Usage of artificial intelligence capabilities in immunology: from literature review to statistical processing and analysis of obtained data
Year: 2025
Type: article
Abstract: Traditional literature review and the standard methods of data analysis in immunology is time-consuming and does not allow rapid solution of large-scale and complex research problems. Artificial intelligence can assist in addressing this issue by automating the processes of search and evaluation of relevant information. The aim of this study was to determine the potential of artificial intelligence in immunological research, based on the automated collection of literature data and analysis of digital information. The theoretical aspect of the study was based on the analysis of materials from PubMed, Scopus, ResearchGate databases covering the years 2000-2025. The practical approach was conducted via comparative analysis of previously published data and conclusions obtained by means of artificial intelligence technologies, e.g., by GPT v.4.0 models. It was found that artificial intelligence technologies may be useful in conducting literature reviews in immunology, including compilation of findings, generation of graphical presentations, as well as visualization of new signaling pathways, cellular interactions, and disease-related factors. In addition to the text analysis, artificial intelligence may be applied to statistical processing and digital data analysis, such as detection of regularities, solution of forecasting issues, design of models. A comparison of previously studied data with the results obtained using GPT v.4.0 revealed several limitations of different chatbot models, including the dependence of responses on the style of query formulation, excessively generalized information synthesis, limited text output (up to 1,000 words), plagiarism risks, difficulties in generating figures, diagrams, and tables, presentation of comprehensive information predominantly in English, and spontaneous creation of non-existent references to the literature sources. Conclusions: 1. Artificial intelligence may sufficiently change immunological research, allowing for a more effective literature reviewing and deeper data analysis, however, requiring expert supervision at the initial stages. 2. Upon development of artificial intelligence technologies, they will become an integral part of the immunologist’s toolkit.
Source: Russian Journal of Immunology
Author: Olga V. Berdiugina
Language: English
Subfield: Health Informatics
Field: Medicine
Domain: Health Sciences
Open Access status: diamond
OpenAlex ID: W4414326571