RF Safe

A Decision Support System for Managing Health Symptoms of Living Near Mobile Phone Base Stations

J Biomed Phys Eng · 2026 · Harm · Moderate evidence

← Research stream
HarmModerate evidencePaper 2690

Machine learning models, especially SVM, predicted health symptoms associated with living near mobile phone base stations with high accuracy and AUC values. Key predictors included distance from base station, age, and duration of residence. There is a connect…

Study snapshot

PublicationJ Biomed Phys Eng · 2026
Study typecross_sectional
Reported directionHarm
Evidence ratingModerate

Abstract

Category: Epidemiology, Public Health, Machine Learning Tags: radiofrequency electromagnetic fields, mobile phone base stations, health symptoms, machine learning, SVM, public health, RF-EMF exposure DOI: 10.31661/jbpe.v0i0.2310-1667 URL: jbpe.sums.ac.ir Overview The rapid increase in the number of Mobile Phone Base Stations (MPBS) worldwide has given rise to significant global concerns about potential adverse health effects stemming from exposure to Radiofrequency Electromagnetic Fields (RF-EMF). This study investigates the use of machine learning models as a proactive means for healthcare professionals and policymakers to address and manage concerns related to RF-EMF exposure—especially among individuals living near MPBS. Objective This research aimed to explore whether machine learning models could accurately predict health symptoms that are associated with living near mobile phone base stations, thereby enabling more effective management of these risks. Material and Methods - Analytical study using Support Vector Machine (SVM) and Random Forest (RF) algorithms. - Included 11 predictors relevant to participants' living conditions. - Total of 699 adult participants. - Model performance measured by sensitivity, specificity, accuracy, and Area Under Curve (AUC). Findings - SVM-based model accuracies for: - Headache: 85.3% - Sleep disturbance: 82% - Dizziness: 84% - Vertigo: 82.4% - Fatigue: 65.1% - Corresponding AUC values: 0.99, 0.98, 0.92, 0.89, and 0.81, respectively. - SVM model consistently outperformed RF and previously developed models, especially for fatigue sensitivity (70% SVM vs. 8% MLPNN vs. 11.1% RF). Key predictors identified were: distance from the base station, age, and duration of residence. Conclusion Machine learning methods—most notably SVM—show considerable promise for the management and prediction of health symptoms in individuals living near or considering residing near mobile phone base stations. There is a connection between proximity to mobile base stations and increased prevalence of headaches, sleep disturbances, dizziness, irritability, concentration issues, and neuropsychiatric risks, reaffirming concerns about health risks from electromagnetic field exposure. Health Effects Linked to EMF Exposure Effect: harm - Studies cited support that people living closer to base stations (

Machine Enhanced Logic

AI evidence extraction

Structured fields below were generated from the paper’s abstract and metadata to make the evidence easier to inspect. Automated extraction may be incomplete or incorrect; verify important details against the original paper.

Study typecross_sectional
Effect directionharm
Population699 adult participants living near mobile phone base stations
Sample size699
ExposureRF
Evidence strengthModerateConfidence: 70% · Peer reviewed: yes

Main findings

Machine learning models, especially SVM, predicted health symptoms associated with living near mobile phone base stations with high accuracy and AUC values. Key predictors included distance from base station, age, and duration of residence. There is a connection between proximity to base stations and increased prevalence of various symptoms.

Outcomes measured

  • headache
  • sleep disturbance
  • dizziness
  • vertigo
  • fatigue
  • irritability
  • concentration issues
  • neuropsychiatric risks

Limitations

  • Cross-sectional design limits causal inference
  • Exposure metrics limited to proximity and living conditions, no direct EMF measurement
  • Potential confounding factors not detailed
View raw extracted JSON
{
  "study_type": "cross_sectional",
  "exposure": {
    "band": "RF",
    "source": "mobile phone base station",
    "frequency_mhz": null,
    "sar_wkg": null,
    "duration": null
  },
  "population": "699 adult participants living near mobile phone base stations",
  "sample_size": 699,
  "outcomes": [
    "headache",
    "sleep disturbance",
    "dizziness",
    "vertigo",
    "fatigue",
    "irritability",
    "concentration issues",
    "neuropsychiatric risks"
  ],
  "main_findings": "Machine learning models, especially SVM, predicted health symptoms associated with living near mobile phone base stations with high accuracy and AUC values. Key predictors included distance from base station, age, and duration of residence. There is a connection between proximity to base stations and increased prevalence of various symptoms.",
  "effect_direction": "harm",
  "limitations": [
    "Cross-sectional design limits causal inference",
    "Exposure metrics limited to proximity and living conditions, no direct EMF measurement",
    "Potential confounding factors not detailed"
  ],
  "evidence_strength": "moderate",
  "confidence": 0.7,
  "peer_reviewed_likely": "yes",
  "keywords": [
    "radiofrequency electromagnetic fields",
    "mobile phone base stations",
    "health symptoms",
    "machine learning",
    "SVM",
    "public health",
    "RF-EMF exposure"
  ],
  "suggested_hubs": [
    {
      "slug": "5g-policy",
      "weight": 0.7,
      "reason": "Study focuses on health effects related to mobile phone base stations, relevant to 5G infrastructure policy."
    }
  ]
}
AI can be wrong. Always verify against the paper.

AI-extracted fields are generated from the abstract and metadata and may be incomplete or incorrect. This material is for informational purposes and is not medical advice.

Identifiers and source

DOI: 10.31661/jbpe.v0i0.2310-1667
PubMed: Not supplied

Read the original paper →

Related papers

← Back to the Research Library