BackgroundSegmentation of the Gross Tumor Volume (GTV) is a crucial step in the brachytherapy (BT) treatment planning workflow. Currently, radiation oncologists segment the GTV manually, which is... Show moreBackgroundSegmentation of the Gross Tumor Volume (GTV) is a crucial step in the brachytherapy (BT) treatment planning workflow. Currently, radiation oncologists segment the GTV manually, which is time-consuming. The time pressure is particularly critical for BT because during the segmentation process the patient waits immobilized in bed with the applicator in place. Automatic segmentation algorithms can potentially reduce both the clinical workload and the patient burden. Although deep learning based automatic segmentation algorithms have been extensively developed for organs at risk, automatic segmentation of the targets is less common. The aim of this study was to automatically segment the cervical cancer GTV on BT MRI images using a state-of-the-art automatic segmentation framework and assess its performance.MethodsA cohort of 195 cervical cancer patients treated between August 2012 and December 2021 was retrospectively collected. A total of 524 separate BT fractions were included and the axial T2-weighted (T2w) MRI sequence was used for this project. The 3D nnU-Net was used as the automatic segmentation framework. The automatic segmentations were compared with the manual segmentations used for clinical practice with Sørensen–Dice coefficient (Dice), 95th Hausdorff distance (95th HD) and mean surface distance (MSD). The dosimetric impact was defined as the difference in D98 (ΔD98) and D90 (ΔD90) between the manual segmentations and the automatic segmentations, evaluated using the clinical dose distribution. The performance of the network was also compared separately depending on FIGO stage and on GTV volume.ResultsThe network achieved a median Dice of 0.73 (interquartile range (IQR) = 0.50–0.80), median 95th HD of 6.8 mm (IQR = 4.2–12.5 mm) and median MSD of 1.4 mm (IQR = 0.90–2.8 mm). The median ΔD90 and ΔD98 were 0.18 Gy (IQR = -1.38–1.19 Gy) and 0.20 Gy (IQR =-1.10–0.95 Gy) respectively. No significant differences in geometric or dosimetric performance were observed between tumors with different FIGO stages, however significantly improved Dice and dosimetric performance was found for larger tumors.ConclusionsThe nnU-Net framework achieved state-of-the-art performance in the segmentation of the cervical cancer GTV on BT MRI images. Reasonable median performance was achieved geometrically and dosimetrically but with high variability among patients. Show less
Barros, H.A. de; Duin, J.J.; Mulder, D.; Noort, V. van der; Noordzij, M.A.; Wit, E.M.K.; ... ; Poel, H.G. van der 2023
Background: Accurate identification of men who harbor nodal metastases is neces-sary to select patients who most likely benefit from whole pelvis radiotherapy (WPRT). Limited sensitivity of... Show moreBackground: Accurate identification of men who harbor nodal metastases is neces-sary to select patients who most likely benefit from whole pelvis radiotherapy (WPRT). Limited sensitivity of diagnostic imaging approaches for the detection of nodal micrometastases has led to the exploration of the sentinel lymph node biopsy (SLNB). Objective: To evaluate whether SLNB can be used as a tool to select pathologically node-positive patients who likely benefit from WPRT. Design, setting, and participants: We included 528 clinically node-negative primary prostate cancer (PCa) patients with an estimated nodal risk of >5% treated between 2007 and 2018. Intervention: A total of 267 patients were directly treated with prostate-only radio-therapy (PORT; non-SLNB group), while 261 patients underwent SLNB to remove lymph nodes directly draining from the primary tumor prior to radiotherapy (SLNB group); pN0 patients were treated with PORT, while pN1 patients were offered WPRT. Outcome measurements and statistical analysis: Biochemical recurrence-free survival (BCRFS) and radiological recurrence-free survival (RRFS) were compared using propensity score weighted (PSW) Cox proportional hazard models. Results and limitations: The median follow-up was 71 mo. Occult nodal metastases were found in 97 (37%) SLNB patients (median metastasis size: 2 mm). Adjusted 7-yr BCRFS rates were 81% (95% confidence interval [CI] 77-86%) in the SLNB group and 49% (95% CI 43-56%) in the non-SLNB group. The corresponding adjusted 7-yr RRFS rates were 83% (95% CI 78-87%) and 52% (95% CI 46-59%), respectively. In the PSW multivariable Cox regression analysis, SLNB was associated with improved BCRFS (hazard ratio [HR] 0.38, 95% CI 0.25-0.59, p < 0.001) and RRFS (HR 0.44, 95% CI 0.28-0.69, p < 0.001). Limitations include the bias inherent to the study's Conclusions: SLNB-based selection of pN1 PCa patients for WPRT was associated efit from the addition of pelvis radiotherapy. This strategy results in a longer dura (c) 2023 The Author(s). Published by Elsevier B.V. on behalf of European Association of Urology. This is an open access article under the CC BY license (http://creativecommons. Show less
