Posterior Urethral Valves Outcomes Prediction (PUVOP): a machine learning tool to predict clinically relevant outcomes in boys with posterior urethral valves

Artificial Intelligence Tools in Pediatric Urology: A Comprehensive Review of Recent Advances

Artificial intelligence (AI) is providing novel answers to long-standing clinical problems, and it is quickly changing pediatric urology. This thorough analysis focuses on current developments in AI technologies that improve pediatric urology diagnosis, treatment planning, and surgery results. Deep learning algorithms help detect problems with previously unheard-of precision in disorders including hydronephrosis, pyeloplasty, and vesicoureteral reflux, where AI-powered prediction models have demonstrated promising outcomes in boosting diagnostic accuracy. AI-enhanced image processing methods have significantly improved the quality and interpretation of medical images. Examples of these methods are deep-learning-based segmentation and contrast limited adaptive histogram equalization (CLAHE). These methods guarantee higher precision in the identification and classification of pediatric urological disorders, and AI-driven ground truth construction approaches aid in the standardization of and improvement in training data, resulting in more resilient and consistent segmentation models. AI is being used for surgical support as well. AI-assisted navigation devices help with difficult operations like pyeloplasty by decreasing complications and increasing surgical accuracy. AI also helps with long-term patient monitoring, predictive analytics, and customized treatment strategies, all of which improve results for younger patients. However, there are practical, ethical, and legal issues with AI integration in pediatric urology that need to be carefully navigated. To close knowledge gaps, more investigation is required, especially in the areas of AI-driven surgical methods and standardized ground truth datasets for pediatric radiologic image segmentation. In the end, AI has the potential to completely transform pediatric urology by enhancing patient care, increasing the effectiveness of treatments, and spurring more advancements in this exciting area.

Long-Term Kidney Outcomes in Children with Posterior Urethral Valves: A Population-Based Cohort Study

Key Points

Background

Posterior urethral valves represent the most common cause of lower urinary tract obstruction in male infants (approximately 1/4000 live births). Long-term kidney outcomes of posterior urethral valves remain uncertain. We aimed to determine the time-varying risk of major adverse kidney events (MAKE) following children with posterior urethral valves into adulthood.

Methods

A population-based retrospective cohort study of all male children (<2 years) diagnosed with posterior urethral valves between 1991 and 2021 in Ontario, Canada. Comparator cohorts were (1) male general population and (2) male children with pyeloplasty (both <2 years). The primary outcome was MAKE (death, long-term KRT [dialysis or kidney transplant], or CKD). Time to MAKE was analyzed using multivariable-adjusted Cox proportional hazards models. We censored for provincial emigration or administrative censoring (March 31, 2022).

Results

We included 727 children with posterior urethral valves, 855 pyeloplasty comparators, and 1,013,052 general population comparators. The median follow-up time was 16.6 years (Q1–3, 8.6–24.5) overall. Throughout follow-up, 32% of children with posterior urethral valves developed MAKE versus 1% of the general population and 6% of pyeloplasty comparators. Their adjusted hazard ratio for MAKE was 36.6 (95% confidence interval, 31.6 to 42.4) versus the general population. The risk of developing MAKE declined over the first 5 years after posterior urethral valves diagnosis but remained elevated for >30-year follow-up. Children with posterior urethral valves were also at higher risk of death, CKD, long-term KRT, hypertension, and AKI than the general population or pyeloplasty comparators.

Conclusions

Children with posterior urethral valves are at higher risk of adverse long-term kidney outcomes well into adulthood.

The potential of the transformer-based survival analysis model, SurvTrace, for predicting recurrent cardiovascular events and stratifying high-risk patients with ischemic heart disease

INTRODUCTION

Ischemic heart disease is a leading cause of death worldwide, and its importance is increasing with the aging population. The aim of this study was to evaluate the accuracy of SurvTrace, a survival analysis model using the Transformer-a state-of-the-art deep learning method-for predicting recurrent cardiovascular events and stratifying high-risk patients. The model's performance was compared to that of a conventional scoring system utilizing real-world data from cardiovascular patients.

METHODS

This study consecutively enrolled patients who underwent percutaneous coronary intervention (PCI) at the Department of Cardiovascular Medicine, University of Tokyo Hospital, between 2005 and 2019. Each patient's initial PCI at our hospital was designated as the index procedure, and a composite of major adverse cardiovascular events (MACE) was monitored for up to two years post-index event. Data regarding patient background, clinical presentation, medical history, medications, and perioperative complications were collected to predict MACE. The performance of two models-a conventional scoring system proposed by Wilson et al. and the Transformer-based model SurvTrace-was evaluated using Harrell's c-index, Kaplan-Meier curves, and log-rank tests.

