Automated Society of Fetal Urology (SFU) grading of hydronephrosis on ultrasound imaging using a convolutional neural network

Ostrowski, David Alfred; Logan, J.; Antony, Maria; Broms, Reilly; Weiß, Dana; Batavia, Jason P. Van; Long, Christopher J.; Smith, Ariana L.; Zderic, Stephen A.; Edwins, Rebecca C.; Pominville, Raymond J.; Hannick, Jessica H.; Woo, Lynn L.; Fan, Yong; Tasian, Gregory E.; Weaver, John

doi:10.1016/j.jpurol.2023.05.014

Cited by 10 publications

(5 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Artificial intelligence shows promise in clinical medicine and could enhance diagnostic processes in the future [47]. However, further research is necessary before AI can be clinically applied to streamline the diagnostic paths of antenatal hydronephrosis [48][49][50]. Nevertheless, contemporary scoring systems based on AI exhibit potential and will no doubt improve future diagnostic processes through further refinement and iterative testing [51].…”

Section: Discussionmentioning

confidence: 99%

Stratifying Antenatal Hydronephrosis: Predicting High-Grade VUR Using Ultrasound and Scintigraphy

Pakkasjärvi,

Belov,

Jahnukainen

et al. 2024

Diagnostics

View full text Add to dashboard Cite

(1) Background: Antenatal hydronephrosis (AHN), detected in approximately one percent of prenatal ultrasounds, is caused by vesicoureteral reflux (VUR) in 15–21% of cases, a condition with significant risks such as urinary tract infections and renal scarring. Our study addresses the diagnostic challenges of VUR in AHN. Utilizing renal ultrasonography and scintigraphy, we developed a novel scoring system that accurately predicts high-grade VUR, optimizing diagnostic precision while minimizing the need for more invasive methods like voiding cystourethrogram (VCUG); (2) Methods: This retrospective study re-analyzed renal ultrasonography, scintigraphy, and VCUG images from infants admitted between 2003 and 2013, excluding cases with complex urinary anomalies; (3) Results: Our analysis included 124 patients (75% male), of whom 11% had high-grade VUR. The multivariate analysis identified visible ureter, reduced renal length, and decreased differential renal function (DRF) as primary predictors. Consequently, we established a three-tier risk score, classifying patients into low, intermediate, and high-risk groups for high-grade VUR, with corresponding prevalences of 2.3%, 22.2%, and 75.0%. The scoring system demonstrated 86% sensitivity and 79% specificity; (4) Conclusions: Our scoring system, focusing on objective parameters of the visible ureter, renal length, and DRF, effectively identifies high-grade VUR in AHN patients. This method enhances diagnostics in ANH by reducing reliance on VCUG and facilitating more tailored and less invasive patient care.

show abstract

Section: Discussionmentioning

confidence: 99%

Stratifying Antenatal Hydronephrosis: Predicting High-Grade VUR Using Ultrasound and Scintigraphy

Pakkasjärvi,

Belov,

Jahnukainen

et al. 2024

Diagnostics

View full text Add to dashboard Cite

show abstract

“…There are various studies in the diagnosis of hydronephrosis using artificial intelligence (AI)/machine learning (ML). Both deep learning and conventional radiomics analysis have been employed in these studies to detect renal abnormalities and predict hydronephrosis grade and severity [ 11 12 13 14 15 16 17 ]. Cerrolaza et al [ 11 ] performed quantitative imaging analysis on US in predicting obstructive severity in children with hydronephrosis.…”

Section: Introductionmentioning

confidence: 99%

“…This ratio was used as a biomarker for hydronephrosis severity prediction with high accuracy [ 14 ]. Ostrowski et al [ 15 ] developed an automated convolutional neural network model based on a VGG16 model pre-trained on ImageNet to classify hydronephrosis grade. The proposed ML model achieved an overall accuracy of 82.0% with three-fold cross-validation on 710 patients [ 15 ].…”

Section: Introductionmentioning

confidence: 99%

“…Ostrowski et al [ 15 ] developed an automated convolutional neural network model based on a VGG16 model pre-trained on ImageNet to classify hydronephrosis grade. The proposed ML model achieved an overall accuracy of 82.0% with three-fold cross-validation on 710 patients [ 15 ]. Lien et al [ 16 ] built three deep-learning models for hydronephrosis detection.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Pilot study of machine learning in the task of distinguishing high and low-grade pediatric hydronephrosis on ultrasound

