Evolution of medical reports over time: an analysis using Dynamic Topic Modeling

In recent years, the availability of data has increased across various fields, with a greater focus on the medical sector. Medical records, clinical charts, and diagnostic reports provide a large amount of information. The complexity of this data requires the use of advanced analytical methods to extract reliable and important information. This study aims to test Text Mining techniques, particularly Dynamic Topic Modeling (DTM), to monitor the evolution of medical practices within surgical records related to the descriptions of surgeries over time. By applying DTM to a large dataset of medical reports from a urology clinic, we were able to identify emerging topics, highlighting changes in topics over time and whether one or more surgeries are more or less frequent in the specific years 2022 and 2023. This method not only extracted emerging topics from the descriptions of surgical interventions but also provided a precise overview of how clinical practices and patient outcomes have evolved. Through this analysis, we can gain a more detailed understanding of therapeutic strategies and how they have evolved during the specified years, contributing to future clinical decision-making and improving patient care. The findings of this research provide detailed information on the importance of using advanced techniques to better understand medical data and its implications for healthcare practices.