OBJECTIVETo compare the diagnostic properties of a nonmydriatic 200° ultra-widefield scanning laser ophthalmoscope (SLO) versus mydriatic Early Treatment of Diabetic Retinopathy Study (ETDRS) 7-field photography for diabetic retinopathy (DR) screening.RESEARCH DESIGN AND METHODSA consecutive series of 212 eyes of 141 patients with different levels of DR were examined. Grading of DR and clinically significant macular edema (CSME) from mydriatic ETDRS 7-field stereo photography was compared with grading obtained by Optomap Panoramic 200 SLO images. All SLO scans were performed through an undilated pupil, and no additional clinical information was used for evaluation of all images by the two independent, masked, expert graders.RESULTSTwenty-two eyes from ETDRS 7-field photography and 12 eyes from Optomap were not gradable by at least one grader because of poor image quality. A total of 144 eyes were analyzed regarding DR level and 155 eyes regarding CSME. For ETDRS 7-field photography, 22 eyes (18 for grader 2) had no or mild DR (ETDRS levels ≤ 20) and 117 eyes (111 for grader 2) had no CSME. A highly substantial agreement between both Optomap DR and CSME grading and ETDRS 7-field photography existed with κ = 0.79 for DR and 0.73 for CSME for grader 1, and κ = 0.77 (DR) and 0.77 (CSME) for grader 2.CONCLUSIONSDetermination of CSME and grading of DR level from Optomap Panoramic 200 nonmydriatic images show a positive correlation with mydriatic ETDRS 7-field stereo photography. Both techniques are of sufficient quality to assess DR and CSME. Optomap Panoramic 200 images cover a larger retinal area and therefore may offer additional diagnostic properties.

Background: There is limited data on the use of targeted or immunotherapy (TT/IT) in combination with single fraction stereotactic radiosurgery (SRS) in patients with melanoma brain metastasis (MBM). Therefore, we analyzed the outcome and toxicity of SRS alone compared to SRS in combination with TT/IT. Methods: Patients with MBM treated with single session SRS at our department between 2014 and 2017 with a minimum follow-up of 3 months after first SRS were included. The primary endpoint of this study was local control (LC). Secondary endpoints were distant intracranial control, radiation necrosis-free survival (RNFS), and overall survival (OS). The local/ distant intracranial control rates, RNFS and OS were analyzed using the Kaplan-Meier method. The log-rank test was used to test differences between groups. Cox proportional hazard model was performed for univariate continuous variables and multivariate analyses. Results: Twenty-eight patients (17 male and 11 female) with 52 SRS-lesions were included. The median follow-up was 19 months (range 14-24 months) after first SRS. Thirty-six lesions (69.2%) were treated with TT/IT simultaneously (4 weeks before and 4 weeks after SRS), while 16 lesions (30.8%) were treated with SRS alone or with sequential TT/ IT. The 1-year local control rate was 100 and 83.3% for SRS with TT/IT and SRS alone (p = 0.023), respectively. The estimated 1-year RNFS was 90.0 and 82.1% for SRS in combination with TT/IT and SRS alone (p = 0.935). The distant intracranial control rate after 1 year was 47.7 and 50% for SRS in combination with TT/IT and SRS alone (p = 0.933). On univariate analysis, the diagnosis-specific Graded Prognostic Assessment including the BRAF status (Melanoma-molGPA) was associated with a significantly improved LC. Neither gender nor SRS-PTV margin had a prognostic impact on LC. V10 and V12 were significantly associated with RNFS (p < 0.001 and p = 0.004). Conclusion: SRS with simultaneous TT/IT significantly improved LC with no significant difference in radiation necrosis rate. The therapy combination appears to be effective and safe. However, prospective studies on SRS with simultaneous TT/IT are necessary and ongoing. Trial registration: The institutional review board approved this analysis on 10th of February 2015 and all patients signed informed consent (UE nr. 128-14).

Background In this dosimetric study, a dedicated planning tool for single isocenter stereotactic radiosurgery for multiple brain metastases using dynamic conformal arc therapy (DCAT) was compared to standard volumetric modulated arc therapy (VMAT). Methods Twenty patients with a total of 66 lesions who were treated with the DCAT tool were included in this study. Single fraction doses of 15–20 Gy were prescribed to each lesion. Patients were re-planned using non-coplanar VMAT. Number of monitor units as well as V 4Gy , V 5Gy and V 8Gy were extracted for every plan. Using a density-based clustering algorithm, V 10Gy and V 12Gy and the volume receiving half of the prescribed dose were extracted for every lesion. Gradient indices and conformity indices were calculated. The correlation of the target sphericity, a measure of how closely the shape of the target PTV resembles a sphere, to the difference in V 10Gy and V 12Gy between the two techniques was assessed using Spearman’s correlation coefficient. Results The automated DCAT planning tool performed significantly better in terms of all investigated metrics ( p < 0.05), in particular healthy brain sparing (V 10Gy : median 3.2 cm 3 vs. 4.9 cm 3 ), gradient indices (median 5.99 vs. 7.17) and number of monitor units (median 4569 vs. 5840 MU). Differences in conformity indices were minimal (median 0.75 vs. 0.73) but still significant ( p < 0.05). A moderate correlation between PTV sphericity and the difference of V 10Gy and V 12Gy between the two techniques was found (Spearman’s rho = 0.27 and 0.30 for V 10Gy and V 12Gy , respectively, p < 0.05). Conclusions The dedicated DCAT planning tool performed better than VMAT in terms of healthy brain sparing and treatment efficiency, in particular for nearly spherical lesions. In contrast, VMAT can be superior in cases with irregularly shaped lesions.

No abstract

Background: Online adaptive radiation therapy (RT) using hybrid magnetic resonance linear accelerators (MR-Linacs) can administer a tailored radiation dose at each treatment fraction. Daily MR imaging followed by organ and target segmentation adjustments allow to capture anatomical changes, improve target volume coverage, and reduce the risk of side effects. The introduction of automatic segmentation techniques could help to further improve the online adaptive workflow by shortening the re-contouring time and reducing intra-and inter-observer variability. In fractionated RT, prior knowledge, such as planning images and manual expert contours, is usually available before irradiation, but not used by current artificial intelligence-based autocontouring approaches. Purpose: The goal of this study was to train convolutional neural networks (CNNs) for automatic segmentation of bladder, rectum (organs at risk, OARs), and clinical target volume (CTV) for prostate cancer patients treated at 0.35 T MR-Linacs. Furthermore, we tested the CNNs generalization on data from independent facilities and compared them with the MR-Linac treatment planning system (TPS) propagated structures currently used in clinics. Finally, expert planning delineations were utilized for patient-(PS) and facility-specific (FS) transfer learning to improve auto-segmentation of CTV and OARs on fraction images. Methods: In this study, data from fractionated treatments at 0.35 T MR-Linacs were leveraged to develop a 3D U-Net-based automatic segmentation. Cohort C1 had 73 planning images and cohort C2 had 19 planning and 240 fraction images. The baseline models (BMs) were trained solely on C1 planning data using 53 MRIs for training and 10 for validation. To assess their accuracy, the models were tested on three data subsets: (i) 10 C1 planning images not used for training, (ii) 19 C2 planning, and (iii) 240 C2 fraction images. BMs also served as a starting point for FS and PS transfer learning, where the planning images from C2 were used for network parameter fine tuning. The segmentation output of the different trained models was compared against expert ground truth by means of geometric metrics. Moreover, a trained physician graded the network segmentations as well as the segmentations propagated by the clinical TPS.