Artificial Intelligence Techniques to Predict the Airway Disorders Illness: A Systematic Review

Our plain-English summary of Artificial Intelligence Techniques to Predict the Airway Disorders Illness: A Systematic Review, by A. Koul, R.K. Bawa, Y. Kumar (Archives of Computational Methods in Engineering, 2023).

The study

• Authors: A. Koul, R.K. Bawa, Y. Kumar
• Journal: Archives of Computational Methods in Engineering
• Published: 2023;30(2):831-864
• Type of article: Review
• Cited by: 25 articles, as counted by NIH iCite
• Read it: PubMed · Publisher · Free full text

Abstract

Airway disease is a major healthcare issue that causes at least 3 million fatalities every year. It is also considered one of the foremost causes of death all around the globe by 2030. Numerous studies have been undertaken to demonstrate the latest advances in artificial intelligence algorithms to assist in identifying and classifying these diseases. This comprehensive review aims to summarise the state-of-the-art machine and deep learning-based systems for detecting airway disorders, envisage the trends of the recent work in this domain, and analyze the difficulties and potential future paths. This systematic literature review includes the study of one hundred fifty-five articles on airway diseases such as cystic fibrosis, emphysema, lung cancer, Mesothelioma, covid-19, pneumoconiosis, asthma, pulmonary edema, tuberculosis, pulmonary embolism as well as highlights the automated learning techniques to predict them. The study concludes with a discussion and challenges about expanding the efficiency and machine and deep learning-assisted airway disease detection applications.

© The Author(s) under exclusive licence to International Center for Numerical Methods in Engineering (CIMNE) 2022, Springer Nature or its licensor holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

In plain English

This article was a systematic literature review, not an original study. Following PRISMA guidelines, the authors searched six databases for English-language research from 2010 to 2022 on machine learning and deep learning methods for detecting airway and lung diseases. They reviewed 155 papers, 100 of them from 2019 to 2022. The diseases covered were mesothelioma, lung cancer, pneumoconiosis, emphysema, tuberculosis, asthma, cystic fibrosis, pulmonary edema, pulmonary embolism and COVID-19. Convolutional neural networks were the deep learning models used most often. One reviewed mesothelioma study used data on 324 patients from southeast Turkey. The best mesothelioma model the review listed had an accuracy of 96%. Two of the reviewed mesothelioma studies looked for risk factors. One reported that asbestos exposure, its length and duration of symptoms had a significant influence on mesothelioma frequency. The other reported that length of symptoms together with asbestos exposure had a significant impact on the mesothelioma rate. The authors concluded that several reviewed models worked only with limited datasets, that some could not distinguish between diseases, and that researchers had focused on only one or two disorders. They named shortage of high-quality data, overfitting and class imbalance as key challenges.

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: Systematic literature review following PRISMA guidelines
• Who was studied: 155 published research papers from 2010 to 2022, found in six databases (ScienceDirect, Google Scholar, Scopus, Web of Science, PubMed, EMBASE), English language only. The papers covered AI detection of airway diseases, including mesothelioma, lung cancer, pneumoconiosis, emphysema, tuberculosis, asthma, cystic fibrosis, pulmonary edema, pulmonary embolism and COVID-19.
• Exposure: Not an exposure study. The review examined machine learning and deep learning methods (and a few traditional methods) for predicting or detecting airway diseases. Asbestos appears only in the findings of reviewed mesothelioma studies.
• Main results:
  • Of the 155 papers reviewed, 100 were from 2019 to 2022 and 55 were from 2010 to 2018.
  • Across the reviewed literature, convolutional neural networks were the deep learning models used most often. The most common machine learning models were linear regression, support vector machine, random forest, decision tree, naive Bayes, fuzzy particle, logistic regression and ensemble learning.
  • One reviewed study of mesothelioma risk factors (reference 102 in the review) used the Apriori algorithm. It reported that asbestos exposure, its length and duration of symptoms had a significant influence on the frequency of malignant mesothelioma. This is a finding of that study, not of the review itself.
  • Another reviewed study of mesothelioma risk variables (reference 104) also used the Apriori algorithm. It reported that length of symptoms, together with exposure to asbestos, had a significant impact on the malignant mesothelioma rate. This study does not mention length of asbestos exposure.
  • A third reviewed mesothelioma study (reference 101) looked for prognostic variables, also using the Apriori method. It used data on 324 patients from the Diyarbakir district of southeast Turkey.
  • In the review's overall comparison table, the mesothelioma model with the best accuracy came from one reviewed study (reference 105). It used SMOTE, ADASYN, an artificial neural network and principal component analysis, and reached 96% accuracy. Another reviewed study (reference 103) reported 71.29% accuracy.
  • The authors identified the main problems in the reviewed research as shortage of high-quality data, modeling errors (overfitting and underfitting), difficulty in continually improving models, and class imbalance.
• The authors' conclusion: The authors concluded that models examined by certain researchers could not distinguish between diseases or detect anomalies. They found that several models worked only with limited datasets, failed to pre-process data, or gave no localization information. They also concluded that researchers had focused on only one or two airway disorders, which limits the ability to distinguish between other airway conditions.

Topics

Medical terms in this study, explained in our Lexicon: Systematic Review, Fibrosis, Emphysema, Lung Cancer, Mesothelioma, Pneumoconiosis.

Related studies

• Rapid fiber-detection technique by artificial intelligence in phase-contrast microscope images of simulated atmospheric samples — T. Yamamoto, K. Iwasaki, et al. (2024)
• Pre-treatment Tumor Volume Growth Rate and Baseline Volume Measured by an Artificial Intelligence Program Predict Overall Survival in Patients With Pleural Mesothelioma Undergoing Deferred Systemic Therapy — I. Smesseim, V. Pugliese, et al. (2026)
• Gender Differences in the Use of ChatGPT as Generative Artificial Intelligence for Clinical Research and Decision-Making in Occupational Medicine — P. Mashburn, F.A. Weuthen, et al. (2025)
• Artificial Intelligence-based Quantification of Pleural Plaque Volume and Association With Lung Function in Asbestos-exposed Patients — K.B.W. Groot Lipman, T.N. Boellaard, 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)
• Predicting Mesothelioma Using Artificial Intelligence: A Scoping Review of Common Models and Applications — M. Ram, M.R. Afrash, et al. (2025)

Browse return to medical research.