6,486 Studies Reviewed. 85.2% of Original Studies Found Biological Effects. The Evidence is Clear.
Shield Your Body
Reproductive Health

Analyzing the impact of occupational exposures on male fertility indicators: A machine learning approach

Bioeffects Seen

Authors not listed · 2025

Share:

Machine learning analysis of 80 auto parts workers identified magnetic and electric field exposure among the strongest predictors of lower free testosterone.

Plain English Summary

Summary written for general audiences

Researchers studied 80 male auto workers exposed to magnetic fields, electric fields, and other workplace hazards to predict reproductive health impacts. Machine learning models found that magnetic field exposure was the strongest predictor of reduced free testosterone levels (SHAP importance 0.339), and a random forest model ranked electric field exposure (5% increase in MSE) and magnetic field exposure (4.7%) as having the most substantial negative impact. The authors conclude that targeted interventions, such as improved workplace safety protocols and regular health monitoring, are needed to protect workers' reproductive health.

Why This Matters

This research links workplace magnetic field exposure to free testosterone levels in a machine learning model. Magnetic field exposure ranked as the top exposure predictor of reduced free testosterone (SHAP importance 0.339), ahead of wet bulb globe temperature (0.138). The models were trained on data from 80 male workers at an automobile part manufacturing plant.

Cite This Study
Unknown (2025). Analyzing the impact of occupational exposures on male fertility indicators: A machine learning approach.
Show BibTeX
@article{analyzing_the_impact_of_occupational_exposures_on_male_fertility_indicators_a_machine_learning_approach_ce4635,
  author = {Unknown},
  title = {Analyzing the impact of occupational exposures on male fertility indicators: A machine learning approach},
  year = {2025},
  doi = {10.1016/j.reprotox.2025.108959},
  
}
DOI: 10.1016/j.reprotox.2025.108959PubMed: 40449633

Quick Questions About This Study

Yes, this study found magnetic field exposure was the strongest predictor of reduced free testosterone in male auto workers, with a SHAP importance score of 0.339, indicating substantial impact on reproductive health.
XGBoost and random forest achieved the highest AUC (0.99), outperforming the other models in predictive accuracy.
In the random forest model, electric field exposure (5% increase in MSE) and magnetic field exposure (4.7%) had the most substantial negative impact on free testosterone, followed by worker age (4.1%). In the SHAP ranking, magnetic field exposure scored 0.339.
Worker age ranked as the strongest demographic factor (0.244 SHAP importance), but magnetic field exposure (0.339) was even more influential, suggesting EMF poses greater reproductive risks than natural aging.
Five models (logistic regression, bagging classifier, extreme gradient boosting, random forest, and support vector machine) were trained and evaluated with 5-fold cross-validation on data from 80 male workers at an automobile part manufacturing plant.