RESUMEN
Background and purpose: Interactive segmentation seeks to incorporate human knowledge into segmentation models and thereby reducing the total amount of editing of auto-segmentations. By performing only interactions which provide new information, segmentation performance may increase cost-effectively. The aim of this study was to develop, evaluate and test feasibility of a deep learning-based single-cycle interactive segmentation model with the input being computer tomography (CT) and a small amount of information rich contours. Methods and Materials: A single-cycle interactive segmentation model, which took CT and the most cranial and caudal contour slices for each of 16 organs-at-risk for head-and-neck cancer as input, was developed. A CT-only model served as control. The models were evaluated with Dice similarity coefficient, Hausdorff Distance 95th percentile and average symmetric surface distance. A subset of 8 organs-at-risk were selected for a feasibility test. In this, a designated radiation oncologist used both single-cycle interactive segmentation and atlas-based auto-contouring for three cases. Contouring time and added path length were recorded. Results: The medians of Dice coefficients increased with single-cycle interactive segmentation in the range of 0.004 (Brain)-0.90 (EyeBack_merged) when compared to CT-only. In the feasibility test, contouring time and added path length were reduced for all three cases as compared to editing atlas-based auto-segmentations. Conclusion: Single-cycle interactive segmentation improved segmentation metrics when compared to the CT-only model and was clinically feasible from a technical and usability point of view. The study suggests that it may be cost-effective to add a small amount of contouring input to deep learning-based segmentation models.
RESUMEN
(1) The current literature contains several studies investigating the correlation between dual-energy-derived iodine concentration (IC) and positron emission tomography (PET)-derived Flourodeoxyglucose (18F-FDG) uptake in patients with non-small-cell lung cancer (NSCLC). In previously published studies, either the entire tumor volume or a region of interest containing the maximum IC or 18F-FDG was assessed. However, the results have been inconsistent. The objective of this study was to correlate IC with FDG both within the entire volume and regional sub-volumes of primary tumors in patients with NSCLC. (2) In this retrospective study, a total of 22 patients with NSCLC who underwent both dual-energy CT (DE-CT) and 18F-FDG PET/CT were included. A region of interest (ROI) encircling the entire primary tumor was delineated, and a rigid registration of the DE-CT, iodine maps and FDG images was performed for the ROI. The correlation between tumor measurements and area-specific measurements of ICpeak and the peak standardized uptake value (SUVpeak) was found. Finally, a correlation between tumor volume and the distance between SUVpeak and ICpeak centroids was found. (3) For the entire tumor, moderate-to-strong correlations were found between SUVmax and ICmax (R = 0.62, p = 0.002), and metabolic tumor volume vs. total iodine content (R = 0.91, p < 0.001), respectively. For local tumor sub-volumes, a negative correlation was found between ICpeak and SUVpeak (R = −0.58, p = 0.0046). Furthermore, a strong correlation was found between the tumor volume and the distance in millimeters between SUVpeak and ICpeak centroids (R = 0.81, p < 0.0001). (4) In patients with NSCLC, high FDG uptakes and high DE-CT-derived iodine concentrations correlated on a whole-tumor level, but the peak areas were positioned at different locations within the tumor. 18F-FDG PET/CT and DE-CT provide complementary information and might represent different underlying patho-physiologies.