PET segmentation of bulky tumors: Strategies and workflows to improve inter-observer variability

Abstract

Background PET-based tumor delineation is an error prone and labor intensive part of image analysis. Especially for patients with advanced disease showing bulky tumor FDG load, segmentations are challenging. Reducing the amount of user-interaction in the segmentation might help to facilitate segmentation tasks especially when labeling bulky and complex tumors. Therefore, this study reports on segmentation workflows/strategies that may reduce the inter-observer variability for large tumors with complex shapes with different levels of userinteraction. Methods Twenty PET images of bulky tumors were delineated independently by six observers using four strategies: (I) manual, (II) interactive threshold-based, (III) interactive threshold-based segmentation with the additional presentation of the PET-gradient image and (IV) the selection of the most reasonable result out of four established semi-automatic segmentation algorithms (Select-the-best approach). The segmentations were compared using Jaccard coefficients (JC) and percentage volume differences. To obtain a reference standard, a majority vote (MV) segmentation was calculated including all segmentations of experienced observers. Performed and MV segmentations were compared regarding positive predictive value (PPV), sensitivity (SE), and percentage volume differences. Results The results show that with decreasing user-interaction the inter-observer variability decreases. JC values and percentage volume differences of Select-the-best and a workflow including gradient information were significantly better than the measurements of the other

Interobserver Agreement on Automated Metabolic Tumor Volume Measurements of Deauville Score 4 and 5 Lesions at Interim 18F-FDG PET in Diffuse Large B-Cell Lymphoma

“…However, no single semi-automated delineation method, including MV3 performs optimally for different types of lymphoma at different therapeutic stages without the need for manual correction (25). Therefore a workflow where observers select the visually best performing method per lesion might improve overall delineation success while minimizing interobserver variability compared to manual segmentation (36). Translating this workflow at I-PET DBLCL might imply that lesions with SUVmax<10 should be delineated using method MV3 and that lesions with SUVmax>10 should be delineated using method SUV4.0, or only in case observers consider MV3 contouring as failed.…”

Plausibility and redundancy analysis to select FDG‐PET textural features in non‐small cell lung cancer

“…It is, however, highly recommended to keep the level of user interaction as low as possible, as we recently showed in Ref. [44]. By using a semi‐automated segmentation approach and workflow, which required delineation adjustment in a few cases (when the tumor was located to another high‐uptake region), we tried to limit the amount of user interaction and we found that this reduces the observer variability as much as possible.…”

Locoregional Recurrence Prediction Using a Deep Neural Network of Radiological and Radiotherapy Images

“…Although tumor segmentation and feature extraction are essential parts of -omics prediction, they are not integrated into a single model. Consequently, such -omics methods are hardly reproducible in other institutions [17][18][19][20]. Therefore, it is necessary to deal with radiological and radiotherapy images themselves, instead of radiomics and dosiomics.…”