MP15-03 NOVEL APPLICATION FOR SEXUAL TRANSMITTED INFECTION SCREENING WITH AN AI CHATBOT

Abstract

You have accessJournal of UrologyInfections/Inflammation/Cystic Disease of the Genitourinary Tract: Prostate & Genitalia I1 Apr 2018MP15-03 NOVEL APPLICATION FOR SEXUAL TRANSMITTED INFECTION SCREENING WITH AN AI CHATBOT Yoshitomo Kobori, Akiyoshi Osaka, Shigehiro Soh, and Hiroshi Okada Yoshitomo KoboriYoshitomo Kobori More articles by this author , Akiyoshi OsakaAkiyoshi Osaka More articles by this author , Shigehiro SohShigehiro Soh More articles by this author , and Hiroshi OkadaHiroshi Okada More articles by this author View All Author Informationhttps://doi.org/10.1016/j.juro.2018.02.516AboutPDF ToolsAdd to favoritesDownload CitationsTrack CitationsPermissionsReprints ShareFacebookTwitterLinked InEmail INTRODUCTION AND OBJECTIVES An outbreak of syphilis is a serious problem in the past several years in Japan. An accurate diagnosis of STI is necessary for proper use of antibiotics, successful treatment and prevention against the generation of antibiotic-resistant bacteria. In current practice people must visit a clinic or other hospital facility to diagnose their STI. Once at the clinic, they are asked a medical interview, which is then analyzed by staff at the clinic. Many subjects are not comfortable with this procedure, which they often find embarrassing. An AI chatbot is a computer program which conducts a conversation via auditory or textual methods. Chatbots are typically used in dialog systems for various practical purposes including customer service or information acquisition. Chatbots began to be used in a medical field recently. Here we report on a novel chatbot system to diagnose STI on the internet with a smartphone. METHODS We built the database which contained information, pictures and questions to diagnosis male and female STI. This chatbot system was composed with autoanswer systems in Japanese and English comprising constant algorithm on a server. This program was designed to work on message applications such as PC browser or Facebook messenger. If patients answered a couple of questions and/or chose a picture, this chatbot could diagnose male STI such as gonorrheal urethritis, chlamydial urethritis, syphilis, condyloma acuminatum, the herpes simplex infectious disease, AIDS and others. In addition, it could diagnose non-STI diseases such as testicular tumor, orchitis, hydrocele testis, pearly penile papules and others. We investigated efficacy and usability of this chatbot. And we analyzed its accuracy of STI results of a total of 70 male patients who visited a clinic to treat STI. RESULTS The total accuracy rate of a diagnosis of STI with this chatbot was 77.7%. The accuracy rates of gonorrheal urethritis, chlamydial urethritis, syphilis and condyloma acuminatum were 65%, 70%, 85% and 95%, respectively. Surprisingly enough, 97.7% patients thought to visit to clinic earlier after they used a chatbot. CONCLUSIONS A chatbot have great potential to screen STI because we can use it anytime and anywhere on smartphone. Given its usability and accessibility, the application can support examining and learning STI at home, making it much more convenient and economical than current practice. And patients will be promoted to visit clinic effectively. This chatbot system to diagnosis STI provides an easy solution for global users to rapidly screen STI by themselves. © 2018FiguresReferencesRelatedDetails Volume 199Issue 4SApril 2018Page: e189-e190 Advertisement Copyright & Permissions© 2018MetricsAuthor Information Yoshitomo Kobori More articles by this author Akiyoshi Osaka More articles by this author Shigehiro Soh More articles by this author Hiroshi Okada More articles by this author Expand All Advertisement Advertisement PDF downloadLoading ...