Machine learning based prediction models in male reproductive health: Development of a proof‐of‐concept model for Klinefelter Syndrome in azoospermic patients

Abstract

Background: Due to the highly variable clinical phenotype, Klinefelter Syndrome is underdiagnosed.Objective: Assessment of supervised machine learning based prediction models for identification of Klinefelter Syndrome among azoospermic patients, and comparison to expert clinical evaluation. Materials and methods:Retrospective patient data (karyotype, age, height, weight, testis volume, follicle-stimulating hormone, luteinizing hormone, testosterone, estradiol, prolactin, semen pH and semen volume) collected between January 2005 and June 2019 were retrieved from a patient data bank of a University Centre. Models were trained, validated and benchmarked based on different supervised machine learning algorithms. Models were then tested on an independent, prospectively acquired set of patient data (between July 2019 and July 2020). Benchmarking against physicians was performed in addition.Results: Based on average performance, support vector machines and CatBoost were particularly well-suited models, with 100% sensitivity and >93% specificity on the test dataset. Compared to a group of 18 expert clinicians, the machine learning models had significantly better median sensitivity (100% vs. 87.5%, p = 0.0455) and fared comparably with regards to specificity (90% vs. 89.9%, p = 0.4795), thereby possibly improving diagnosis rate. A Klinefelter Syndrome Score Calculator based on the prediction models is available on http://klinefelter-score-calculator.uni-muenster.de.Discussion: Differentiating Klinefelter Syndrome patients from azoospermic patients with normal karyotype (46,XY) is a problem that can be solved with supervised machine learning techniques, improving patient care.Conclusions: Machine learning could improve the diagnostic rate of Klinefelter Syndrome among azoospermic patients, even more for less-experienced physicians.

Biochemical identification of prepubertal boys with Klinefelter syndrome by combined reproductive hormone profiling using machine learning

“…Although the control and KS clinical profiles were not arranged as disjoined clusters in the PCA biplot or satisfactorily distinguishable by singular variable ROC analyses, we proposed that a decision tree-based ML approach could achieve an acceptable classification accuracy. In line with this, a recent study evaluated a model based on age, TV, semen volume, height, and concentrations of LH and FSH for identifying patients with KS among patients with azoospermia (13). The model was based on data from a total of 1339 azoospermic patients including 345 patients with KS, and the sensitivity of that model was 100% and the specificity >93%, which was significantly better compared to 18 expert clinicians.…”

“…Supervised machine learning (ML) is an artificial intelligence method that is used for optimizing the interpretation of complex or large datasets. During recent years, ML has increasingly been implemented in the field of medicine as a tool for optimizing diagnostics (11,12,13,14). By bootstrapping and cross-validating training data, the 'random forest' algorithm produces several decision trees that 'vote' to classify new cases according to the pattern established by the training process (15).…”

“…By bootstrapping and cross-validating training data, the 'random forest' algorithm produces several decision trees that 'vote' to classify new cases according to the pattern established by the training process (15). A recent publication outlined a ML model to detect KS patients among adult male patients with azoospermia (13).…”

Accurate Quantitative Histomorphometric-Mathematical Image Analysis Methodology of Rodent Testicular Tissue and Its Possible Future Research Perspectives in Andrology and Reproductive Medicine

“…Benefits of slide-scanning technique is not unknown in andrology, similar ones have been used for quantitative sperm morphology analysis [67,68] which could be complemented with the help of convolutional neuronal network deep-learning [69]. A supervised machine learning-based prediction model that has been developed for the identification of patients with Klinefelter-syndrome among azoospermic patients, has significantly better sensitivity and can improve diagnostic rate of the illness [70]. Another deep learning-based method was developed to assess the stages of Wistar rat spermatogenic cycle on hematoxylin-eosin-stained digital slides, which makes the quick evaluation of stage-frequency possible [71].…”

Prediction model for missed abortion of patients treated with IVF-ET based on XGBoost: a retrospective study

“…Predictive models are widely used in the medical field, and the common modeling methods include traditional logical regression ( Lian et al, 2022 ; Xu et al, 2022 ) and machine learning ( Krenz et al, 2022 ; Yan et al, 2021 ). A traditional prediction model is a well-known modeling and analysis method in the field of ART, but its application is limited.…”