RESULTS

A total of 3938 cases were included in the study, with 394 used as the test dataset and the remaining 3544 used for model training. SurvTrace exhibited a mean c-index of 0.72 (95% confidence intervals (CI): 0.69-0.76), which indicated higher prognostic accuracy compared with the conventional scoring system's 0.64 (95% CI: 0.64-0.64). Moreover, SurvTrace demonstrated superior risk stratification ability, effectively distinguishing between the high-risk group and other risk categories in terms of event occurrence. In contrast, the conventional system only showed a significant difference between the low-risk and high-risk groups.

CONCLUSION

This study based on real-world cardiovascular patient data underscores the potential of the Transformer-based survival analysis model, SurvTrace, for predicting recurrent cardiovascular events and stratifying high-risk patients.

Bioinformatics in urology - molecular characterization of pathophysiology and response to treatment

The application of bioinformatics has revolutionized the practice of medicine in the past 20 years. From early studies that uncovered subtypes of cancer to broad efforts spearheaded by the Cancer Genome Atlas initiative, the use of bioinformatics strategies to analyse high-dimensional data has provided unprecedented insights into the molecular basis of disease. In addition to the identification of disease subtypes - which enables risk stratification - informatics analysis has facilitated the identification of novel risk factors and drivers of disease, biomarkers of progression and treatment response, as well as possibilities for drug repurposing or repositioning; moreover, bioinformatics has guided research towards precision and personalized medicine. Implementation of specific computational approaches such as artificial intelligence, machine learning and molecular subtyping has yet to become widespread in urology clinical practice for reasons of cost, disruption of clinical workflow and need for prospective validation of informatics approaches in independent patient cohorts. Solving these challenges might accelerate routine integration of bioinformatics into clinical settings.

Kidney disease progression in pediatric and adult posterior urethral valves (PUV) patients

BACKGROUND

Posterior urethral valves (PUV) is the most common cause of obstructive uropathy in boys; approximately 15% develop kidney failure by early adulthood. However, rates of kidney function decline are poorly defined in PUV children and adults, as is the impact of potentially modifiable chronic kidney disease (CKD) progression risk factors.

METHODS

We conducted a retrospective review of all PUV patients followed at our institution from 1995 to 2018. Inclusion criteria were estimated glomerular filtration rate (eGFR) > 20 ml/min/1.73 m2 after 1 year of age, no dialysis or kidney transplant history, and ≥ 2 yearly serum creatinine values after age 1 year. eGFRs were calculated using creatinine-based estimating formulas for children (CKID U25) or adults (CKD-EPI). The primary outcome was annualized change in eGFR, assessed with linear mixed effects models. We also examined the association of acute kidney injury (AKI), proteinuria, hypertension (HTN), and recurrent febrile urinary tract infections (UTIs) with eGFR decline.

RESULTS

Fifty-two PUV patients met the inclusion criteria. Median (interquartile range) eGFR decline was 2.6 (2.1, 3.1) ml/min/1.73 m2/year. Children (n = 35) and adults (n = 17) demonstrated progressive decline. Proteinuria and recurrent UTIs were significantly associated with faster progression; AKI and HTN were also associated but did not reach significance.

CONCLUSION

PUV patients show progressive loss of kidney function well into adulthood. Proteinuria and recurrent UTIs are associated with faster progression, suggesting potential modifiable risk factors. This is the first study to report annualized eGFR decline rates in PUV patients, which could help inform the design of clinical trials of CKD therapies.

Artificial Intelligence in Pediatric Urology

Application of artificial intelligence (AI) is one of the hottest topics in medicine. Unlike traditional methods that rely heavily on statistical assumptions, machine learning algorithms can identify highly complex patterns from data, allowing robust predictions. There is an abundance of evidence of exponentially increasing pediatric urologic publications using AI methodology in recent years. While these studies show great promise for better understanding of disease and patient care, we should be realistic about the challenges arising from the nature of pediatric urologic conditions and practice, in order to continue to produce high-impact research.

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

INTRODUCTION

Grading of hydronephrosis severity on postnatal renal ultrasound guides management decisions in antenatal hydronephrosis (ANH). Multiple systems exist to help standardize hydronephrosis grading, yet poor inter-observer reliability persists. Machine learning methods may provide tools to improve the efficiency and accuracy of hydronephrosis grading.

OBJECTIVE

To develop an automated convolutional neural network (CNN) model to classify hydronephrosis on renal ultrasound imaging according to the Society of Fetal Urology (SFU) system as potential clinical adjunct.

