AI model developed using machine learning for predicting sperm retrieval in micro TESE for non-obstructive azoospermia patients

Abstract

Azoospermia is a severe problem that prevents couples from having their own children through natural pregnancy. To obtain sperm from patients with azoospermia testicular sperm extraction (TESE) is required. In non-obstructive azoospermia (NOA), a particularly severe type of male infertility, patients need to undergo microdissection testicular sperm extraction (micro TESE) to collect sperm and, at 40 to 60%, the sperm retrieval success rate is not very high. Therefore, our objective in the present study was to make a model for predicting the possibility of sperm retrieval in patients with NOA before performing micro TESE, using machine learning. We retrospectively obtained data from the medical records of 430 patients who underwent micro TESE from 2011 to 2020, and used Prediction One, which does not require coding, to create the model. We successfully achieved our objective. The AUC for the AI model was 0.7246, which is acceptable. In addition, among the variables, we found that T/E2 ratios contributed most to predicting whether sperm retrieval was possible or not. T/E2 ratios have potential as a clinical predictor of sperm retrieval in NOA.

Potential for artificial intelligence in medicine and its application to male infertility

“…The AI model achieved an acceptable AUC of 72.46%. T/E2 ratios were found to be the most important variable for predicting sperm retrieval 48 (Figure 3 ).…”

“… 46 In addition, although a sperm retrieval AI prediction model for conventional TESE in patients with NOA had been previously reported, 60 our study is the first on a sperm retrieval AI prediction model for micro‐TESE in NOA patients. 48 …”