Machine Learning Models for Predicting Breast Cancer Risk in Women Exposed to Blue Light from Digital Screens

Authors: Mortazavi, S. A., Tahmasebi, S., Parsaei, H., Taleie, A., Faraz, M., Rezaianzadeh, A., Zamani, A., Mortazavi, S. M. J.

Year: 2022

Category: Biomedical Physics and Engineering

Journal: Journal of Biomedical Physics and Engineering

DOI: 10.31661/jbpe.v0i0.2105-1341

URL: https://jbpe.sums.ac.ir/article_48176.html

Abstract

Abstract

Overview

The growing concern over increased screen time involves the detrimental effects of blue light from digital devices. This exposure is now linked to potential cancer risks.

Objective

The main goal of this study is to develop machine learning models that help predict the risk of breast cancer from parameters associated with both ionizing and non-ionizing radiation exposure.

Material and Methods

  • Utilization of three machine learning models: Random Forest, Support Vector Machine, and Multi-Layer Perceptron Neural Network.
  • Data analysis from 603 participants, including both breast cancer patients and matched controls.
  • Data collection through standardized interviews and questionnaires.

Results

The machine learning models demonstrated high efficacy, with sensitivities over 97.2%, specificities over 96.4%, and an overall accuracy rate of more than 97.1% in distinguishing between cancer patients and healthy controls.

Conclusion

Machine learning technics are effective in predicting breast cancer risks related to radiation exposure. These preliminary results are promising, though additional research is necessary to validate these findings fully.

machine learning breast cancer risk blue light exposure digital screens ionizing radiation
← Back to Stats