1169

ESTRO 37

segmented. Finally, we obtain the following values for the gamma indicators: γPR (95.25±4.86)%, γ95 (0.85±0.53). The γ95 indicator correlates well with the geometric metrics (Fig. 2), except for a few outliers.

automated ABAS contours (UABAS) and DLC contours (UDLC). MC was used as the baseline method. User contouring/editing was performed with software employed in clinical practice (Eclipse, version 11.0, Varian, Palo Alto, U.S.A.), which includes semi-automatic contouring tools. For each contouring method MC, UABAS and UDLC, the time required to complete the task was recorded per patient and OAR. Only the times requires to manually edit the DLC and ABAS contours were recorded, not the times to generate them. Results Table 1 reports the median and inter-quartile ranges of user editing time for MC, ABAS and DLC contours. UABAS led to a median contouring time reduction of 7.8 minutes compared to MC, while UDLC led to a median time reduction of 10 minutes compared to MC. For the lungs, spinal cord and esophagus, UDLC times were significantly lower than UABAS (p<0.05). Times for both automatic methods were significantly lower than MC for all organs.

Figure 1 - Average surface distances (in mm) projected onto the mean breast shape (cranial, posterior and frontal views).

Figure 2 - γ95 vs. the geometric metrics; R: Pearson correlation coefficient, p: p-value. Conclusion Mirada’s Workflow Box™ provides accurate CTV segmentations. The geometric differences occur in the same locations as the inter-observer variability reported in the literature. A visual analysis of the differences indicates that a larger variety of breast sizes should be included in the set of atlases used. The dosimetric impact of the manual adjustments is overall limited with most γ95 values lying below 1, indicating potential for an automated workflow that would require the manual revision of the contours in only a few cases. The proposed evaluation framework can be readily applied to other treatment sites. EP-2124 Time-saving evaluation of deep learning contouring of thoracic organs at risk T. Lustberg 1 , J. Van der Stoep 1 , D. Peressutti 2 , P. Aljabar 2 , W. Van Elmpt 1 , J. Van Soest 1 , M. Gooding 2 , A. Dekker 1 1 MAASTRO Clinic, Department of Radiation Oncology MAASTRO- GROW – School for Oncology and Developmental Biology- Maastricht University Medical Centre, Maastricht, The Netherlands 2 Mirada Medical Ltd, Science and Medical Technology, Oxford, United Kingdom Purpose or Objective Accurate contouring of organs-at-risk (OARs) in radiotherapy treatment planning is recommended for a successful outcome, but is often a time-consuming process. Auto-contouring methods have been proposed to improve consistency and reduce contouring time. This study reports on time savings generated by two automatic contouring methods: an atlas-based automatic segmentation (ABAS) (WorkflowBox 1.4, Mirada Medical Ltd, Oxford, UK) and a deep learning contouring (DLC) (WorkflowBox 2.0alpha, Mirada Medical Ltd, Oxford, UK). Their performance was evaluated on thoracic OARs for treatment of lung cancer patients. Material and Methods The automatic methods were evaluated on 20 stage I-III NSCLC patient CT images. The DLC system was trained on 450 lung cases collected at the same clinic. For the ABAS system 10 stage I NSCLC patients were carefully contoured according to the institutional guidelines and employed as atlases. All cases were acquired at a single institution. Evaluation was performed on the lungs, spinal cord, esophagus, heart and mediastinum envelope. A single user, an experienced radiotherapy technician, performed the manual contouring (MC) and the editing of

Figure 1 shows the fraction of time saved with respect to MC for the median values. For the lungs and the spinal cord, editing time for UDLC corresponded to a visual assessment of the contours with limited or no manual adjustments. Over all OARs, UABAS and UDLC led to a time saving of 41% and 64%, respectively, compared to MC.

Conclusion This investigation has demonstrated time savings for two OAR auto-contouring methods in the thorax. Editing the results of both methods provided significant time reductions for all OARs considered with respect to MC. DLC editing generated significant contouring time savings compared to ABAS for the lungs, spinal cord and esophagus. Future work will extend the time saving analysis to other body regions. EP-2125 Abdominal compression for SBRT: influence on 3D MRI image quality and diaphragm motion reduction O.L. Wong 1 , K.F. Cheng 2 , J. Yuan 1 , P.H. Fok 2 , Y.H. Zhou 1 , G. Chiu 2 , S.K. Yu 1 , K.Y. Cheung 1 1 Hong Kong Sanatorium & Hospital, Medical physics and Research, Hong Kong, Hong Kong SAR China 2 Hong Kong Sanatorium & Hospital, Department of Radiotherapy, Hong Kong, Hong Kong SAR China