Abstract
By text mining using artificial intelligence IBM Watson®, we analyzed the relationship between dental hygienist clinical records at Fujiwara Dental Clinic and extraction. Text mining is an analysis method that divides freely written sentences into words and calculates their appearance frequency and correlation with specific items. As a result, expressions, such as “破”(kanji that stands for “break”) / “排”(drainage) / “膿”(pus) or “義歯”(denture)/ “印象”(impression) seemingly associated with tooth extraction or dentures respectively were frequently used in the medical records with 147 out of 1498 patients (medical records amounting to 9724) from January 2016 to June 2018, who underwent tooth extraction during the period. On the other hand, words related to oral hygiene, such as “floss”, “early”, “tartar”, and “plaque”, were frequently used in the medical records of 1351 patients who did not have their teeth extracted. How to apply these results to clinical practice is a topic for the future, but data can be analyzed from a different perspective than human insights, suggesting future usefulness in dental practice.