STUDY DESIGN

A cross-sectional, single-institution cohort of postnatal renal ultrasounds with radiologist SFU grading from pediatric patients with and without hydronephrosis of stable severity was obtained. Imaging labels were used to automatedly select sagittal and transverse grey-scale renal images from all available studies from each patient. A VGG16 pre-trained ImageNet CNN model analyzed these preprocessed images. Three-fold stratified cross-validation was used to build and evaluate the model that was used to classify renal ultrasounds on a per patient basis into five classes based on the SFU system (normal, SFU I, SFU II, SFU III, or SFU IV). These predictions were compared to radiologist grading. Confusion matrices evaluated model performance. Gradient class activation mapping demonstrated imaging features driving model predictions.

RESULTS

We identified 710 patients with 4659 postnatal renal ultrasound series. Per radiologist grading, 183 were normal, 157 were SFU I, 132 were SFU II, 100 were SFU III, and 138 were SFU IV. The machine learning model predicted hydronephrosis grade with 82.0% (95% CI: 75-83%) overall accuracy and classified 97.6% (95% CI: 95-98%) of the patients correctly or within one grade of the radiologist grade. The model classified 92.3% (95% CI: 86-95%) normal, 73.2% (95% CI: 69-76%) SFU I, 73.5% (95% CI: 67-75%) SFU II, 79.0% (95% CI: 73-82%) SFU III, and 88.4% (95% CI: 85-92%) SFU IV patients accurately. Gradient class activation mapping demonstrated that the ultrasound appearance of the renal collecting system drove the model's predictions.

DISCUSSION

The CNN-based model classified hydronephrosis on renal ultrasounds automatically and accurately based on the expected imaging features in the SFU system. Compared to prior studies, the model functioned more automatically with greater accuracy. Limitations include the retrospective, relatively small cohort, and averaging across multiple imaging studies per patient.

CONCLUSIONS

An automated CNN-based system classified hydronephrosis on renal ultrasounds according to the SFU system with promising accuracy based on appropriate imaging features. These findings suggest a possible adjunctive role for machine learning systems in the grading of ANH.

Posterior urethral valves, pressure pop-offs, and kidney function: systematic review and meta-analysis

PURPOSE

To determine the role of pressure pop-off mechanisms, including vesicoureteral reflux and renal dysplasia (VURD) syndrome, in determining long-term kidney outcomes in boys with posterior urethral valves (PUV).

METHODS

A systematic search was performed in December 2022. Descriptive and comparative studies with a defined pressure pop-off group were included. Assessed outcomes included end-stage renal disease (ESRD), kidney insufficiency (defined as chronic kidney disease [CKD] stage 3 + or SCr > 1.5 mg/dL), and kidney function. Pooled proportions and relative risks (RR) with 95% confidence intervals (CI) were extrapolated from available data for quantitative synthesis. Random-effects meta-analyses were performed according to the study design and techniques. The risk of bias was assessed with the QUIPS tool and GRADE quality of evidence. The systematic review was prospectively registered on PROSPERO (CRD42022372352).

RESULTS

A total of 15 studies describing 185 patients with a median follow-up of 6.8 years were included. By the last follow-up, overall effect estimates demonstrate the prevalence of CKD and ESRD to be 15.2% and 4.1%, respectively. There was no significant difference in the risk of ESRD in patients with pop-off compared to no pop-off patients [RR 0.34, 95%CI 0.12, 1.10; p = 0.07]. There was a significantly reduced risk for kidney insufficiency in boys with pop-off [RR 0.57, 95%CI 0.34, 0.97; p = 0.04], but this protective effect was not re-demonstrated after excluding studies with inadequate reporting of CKD outcomes [RR 0.63, 95%CI 0.36, 1.10; p = 0.10]. Included study quality was low, with 6 studies having moderate risk and 9 having a high risk of bias.

CONCLUSIONS

Pop-off mechanisms may be associated with reducing the risk of kidney insufficiency, but current certainty in the evidence is low. Further research is warranted to investigate sources of heterogeneity and long-term sequelae in pressure pop-offs.

Deep learning imaging features derived from kidney ultrasounds predict chronic kidney disease progression in children with posterior urethral valves

BACKGROUND

We sought to use deep learning to extract anatomic features from postnatal kidney ultrasounds and evaluate their performance in predicting the risk and timing of chronic kidney disease (CKD) progression for boys with posterior urethral valves (PUV). We hypothesized that these features would predict CKD progression better than clinical characteristics such as nadir creatinine alone.

METHODS

We performed a retrospective cohort study of boys with PUV treated at two pediatric health systems from 1990 to 2021. Features of kidneys were extracted from initial postnatal kidney ultrasound images using a deep learning model. Three time-to-event prediction models were built using random survival forests. The Imaging Model included deep learning imaging features, the Clinical Model included clinical data, and the Ensemble Model combined imaging features and clinical data. Separate models were built to include time-dependent clinical data that were available at 6 months, 1 year, 3 years, and 5 years.

