RESUMO
Background: Voiding cystourethrography (VCUG) is the gold standard for the diagnosis and grading of vesicoureteral reflux (VUR). However, VUR grading from voiding cystourethrograms is highly subjective with low reliability. This study aimed to develop a deep learning model to improve reliability for VUR grading on VCUG and compare its performance to that of clinicians. Methods: In this retrospective study in China, VCUG images were collected between January 2019 and September 2022 from our institution as an internal dataset for training and 4 external data sets as external testing set for validation. Samples were divided into training (N = 1000) and validation sets (N = 500), internal testing set (N = 168), and external testing set (N = 280). An ensemble learning-based model, Deep-VCUG, using Res-Net 101 and the voting methods was developed to predict VUR grade. The grading performance was assessed using heatmaps, area under the receiver operating characteristic curve (AUC), sensitivity, specificity, accuracy, and F1 score in the internal and external testing set. The performances of four clinicians (2 pediatric urologists and 2 radiologists) with and without the Deep-VCUG assisted to predict VUR grade were explored in external testing sets. Findings: A total of 1948 VCUG images were collected (Internal dataset = 1668; multi-center external dataset = 280). For assessing unilateral VUR grading, the Deep-VCUG achieved AUCs of 0.962 (95% confidence interval [CI]: 0.943-0.978) and 0.944 (95% [CI]: 0.921-0.964) in the internal and external testing sets, respectively, for bilateral VUR grading, the Deep-VCUG also achieved high AUCs of 0.960 (95% [CI]: 0.922-0.983) and 0.924 (95% [CI]: 0.887-0.957). The Deep-VCUG model using voting method outperformed single model and clinician in terms of classification based on VCUG image. Moreover, Under the Dee-VCUG assisted, the classification ability of junior and senior clinicians was significantly improved. Interpretation: The Deep-VCUG model is a generalizable, objective, and accurate tool for vesicoureteral reflux grading based on VCUG imaging and had good assistance with clinicians to VUR grading applicability. Funding: This study was supported by Natural Science Foundation of China, "Fuqing Scholar" Student Scientific Research Program of Shanghai Medical College, Fudan University, and the Program of Greater Bay Area Institute of Precision Medicine (Guangzhou).
RESUMO
INTRODUCTION AND OBJECTIVE: Accurate and objective assessment of penile curvature is considered a critical evaluation in patients with hypospadias, as it often determines if a 1 or 2-stage procedure should be done. Due to the ease of acquisition and reusability of digital images, more research is focused on digital images; however, the current method based on digital images is not an easily accurate and objective evaluation of penile curvature amongst surgeons. In scoliosis, the Cobb method is a standard method to quantify spinal curvature. Therefore, this study introduces a new accurate, and standardized method for measuring penile curvature based on the digital image concerning the Cobb method. METHODS: Twenty-two subjects were randomly selected, including 11 pediatric urologists with experience in goniometry(professional group)and 11 non-pediatric urologists without experience in goniometry (non-professional group). A total of 9 two-dimensional images of penile curvatures from 10° to 90°were obtained and stored in the research project notebook. Subjects measured 9 digital images using the new method (fixed anatomical position method) and classical method (the angle created by the interception of two ideal lines, one passing along the proximal portion of the corpora and a second passing through the tip of the penis), respectively. Measurement error was calculated as the absolute difference between the true curvature and the subject estimation. A t-test was used to evaluate the significant differences between the methods. RESULTS: A total of 22 subject measurement data were obtained. Mean errors using the new method ranged from 1.06° to 3.50°, compared to 3.84°to 11.83°for classical method. Mean errors were significantly lower (p < 0.05) when using the new method compared to the classical method. A subgroup comparing subjects with and without prior experience with goniometry showed a statistically significant difference only for the classical method when the penis curvature is 10-40°, the mean error range of the professional group was 7.8°-9.56°, compared to 10.34°-13.02°for nonprofessional group. DISCUSSION: We emphasize the importance of penile curvature measurement and urgent need for an accurate measurement method, and then we focus on the new method compare with the previously described methods looking at mean errors and explain the reason that the new method why is accurate. Subsequently, we focus on explain the impact of experience measurement methods. Finally, the shortcomings of this paper and the prospective points are discussed:1) how to obtain more photos in practical situations; 2) using artificial intelligence methods for automatic marking of key points to achieve efficient measurement of penile curvature. CONCLUSIONS: In this preliminary study, we demonstrated better penile curvature estimations using the new method compared to the classical methods currently used by pediatric urologists.