year 12, Issue 1 (5-2025)
CJS 2025, 12(1): 65-75 |
Back to browse issues page
Download citation:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:
Zoumakzehi S, Makkizadeh F, Ebrahimi F, Hazeri A. An investigation of the conceptual network of scientific productions in the field of artificial intelligence and therapy through topic modeling. CJS 2025; 12 (1) :65-75
URL: http://cjs.mubabol.ac.ir/article-1-377-en.html
URL: http://cjs.mubabol.ac.ir/article-1-377-en.html
Department of Information Science and Knowledge, Faculty of Psychology and Educational Sciences, Yazd University, Yazd, Iran & Yazd university , makkizadeh@yazd.ac.ir
Abstract: (1396 Views)
Background and aim: Artificial intelligence has been used since 1950 in the areas of personalized health services and analyzing patient data, such as medical history and lifestyle factors. The aim of the present study is to model the thematic content of scientific products in the field of artificial intelligence in the healthcare sector based on data from the PubMed database.
Materials and methods: This study is descriptive-exploratory in nature. The statistical population includes all scientific publications (16,011 articles) related to artificial intelligence in medical treatment, indexed in the PubMed biomedical literature database from the beginning to the end of 2023. For modeling and analysis, abstracts and titles are used by combining LDA and TF-IDF algorithms.
Findings: The clustering findings led to the formation of eight thematic clusters: “Robotic-Assisted Laparoscopy”, “Deep Learning Models for Disease Prediction”, “Robotic Surgery”, “Clinical Applications of Artificial Intelligence”, “Robotic Rehabilitation”, “Medical Imaging with Deep Learning”, “Robotic Radical Prostatectomy”, and “Social Robotics”. This diversity indicates the wide-ranging scope of research within AI applications in medical treatment. Heat map analysis revealed a strong correlation (0.91) between the clusters “Robotic-Assisted Laparoscopy” and “Disease Prediction Models”, highlighting the interdisciplinary nature of this field. The most frequently weighted keywords in the literature were “robotics”, “surgery”, “patients”, “model”, and “prostatectomy”.
Conclusion: The “robotics” is the most important keyword in the field of artificial intelligence and treatment. The results also show that the extracted clusters not only have thematic coherence, but also a logical and meaningful connection is seen between them.
Materials and methods: This study is descriptive-exploratory in nature. The statistical population includes all scientific publications (16,011 articles) related to artificial intelligence in medical treatment, indexed in the PubMed biomedical literature database from the beginning to the end of 2023. For modeling and analysis, abstracts and titles are used by combining LDA and TF-IDF algorithms.
Findings: The clustering findings led to the formation of eight thematic clusters: “Robotic-Assisted Laparoscopy”, “Deep Learning Models for Disease Prediction”, “Robotic Surgery”, “Clinical Applications of Artificial Intelligence”, “Robotic Rehabilitation”, “Medical Imaging with Deep Learning”, “Robotic Radical Prostatectomy”, and “Social Robotics”. This diversity indicates the wide-ranging scope of research within AI applications in medical treatment. Heat map analysis revealed a strong correlation (0.91) between the clusters “Robotic-Assisted Laparoscopy” and “Disease Prediction Models”, highlighting the interdisciplinary nature of this field. The most frequently weighted keywords in the literature were “robotics”, “surgery”, “patients”, “model”, and “prostatectomy”.
Conclusion: The “robotics” is the most important keyword in the field of artificial intelligence and treatment. The results also show that the extracted clusters not only have thematic coherence, but also a logical and meaningful connection is seen between them.
Type of Study: Orginal |
Subject:
Scientometrics
Received: 2025/01/29 | Revised: 2025/07/15 | Accepted: 2025/07/26 | ePublished: 2025/08/11
Received: 2025/01/29 | Revised: 2025/07/15 | Accepted: 2025/07/26 | ePublished: 2025/08/11
Send email to the article author
| Rights and permissions | |
 |
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License. |
