ESTRO 2023 - Abstract Book

S669

Monday 15 May 2023

ESTRO 2023

Conclusion We present and apply a framework that allows the evaluation of DL autocontouring of the CTV. Model performance was comparable to the IOV and minor changes were required in regions of significant IOV. We showed the potential clinical relevance of the corrections, since target coverage increased after correcting. These findings combined allow the safe clinical commissioning of a DL model for CTV autocontouring. MO-0799 Single-click user input reduces false detection in deep learning head and neck tumor segmentation J. Ren 1,2,3 , J. Nijkamp 1,3 , M.E. Rasmussen 1,2,3 , J.G. Eriksen 4 , S.S. Korreman 1,2,3 1 Aarhus University, Department of Clinical Medicine, Aarhus, Denmark; 2 Aarhus University, Department of Oncology, Aarhus, Denmark; 3 Aarhus University Hospital, Danish Center for Particle Therapy, Aarhus, Denmark; 4 Aarhus University Hospital, Department of Experimental Clinical Oncology, Aarhus, Denmark Purpose or Objective Gross tumor volumes (GTV) of head and neck cancer (HNC) are difficult to identify in images, even with deep learning (DL), particularly when the primary tumor (GTV-T) and multiple nodal metastases (GTV-N) are present. Using only imaging for DL segmentation may lead to false segmentations. This study examines whether minimal user input, in which the oncologist only needs to single-click the lesions to be segmented, improves the detection ratio and segmentation performance of deep learning-based auto-segmentation for GTV-T and GTV-N of HNC. Materials and Methods We have included treatment planning CT, PET, and MRI (T1w mDixon and T2w) images for 567 HNC patients with a wide variety of tumor sites (larynx, pharynx, oral cavity, sinonasal, and salivary gland carcinomas). GTV-T and GTV-N clinical delineations were treated as two separate DL targets. The data was randomly split into training(n=375), validation(n=95), and test sets(n=97). To simulate user input clicks, we generated a dot of random size between 5 and 10 mm ³ at a random location inside each distinct target volume. We used this simulated user feed in conjunction with CT, PET, T1w, and T2w MRI scans as inputs to a 3D UNet. We compared the segmentation results to the UNet using only the scans as input. We evaluated the detection ratio(%) on all the distinct GTV-Ts and GTV-Ns. The segmentation performance was evaluated using Dice Similarity Coefficient(Dice), Hausdorff distance 95%(HD95), mean surface distance(MSD), and Surface-Dice with 2mm tolerance. The voxel-based false discovery rate (FDR) and false negative rate (FNR) were used to measure false segmentation and compared using a Wilcoxon signed-rank test(p<0.05). FDR can be interpreted as an indicator of false positive segmentations, whereas FNR indicates false negative segmentations. For all metrics, the mean and 95% confidence interval (CI95, bootstrapping 10000 samples) were reported. Results On the test set of 97 patients with 100 GTV-Ts and 177 GTV-Ns, after incorporating user feed, the detection ratio increased from GTV-T(95%)/GTV-N(79%) to GTV-T(99%)/GTV-N(99%). All segmentation metrics were improved, especially for GTV-N (Figure 1-A). The mean(CI95) of FDR for GTV-T/-N decreased from 0.27(0.23-0.30)/0.31(0.26-0.37) to 0.22(0.20- 0.25)/0.16(0.14-0.19) with p<0.001. FNR of GTV-T was marginally affected by user input, reduced from 0.32(0.28-0.36) to 0.29(0.25-0.32) with p>0.05. Whereas FNR of GTV-N decreased significantly from 0.29(0.25-0.32) to 0.17(0.16-0.19) with p<0.001 (Figure 1-B). Two cases with significant improvement in GTV-N segmentation are plotted in Figure 2.