Maintenance Notice

Due to necessary scheduled maintenance, the JMIR Publications website will be unavailable from Wednesday, July 01, 2020 at 8:00 PM to 10:00 PM EST. We apologize in advance for any inconvenience this may cause you.

Who will be affected?

Accepted for/Published in: Journal of Medical Internet Research

Date Submitted: Nov 4, 2022
Date Accepted: Apr 7, 2023

The final, peer-reviewed published version of this preprint can be found here:

A Machine Learning Approach for the Prediction of Testicular Sperm Extraction in Nonobstructive Azoospermia: Algorithm Development and Validation Study

Bachelot G, Dhombres F, Sermondade N, Haj Hamid R, Berthaut I, Frydman V, Prades M, Kolanska K, Selleret L, Mathieu-D’Argent E, Rivet-Danon D, Levy R, Lamazière A, Dupont C

A Machine Learning Approach for the Prediction of Testicular Sperm Extraction in Nonobstructive Azoospermia: Algorithm Development and Validation Study

J Med Internet Res 2023;25:e44047

DOI: 10.2196/44047

PMID: 37342078

PMCID: 10337455

Warning: This is an author submission that is not peer-reviewed or edited. Preprints - unless they show as "accepted" - should not be relied on to guide clinical practice or health-related behavior and should not be reported in news media as established information.

Machine learning-based prediction of testicular sperm extraction: A comparison of different models.

  • Guillaume Bachelot; 
  • Ferdinand Dhombres; 
  • Nathalie Sermondade; 
  • Rahaf Haj Hamid; 
  • Isabelle Berthaut; 
  • Valentine Frydman; 
  • Marie Prades; 
  • Kamila Kolanska; 
  • Lise Selleret; 
  • Emmanuelle Mathieu-D’Argent; 
  • Diane Rivet-Danon; 
  • Rachel Levy; 
  • Antonin Lamazière; 
  • Charlotte Dupont

ABSTRACT

Background:

Testicular sperm extraction (TESE) is an essential therapeutic tool for the male infertility management. However, it is an invasive procedure and is successful in up to 50%. Until now, no model, based on the clinical and laboratory parameters, is sufficiently powerful to accurately predict the success of sperm retrieval in TESE.

Objective:

The objective of this study is to compare a wide range of predictive models under similar conditions for TESE outcomes in patients with non-obstructive azoospermia (NOA) to identify the correct mathematical approach to apply thereto, as well as the most appropriate study size and the relevance of input biomarkers.

Methods:

Two-hundred and one patients who underwent TESE at Tenon Hospital (AP-HP, Sorbonne University, Paris) distributed in a retrospective training cohort of 175 patients (2012 to April 2021) and a prospective testing cohort (May 2021 to December 2021) of 26 patients were analyzed. Eight machine learning (ML) models were evaluated (temporal validation) and compared.

Results:

The best model was the random forest (RF): (Area under curve) AUC = 89.6%, sensitivity = 100%, and specificity = 69,2%. Moreover, a study size of 120 patients seemed to be sufficient to properly exploit the preoperative data during the modeling process. Furthermore, inhibin B and a history of varicoceles exhibited the highest predictive capacity.

Conclusions:

Using the appropriate approach, an ML algorithm can accurately predict successful sperm retrieval in men with NOA under-going TESE. However, it will be necessary to confirm and validate these results in a multicentric prospective study prior to practical use. The applicability of seminal plasma biomarkers, especially noncoding RNAs, as markers of residual spermatogenesis in NOA patients also requires further study.


 Citation

Please cite as:

Bachelot G, Dhombres F, Sermondade N, Haj Hamid R, Berthaut I, Frydman V, Prades M, Kolanska K, Selleret L, Mathieu-D’Argent E, Rivet-Danon D, Levy R, Lamazière A, Dupont C

A Machine Learning Approach for the Prediction of Testicular Sperm Extraction in Nonobstructive Azoospermia: Algorithm Development and Validation Study

J Med Internet Res 2023;25:e44047

DOI: 10.2196/44047

PMID: 37342078

PMCID: 10337455

Download PDF


Request queued. Please wait while the file is being generated. It may take some time.

© The authors. All rights reserved. This is a privileged document currently under peer-review/community review (or an accepted/rejected manuscript). Authors have provided JMIR Publications with an exclusive license to publish this preprint on it's website for review and ahead-of-print citation purposes only. While the final peer-reviewed paper may be licensed under a cc-by license on publication, at this stage authors and publisher expressively prohibit redistribution of this draft paper other than for review purposes.