Abstract
INTRODUCTION: This study aimed to investigate the landscape of Multiple Endocrine Neoplasia Type 1 research during the last 22 years using machine learning and text analysis. METHOD: In December 2021, all publications indexed under the MeSH term "Multiple Endocrine Neoplasia Type 1" were obtained from PubMed. The whole set of search results was downloaded in XML format, and metadata such as title, abstract, keywords, mesh words, and year of publication were extracted from the original XML files for bibliometric evaluation. The Latent Dirichlet allocation (LDA) topic modeling method was used to analyze specific themes. RESULTS: This study eventually contained 1,407 publications. Among them, there are 768 (54.58%) case reports and reviews. Text analysis based on MeSH words revealed that the most often studied clinical areas include therapy efficacy, prognosis, and genetic diagnosis. The majority of basic study is focused on genetic alterations. The LDA topic model further identifies three topic clusters include basic research, treatment cluster, and diagnosis cluster. In the basic research cluster, many studies are focused on the expression of Menin. The primary focus of the therapy cluster is pancreatic resections and parathyroidectomy. In the diagnose cluster, the main focus is on Genetic Diagnosis and screening strategies for Hereditary Cancer Syndrome. CONCLUSION: The current state of research on MEN1 is far from adequate. Research on rare diseases MEN1 necessitates implementing a broad research program involving multiple centers to advance MEN1 research together.