Search

Item does not contain fulltext

Contains fulltext : 205458.pdf (publisher's version ) (Open Access), The approximity of the inferior alveolar nerve (IAN) to the roots of lower third molars (M3) is a risk factor for the occurrence of nerve damage and subsequent sensory disturbances of the lower lip and chin following the removal of third molars. To assess this risk, the identification of M3 and IAN on dental panoramic radiographs (OPG) is mandatory. In this study, we developed and validated an automated approach, based on deep-learning, to detect and segment the M3 and IAN on OPGs. As a reference, M3s and IAN were segmented manually on 81 OPGs. A deep-learning approach based on U-net was applied on the reference data to train the convolutional neural network (CNN) in the detection and segmentation of the M3 and IAN. Subsequently, the trained U-net was applied onto the original OPGs to detect and segment both structures. Dice-coefficients were calculated to quantify the degree of similarity between the manually and automatically segmented M3s and IAN. The mean dice-coefficients for M3s and IAN were 0.947 +/- 0.033 and 0.847 +/- 0.099, respectively. Deep-learning is an encouraging approach to segment anatomical structures and later on in clinical decision making, though further enhancement of the algorithm is advised to improve the accuracy.

Contains fulltext : 204008.pdf (publisher's version ) (Open Access)

Contains fulltext : 205463.pdf (publisher's version ) (Open Access), Importance: Nephron-sparing surgery can be considered in well-defined cases of unilateral and bilateral Wilms tumors, but the surgical procedure can be very challenging for the pediatric surgeon to perform. Objective: To assess the added value of personalized 3-dimensional (3-D) kidney models derived from conventional imaging data to enhance preoperative surgical planning. Design, Setting, and Participants: In a survey study, the conventional imaging data of 10 Dutch children with Wilms tumors were converted to 3-D prints and augmented reality (AR) holograms and a panel of pediatric oncology surgeons (n = 7) assessed the quality of the different imaging methods during preoperative evaluation. Kidney models were created with 3-D printing and AR using a mixed reality headset for visualization. Main Outcomes and Measures: Differences in the assessment of 4 anatomical structures (tumor, arteries, veins, and urinary collecting structures) using questionnaires. A Likert scale measured differences between the imaging methods, with scores ranging from 1 (completely disagree) to 5 (completely agree). Results: Of the 10 patients, 7 were girls, and the mean (SD) age was 3.7 (1.7) years. Compared with conventional imaging, the 3-D print and the AR hologram models were evaluated by the surgeons to be superior for all anatomical structures: tumor (median scores for conventional imaging, 4.07; interquartile range [IQR], 3.62-4.15 vs 3-D print, 4.67; IQR, 4.14-4.71; P = .008 and AR hologram, 4.71; IQR, 4.26-4.75; P = .002); arteries (conventional imaging, 3.62; IQR, 3.43-3.93 vs 3-D print, 4.54; IQR, 4.32-4.71; P = .002 and AR hologram, 4.83; IQR, 4.64-4.86; P < .001), veins (conventional imaging, 3.46; IQR 3.39-3.62 vs 3-D print, 4.50; IQR, 4.39-4.68; P < .001 and AR hologram, 4.83; IQR, 4.71-4.86; P < .001), and urinary collecting structures (conventional imaging, 2.76; IQR, 2.42-3.00 vs 3-D print, 3.86; IQR, 3.64-4.39; P < .001 and AR hologram, 4.00; IQR, 3.93-4.58; P < .001). T

Item does not contain fulltext

Contains fulltext : 182317.pdf (publisher's version ) (Open Access)