Predicting Mesothelioma Using Artificial Intelligence: A Scoping Review of Common Models and Applications

Our plain-English summary of Predicting Mesothelioma Using Artificial Intelligence: A Scoping Review of Common Models and Applications, by M. Ram, M.R. Afrash, K. Moulaei, et al (Technology in Cancer Research & Treatment, 2025).

The study

• Authors: M. Ram, M.R. Afrash, K. Moulaei, E. Esmaeeli, M.S. Khorashadizadeh, A. Garavand, P. Amiri, A. Sabahi
• Journal: Technology in Cancer Research & Treatment
• Published: 2025;24:15330338251341053
• Cited by: 5 articles, as counted by NIH iCite
• Read it: PubMed · Publisher · Free full text

Abstract

IntroductionMesothelioma is a type of lung cancer caused by asbestos exposure, and early diagnosis is crucial for improving survival chances. Artificial intelligence offers a potential solution for the timely diagnosis and staging of the disease. This study aims to review the latest research conducted in artificial intelligence applications to predict mesothelioma.MethodsUntil April 24, 2023, PubMed, Scopus, and Web of Science databases were searched comprehensively for articles on artificial intelligence in mesothelioma management. The data was gathered using a standardized extraction form, and the findings were reported in figures and tables.ResultsOne hundred and seventy-three articles were identified from database searches, which were then reduced to 151 after eliminating duplicates. Finally, 19 articles were selected for inclusion in our study. The applications of artificial intelligence in these articles primarily focused on tumor diagnosis and classification (73.69%), followed by prevention and prognosis (21.05%) and tumor volumetric measurement of malignant pleural mesothelioma (5.26%). The most frequently used AI models include types of neural networks (NN), decision trees (DT), random forests (RF), logistic regression (LogR), Naïve Bayes (NB), and support vector machines (SVM). SVM, DT, and RF emerged as prominent models, achieving high accuracies ranging from 78.3% to 99.97%. Genetic algorithms, correlation-based algorithms, and Neural Networks were employed for risk factor identification and feature selection.ConclusionArtificial intelligence, particularly machine learning models such as neural networks, decision trees, support vector machines, and random forests, holds promise in predicting and managing mesothelioma, potentially enhancing early detection and improving patient outcomes.

In plain English

This article was a scoping review of earlier research. It did not study patients or asbestos exposure directly. The authors searched PubMed, Scopus and Web of Science for articles published up to April 24, 2023, on artificial intelligence (AI) in mesothelioma. They selected 19 articles, published from 2015 to 2022. Of these, 14 (73.69%) used AI for tumor diagnosis and classification, 4 (21.05%) for prevention and prognosis, and 1 (5.26%) for measuring tumor volume in malignant pleural mesothelioma. The most common model types were neural networks (9), decision trees and random forests together (8), logistic regression and Naïve Bayes together (7), and support vector machines (4). Summarizing the included studies, the authors reported average accuracies of 99.97% for support vector machines, 93.75% for random forests, 88.36% for decision trees, 78.3% for Naïve Bayes and 100% for logistic regression. Among the studies' best models, the authors named 71.29%, from one study using AdaBoost, as the lowest accuracy; some individual models scored lower, down to 52%. The authors concluded that AI models have significant potential for mesothelioma prediction and management. They noted that they searched only English-language studies and only three databases.

We wrote this plain-English summary from the full article with the help of AI, and a second AI model checked each sentence against the article. It is not the authors' wording; the abstract and the article are their own account.

What the study did and found

• Design: scoping review (reported using the PRISMA-ScR checklist)
• Who was studied: 19 published articles on AI in mesothelioma, selected from 173 found in PubMed, Scopus and Web of Science (151 after removing duplicates; 36 read in full), published up to April 24, 2023; the included articles dated from 2015 to 2022
• Exposure: Not an asbestos exposure study: no fiber types, products, jobs or doses were analyzed. The review examined AI technologies (machine learning and deep learning) used to predict mesothelioma.
• Main results:
  • Of the 19 included articles, 14 (73.69%) applied AI to tumor diagnosis and classification, 4 (21.05%) to prevention and prognosis, and 1 (5.26%) to tumor volume measurement of malignant pleural mesothelioma
  • Most commonly used model types across the included articles: neural networks (n = 9), decision trees and random forests (n = 8), logistic regression and Naïve Bayes (n = 7), support vector machines (n = 4)
  • Average accuracy reported for support vector machine models across the included studies: 99.97%
  • Accuracy reported for random forest models (used four times): 93.75%; decision tree models (used five times): 88.36%; Naïve Bayes models (used three times): 78.3%
  • In the two included studies that used logistic regression, the accuracy was 100%
  • In the comparison of the studies' best proposed models, the authors called 71.29%, from one study (Choudhury et al) using AdaBoost on 324 patients, the lowest accuracy among the reviewed studies
  • Table 4 lists lower accuracies for individual models within studies, for example 52% for an MLP model in the Chicco study
  • In the one included tumor-volume study (Kidd et al), human and AI volume responses agreed in 20 of 30 cases (67%), kappa 0.439 (0.178 to 0.700)
• The authors' conclusion: The authors concluded that AI, especially machine learning models such as neural networks, decision trees, support vector machines and random forests, has significant potential to improve mesothelioma prediction and management.
• Limitations the authors state:
  • Only English-language studies were included, so relevant research in other languages may have been missed
  • The search was limited to three databases (PubMed, Scopus and Web of Science)
  • The authors noted that there were relatively few studies and that their sample sizes were limited

Topics

The National Library of Medicine indexes this study under subjects our site covers: Mesothelioma.

Medical terms in this study, explained in our Lexicon: Mesothelioma, Lung Cancer, Asbestos Exposure, Staging, Prognosis, Pleural Mesothelioma.

Related studies

• Application of artificial intelligence in the analysis of asbestos fibers — R. Lee, D. Van Orden, et al. (2025)
• Rapid fiber-detection technique by artificial intelligence in phase-contrast microscope images of simulated atmospheric samples — T. Yamamoto, K. Iwasaki, et al. (2024)
• Artificial intelligence-based diagnosis of asbestosis: analysis of a database with applicants for asbestosis state aid — K.B.W. Groot Lipman, C.J. de Gooijer, et al. (2023)
• Development of rapid and highly accurate method to measure concentration of fibers in atmosphere using artificial intelligence and scanning electron microscopy — Y. Iida, K. Watanabe, et al. (2021)
• Prospective validation of an artificial intelligence assessment in a cohort of applicants seeking financial compensation for asbestosis (PROSBEST) — I. Smesseim, K.B.W.G. Lipman, et al. (2025)
• Artificial Intelligence in Pleural Diseases: Current Applications and Next Steps — F. Karataş, Ö. Dikensoy (2026)

Browse return to medical research.