Urologist validation of an artificial intelligence-based tool for automated estimation of penile curvature.

INTRODUCTION: Severity of penile curvature (PC) is commonly used to select the optimal surgical intervention for hypospadias, either alone or in conjunction with other phenotypic characteristics. Despite this, current literature on the accuracy and precision of different PC measurement techniques in hypospadias patients remains limited. PURPOSE: Assess the feasibility and validity of an artificial intelligence (AI)-based model for automatic measurement of PC. MATERIAL AND METHODS: Seven 3D-printed penile models with variable degrees of ventral PC were used to evaluate and compare interobserver agreement in estimation of penile curvatures using various measurement techniques (including visual inspection, goniometer, manual estimation via a mobile application, and an AI-based angle estimation app. In addition, each participant was required to complete a questionnaire about their background and experience. RESULTS: Thirty-five clinical practitioners participated in the study, including pediatric urologists, pediatric surgeons, and urologists. For each PC assessment method, time required, mean absolute error (MAE), and inter-rater agreement were assessed. For goniometer-based measurement, the lowest MAE achieved was derived from a model featuring 86° PC. When using either UVI (unaid visual inspection), mobile apps, or AI-based measurement, MAE was lowest when assessing a model with 88° PC, indicating that high-grade cases can be quantified more reliably. Indeed, MAE was highest when PC angle ranged between 40° and 58° for all the investigated measurement tools. In fact, among these methodologies, AI-based assessment achieved the lowest MAE and highest level of inter-class correlation, with an average measurement time of only 22 s. CONCLUSION: AI-based PC measurement models are more practical and consistent than the alternative curvature assessment tools already available. The AI method described in this study could help surgeons and hypospadiology researchers to measure PC more accurately.

Perturbation of deep autoencoder weights for model compression and classification of tabular data.

Fully connected deep neural networks (DNN) often include redundant weights leading to overfitting and high memory requirements. Additionally, in tabular data classification, DNNs are challenged by the often superior performance of traditional machine learning models. This paper proposes periodic perturbations (prune and regrow) of DNN weights, especially at the self-supervised pre-training stage of deep autoencoders. The proposed weight perturbation strategy outperforms dropout learning or weight regularization (L1 or L2) for four out of six tabular data sets in downstream classification tasks. Unlike dropout learning, the proposed weight perturbation routine additionally achieves 15% to 40% sparsity across six tabular data sets, resulting in compressed pretrained models. The proposed pretrained model compression improves the accuracy of downstream classification, unlike traditional weight pruning methods that trade off performance for model compression. Our experiments reveal that a pretrained deep autoencoder with weight perturbation can outperform traditional machine learning in tabular data classification, whereas baseline fully-connected DNNs yield the worst classification accuracy. However, traditional machine learning models are superior to any deep model when a tabular data set contains uncorrelated variables. Therefore, the performance of deep models with tabular data is contingent on the types and statistics of constituent variables.

Missing Value Estimation using Clustering and Deep Learning within Multiple Imputation Framework.

Missing values in tabular data restrict the use and performance of machine learning, requiring the imputation of missing values. The most popular imputation algorithm is arguably multiple imputations using chains of equations (MICE), which estimates missing values from linear conditioning on observed values. This paper proposes methods to improve both the imputation accuracy of MICE and the classification accuracy of imputed data by replacing MICE's linear regressors with ensemble learning and deep neural networks (DNN). The imputation accuracy is further improved by characterizing individual samples with cluster labels (CISCL) obtained from the training data. Our extensive analyses involving six tabular data sets, up to 80% missing values, and three missing types (missing completely at random, missing at random, missing not at random) reveal that ensemble or deep learning within MICE is superior to the baseline MICE (b-MICE), both of which are consistently outperformed by CISCL. Results show that CISCL + b-MICE outperforms b-MICE for all percentages and types of missingness. Our proposed DNN-based MICE and gradient boosting MICE plus CISCL (GB-MICE-CISCL) outperform seven state-of-the-art imputation algorithms in most experimental cases. The classification accuracy of GB-MICE imputed data is further improved by our proposed GB-MICE-CISCL imputation method across all missingness percentages. Results also reveal a shortcoming of the MICE framework at high missingness (>50%) and when the missing type is not random. This paper provides a generalized approach to identifying the best imputation model for a data set with a missingness percentage and type.