RESULTS

Two-hundred and twenty-five patients were included in the analysis. All models performed well with C-indices of 0.7 or greater. The Clinical Model outperformed the Imaging Model at all time points with nadir creatinine driving the performance of the Clinical Model. Combining the 6-month Imaging Model (C-index 0.7; 95% confidence interval [CI] 0.6, 0.79) with the 6-month Clinical Model (C-index 0.79; 95% CI 0.71, 0.86) resulted in a 6-month Ensemble Model that performed better (C-index 0.82; 95% CI 0.77, 0.88) than either model alone.

CONCLUSIONS

Deep learning imaging features extracted from initial postnatal kidney ultrasounds may improve early prediction of CKD progression among children with PUV. A higher resolution version of the Graphical abstract is available as Supplementary information.

Chronic Kidney Disease in Boys with Posterior Urethral Valves-Pathogenesis, Prognosis and Management

Posterior urethral valves (PUV) are the most common form of lower urinary tract obstructions (LUTO). The valves can be surgically corrected postnatally; however, the impairment of kidney and bladder development is irreversible and has lifelong implications. Chronic kidney disease (CKD) and bladder dysfunction are frequent problems. Approximately 20% of PUV patients will reach end-stage kidney disease (ESKD). The subvesical obstruction in PUV leads to muscular hypertrophy and fibrotic remodelling in the bladder, which both impair its function. Kidney development is disturbed and results in dysplasia, hypoplasia, inflammation and renal fibrosis, which are hallmarks of CKD. The prognoses of PUV patients are based on prenatal and postnatal parameters. Prenatal parameters include signs of renal hypodysplasia in the analysis of fetal urine. Postnatally, the most robust predictor of PUV is the nadir serum creatinine after valve ablation. A value that is below 0.4 mg/dl implies a very low risk for ESKD, whereas a value above 0.85 mg/dl indicates a high risk for ESKD. In addition, bladder dysfunction and renal dysplasia point towards an unbeneficial kidney outcome. Experimental urinary markers such as MCP-1 and TGF-β, as well as microalbuminuria, indicate progression to CKD. Until now, prenatal intervention may improve survival but yields no renal benefit. The management of PUV patients includes control of bladder dysfunction and CKD treatment to slow down progression by controlling hypertension, proteinuria and infections. In kidney transplantation, aggressive bladder management is essential to ensure optimal graft survival.

Predictors of advanced chronic kidney disease in infancy after definitive vesicoamniotic shunting for congenital lower urinary tract obstruction

BACKGROUND

Severe congenital lower urinary tract obstruction (cLUTO) is associated with poor postnatal outcomes, including chronic and end stage kidney disease, and high mortality. Studies of the impact of fetal intervention through vesicoamniotic shunting are marred by a device malfunction rate of up to 60%. In this study, we delineate the postnatal course and infant kidney function following definitive urinary diversion in utero.

MATERIALS AND METHODS

This is a retrospective, single-center cohort study of 16 male infants who survived the fetal intervention to birth, from 2010 to 2014 at a single center. All had patent shunts in place at birth. Perinatal and biochemical characteristics were collected with patients followed for one year, or until demise, with serial measures of serum creatinine (SCr) and serum cystatin C (CysC).

RESULTS

Of the 16 males, 81% were non-white (38% black, 43% Hispanic). Shunts were placed at a median of 20 weeks (IQR 19,23) gestation, with median fetal bladder volume of 39 cm³ (IQR 9.9,65). All neonates were born preterm [median 34 weeks (IQR 31,35)] and the majority with low birth weight [median 2340 grams (1,895, 2,600)]. 63% required positive pressure ventilation. Advanced chronic kidney disease stage 4-5 at 1 year of age was predicted by neonatal characteristics: peak SCr ≥2 mg/dl, time to peak SCr > 6 days, discharge SCr ≥1.0 mg/dl, CysC ≥2.5 mg/l, urine protein:creatinine ≥4.8 mg/mg, urine microalbumin:creatinine ≥2.2 mg/mg. In infancy, a nadir SCr ≥0.5 mg/dl occurring before 160 days (5.3 months) of age was also predictive of advanced chronic kidney disease stage 4-5 at 1 year. Three patients died in the neonatal period, with 1 receiving kidney replacement therapy (KRT). Three additional patients required KRT before 12 months of age.

CONCLUSIONS

Even with definitive vesicoamniotic shunting for cLUTO, postnatal morbidity and mortality remain high, emphasizing the role of renal dysplasia, in spite of urinary diversion, in postnatal kidney dysfunction. Neonatal and infant biochemical parameters exhibit distinct trends that offer families and physicians a better understanding of the prognosis of childhood kidney function.