Epidemiological exploration of the impact of bluetooth headset usage on thyroid nodules using Shapley additive explanations method

Abstract

Overview

This study investigates the growing concern regarding the impact of Bluetooth headsets on the prevalence of thyroid nodules, attributing potential risks to the non-ionizing radiation emitted by these devices.

Methodology

• Analysis of 600 valid questionnaires from the WenJuanXing platform.
• Employment of Propensity Score Matching (PSM) to minimize biases by balancing baseline characteristic differences.
• Use of the XGBOOST model to predict and measure the risk factors with efficacy gauged by the AUC score.
• Supplementation of SHAP analysis to quantify the influence of individual features on the model's predictions.

Findings

The study processed a subset of matched datasets yielding 96 cases for focused analysis, achieving an AUC value of 0.95, indicating high model accuracy in risk differentiation. Notably, SHAP analysis identifies age and daily usage duration of Bluetooth headsets as significant predictors of thyroid nodule risk, illustrating that prolonged usage correlates positively with increased risk.

Conclusion

This research underscores the significant adverse effects prolonged use of Bluetooth headsets could have on thyroid health. The findings serve as a basis for public health policy-making and personal health decisions, advocating for moderated use of Bluetooth technology to mitigate health risks. The study suggests avenues for future research to explore biological mechanisms and expand on the identified correlations.