Advancements and Applications of Machine Learning in Detecting Radon Nuclear Tracks from 2001 to 2023 - A Bibliometric Analysis

Aug 4, 2026

Upload by Sage

Bibliometric analysis evaluates advancements in machine learning for detecting radon nuclear tracks using publications from 2001 to 2023 retrieved from Scopus and Web of Science. Research output trends are assessed, with emphasis on key contributions from China and the United States. Major themes are identified through frequent terminology such as machine learning, radon, neural networks, and emerging approaches including XGBoost and long short-term memory networks. Authorship networks map collaborative structures and distinguish core versus peripheral research areas. Findings outline the field’s evolution and support continued interdisciplinary collaboration to improve radon risk assessment methods.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages9

Reading Time23 min

AccessFree to download

Introduction

Context of radon detection

Role of machine learning and deep learning

Methodology

Data sources and bibliometric metrics

Visual analysis of research development

ISBN: 978-628-95207-8-1. ISSN: 2414-6390. Digital Object Identifier: 10.18687/LACCEI2024.1.1.1018 Advancements and Applications of Machine Learning in Detecting Radon Nuclear Tracks from 2001 to 2023: A Bibliometric Analysis Félix Díaz1, Luis Sánchez2, Rafael Liza3, Jessica Toribio4, Nhell Cerna5 1Vicerrectorado de Investigación, Universidad Autónoma del Perú, Perú, [fe](fediazdes@autonoma.edu.pe)[diazdes@autonoma.edu.pe](fediazdes@autonoma.edu.pe)[ ](fediazdes@autonoma.edu.pe)2Universidad Privada del Norte, Perú, [j](jusa295@gmail.com)[usa295@gmail.com](jusa295@gmail.com) 3Universidad Tecnológica del Perú, Perú, [c20231@utp.edu.pe](c20231@utp.edu.pe) 4Pontificia Universidad Católica del Perú, Perú, [jessica.toribio@pucp.edu.pe](jessica.toribio@pucp.edu.pe) 5Universidad Norbert Wiener, Perú, [nhelldx@gmail.com](nhelldx@gmail.com) Abstract– We present a bibliometric analysis of the advancements in machine learning for detecting radon nuclear tracks, using publications from 2001 to 2023 sourced from Scopus and Web of Science databases. We analyze the growth in research output, particularly highlighting contributions from China and the United States, and identify key themes such as "machine learning","radon", "neural networks", and emerging methods like"xgboost" and "long short-term memory networks". Our findings underscore the collaborative efforts within the field, as evidenced by the global authorship networks. The research landscape is mapped out, revealing core and peripheral areas of study that define the current state and prospects of radon detection research. The present study encapsulates the evolution of the field and emphasizes the necessity for continued interdisciplinary collaboration to enhance radon risk assessment methods. Keywords--Machine Learning, Nuclear Tracks, Bibliometric. I. INTRODUCTION The significance of radon detection, a radioactive gas classified by the International Agency for Research on Cancer (IARC) as a Class 1 carcinogen [1], lies in its ability to accumulate in enclosed spaces, emanating from geological and construction materials, and its direct association with increased lung cancer risk following prolonged exposures to high concentrations [2-13] . This public health challenge has driven the evolution of advanced techniques for effective radon detection, becoming a crucial area of interest in environmental and residential research [14-21] . Since the early 20th century, radon detection has been a cornerstone in advancing nuclear physics and radioactivity, with traditional methods ranging from nuclear emulsion photography to solid trace detectors, albeit limited by the need for development and manual analysis [21-43] . In parallel, machine learning, rooted in computer science and statistics since the 1950s, has undergone a significant transformation, particularly with the technological renaissance of the 21st century characterized by increased data availability and computing capacity [44-52] . This progress has been particularly notable in developing deep learning algorithmsand neural networks, facilitating advances in natural language processing, computer vision, and the detection and analysis of radiological phenomena [53-61] . Applying these technologies Digital Object Identifier: (only for full papers, inserted by LACCEI) . ISSN, ISBN: (to be inserted by LACCEI) . DO NOT REMOVE to study nuclear traces has revolutionized previous methodologies, allowing for more efficient and accurate analysis and new directions for research in environmental health and public safety [62-65] . In this context, bibliometric analysis is an essential strategic tool for evaluating research trends and identifying existing knowledge gaps through concrete indicators such as citations, publications, and keywords [66-70] . Despite advancements in radon detection and machine learning, the literature needs a comprehensive bibliometric analysis that merges both fields from a global perspe

FAQs

What timeframe and databases does the bibliometric analysis cover?

The study analyzes publications from 2001 to 2023 and sources records from Scopus and Web of Science.

Which research themes and methods are highlighted as prominent in the literature?

The analysis highlights themes such as machine learning, radon, and neural networks, and identifies emerging methods including XGBoost and long short-term memory networks.

How does the study assess research impact and development?

It uses bibliometric metrics such as citations per article, and also examines contributions by authors, journals, and countries, supported by graphical representations based on citations, keywords, and authorships.

Aug 4, 2026

Upload by Sage