[3Xin4] Poster session 1

[3Xin4-36] Analysis of COVID-19 Case Registration Form Toward Automated Registration of Patients through Natural Language Processing

〇Takuya Fukumoto¹, Ami Sakane¹, Shumpei Muramatsu¹, Masanao Igarashi², Yoshinobu Kano¹, Eiji Aramaki³, Hiromasa Horiguchi², Takashi Okumura⁴ (1.Shizuoka University, 2.National Hospital Organization, 3.Nara Institute of Science and Technology, 4.Kitami Institute of Technology)

Keywords:COVID-19, Notification of Occurrence, NLP

During the COVID-19 pandemic, medical institutions were forced to file the case registration forms to the public health authority, when they identify infected patients. However, the case registration form required substantial works to complete, and imposed heavy burdens on physicians who are on the front lines of the pandemic response. This study analysed the case registration form, in the aim of establishing the automated entry of the form out of electric health records used throughout the country nowadays. The analysis by domain experts exhibited that 26% of the fields could be entered automatically, 50% could be filled with the support of Natural Language Processing, 16% necessitated further engineering efforts with investment, and the last 9% has proved infeasible at present.