European Journal of Biomedical and Pharmaceutical Sciences

THE ROLE OF ARTIFICIAL INTELLIGENCE APPROACHES IN THE DIAGNOSIS OF CELIAC DISEASE FROM DUODENAL BIOPSY IMAGES: A SYSTEMATIC REVIEW OF THE LITERATURE

Iman Zahedi, Osahon David Omoregie, Esosa Daniel Omoregie, Tambi Isaac, Blessing Ebong,
Tiwalade Oluwadamilola Ogunlaja*, Madiha Haseeb, Vishal Deshpande, Vamshi Krishna Budda,
Adaku Anyalewechi Okafor, Akpevwoghene Victor Erhiano, Idowu Isaac Ige,
Chidinma Vivian Ikekpeazu, Ibrahim Ahmed Osman and Oluwatobiloba Frances Fasoranti-Sowemimo

ABSTRACT

Background: Celiac disease (CD) is a common autoimmune condition with complex diagnostic pathways. Diagnostic accuracy is critical in ensuring timely treatment and preventing long-term complications. In the past few years, there has been a surge in research exploring the application of artificial intelligence (AI) and machine learning (ML) in the diagnosis of CD. This systematic review aimed to collate and assess recent studies investigating the role of AI and ML in CD diagnosis. Methods: A comprehensive literature search was conducted in PubMed, Scopus, Web of Science, and Google Scholar using relevant keywords and MeSH terms. Original research studies published in the last twenty years were included if they focused on AI/ML application in CD diagnosis and provided sufficient data for analysis. The selection, data extraction, and synthesis were conducted following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. Findings: Nine studies were selected for review, which demonstrated promising results in the application of AI/ML for CD diagnosis. A variety of models and techniques, including Support Vector Machine (SVM), T-cell receptor sequencing, Steerable Pyramid Transform (SPT), and convolutional neural networks (CNNs), demonstrated high accuracies in distinguishing CD from non-CD cases. The review also highlighted the potential role of AI/ML in clinical decision support systems. Conclusion: This systematic review highlights the significant potential of AI/ML in enhancing the diagnostic process of CD. As AI/ML technologies continue to evolve, their integration into routine clinical practice could revolutionize CD management. Nevertheless, continual research and appraisal are essential in navigating the evolving landscape of these technologies and harnessing their full potential for patient benefit.

Keywords: Celiac Disease, Artificial Intelligence, Machine Learning, Diagnosis, Systematic Review.