Sloan,

Li,

Lescay

et al. 2023

Investig Clin Urol

View full text Add to dashboard Cite

Purpose Hydronephrosis is a common pediatric urological condition, characterized by dilation of the renal collecting system. Accurate identification of the severity of hydronephrosis is crucial in clinical management, as high-grade hydronephrosis can cause significant damage to the kidney. In this pilot study, we demonstrate the feasibility of machine learning in differentiating between high and low-grade hydronephrosis in pediatric patients. Materials and Methods We retrospectively reviewed 592 images from 90 unique patients ages 0–8 years diagnosed with hydronephrosis at the University of Chicago’s Pediatric Urology Clinic. The study included 74 high-grade hydronephrosis (145 images) and 227 low-grade hydronephrosis (447 images). Patients were excluded if they had less than 2 studies prior to surgical intervention or had structural abnormalities. We developed a radiomic-based artificial intelligence algorithm incorporating computerized texture analysis and machine learning (support-vector machine) to yield a predictor of hydronephrosis grade. Results Receiver operating characteristic analysis of the classifier output yielded an area under the curve value of 0.86 (95% CI 0.81–0.92) in the task of distinguishing between low and high-grade hydronephrosis using a five-fold cross-validation by kidney. In addition, a Mann–Kendall trend test between computer output and clinical hydronephrosis grade yielded a statistically significant upward trend (p<0.001). Conclusions Our findings demonstrate the potential of machine learning in the differentiation between low and high-grade hydronephrosis. Further studies are warranted to validate our findings and their generalizability for use in clinical practice as a means to predict clinical outcomes and the resolution of hydronephrosis.

show abstract

“…Artificial neural networks (ANN) are the building blocks of AI technologies, which simulate the human brain's analyzing and processing abilities to solve complex problems. The unique characteristics of ANN (such as efficient data handling, low complexity, reduced computation, and storage requirements) have enormous potential for a wide range of disciplines, including medical sciences [1] (especially in the areas of cardiology [2], radiology [3], oncology [4], urology [5]), veterinary [6],…”

Section: Introductionmentioning

confidence: 99%

Neural Networks for the Detection of COVID-19 and Other Diseases: Prospects and Challenges

et al. 2023

View full text Add to dashboard Cite

Artificial neural networks (ANNs) ability to learn, correct errors, and transform a large amount of raw data into beneficial medical decisions for treatment and care has increased in popularity for enhanced patient safety and quality of care. Therefore, this paper reviews the critical role of ANNs in providing valuable insights for patients’ healthcare decisions and efficient disease diagnosis. We study different types of ANNs in the existing literature that advance ANNs’ adaptation for complex applications. Specifically, we investigate ANNs’ advances for predicting viral, cancer, skin, and COVID-19 diseases. Furthermore, we propose a deep convolutional neural network (CNN) model called ConXNet, based on chest radiography images, to improve the detection accuracy of COVID-19 disease. ConXNet is trained and tested using a chest radiography image dataset obtained from Kaggle, achieving more than 97% accuracy and 98% precision, which is better than other existing state-of-the-art models, such as DeTraC, U-Net, COVID MTNet, and COVID-Net, having 93.1%, 94.10%, 84.76%, and 90% accuracy and 94%, 95%, 85%, and 92% precision, respectively. The results show that the ConXNet model performed significantly well for a relatively large dataset compared with the aforementioned models. Moreover, the ConXNet model reduces the time complexity by using dropout layers and batch normalization techniques. Finally, we highlight future research directions and challenges, such as the complexity of the algorithms, insufficient available data, privacy and security, and integration of biosensing with ANNs. These research directions require considerable attention for improving the scope of ANNs for medical diagnostic and treatment applications.

show abstract

Automated Society of Fetal Urology (SFU) grading of hydronephrosis on ultrasound imaging using a convolutional neural network

Cited by 10 publications

References 26 publications

Stratifying Antenatal Hydronephrosis: Predicting High-Grade VUR Using Ultrasound and Scintigraphy

Stratifying Antenatal Hydronephrosis: Predicting High-Grade VUR Using Ultrasound and Scintigraphy

Pilot study of machine learning in the task of distinguishing high and low-grade pediatric hydronephrosis on ultrasound

Neural Networks for the Detection of COVID-19 and Other Diseases: Prospects and Challenges

Contact Info

Product

Resources

About