Performance Improvement of a Natural Language Processing Tool for Extracting Patient Narratives Related to Medical States From Japanese Pharmaceutical Care Records by Increasing the Amount of Training Data: Natural Language Processing Analysis and Validation Study
Performance Improvement of a Natural Language Processing Tool for Extracting Patient Narratives Related to Medical States From Japanese Pharmaceutical Care Records by Increasing the Amount of Training Data: Natural Language Processing Analysis and Validation Study
Yukiko Ohno
1
, BSc ;
Tohru Aomori
2
, PhD ;
Tomohiro Nishiyama
3
, MEng ;
Riri Kato
4
, BSc ;
Reina Fujiki
5
, BSc ;
Haruki Ishikawa
5
, MSc ;
Keisuke Kiyomiya
4
, PhD ;
Minae Isawa
4
, PhD ;
Mayumi Mochizuki
4
, PhD ;
Eiji Aramaki
3
, PhD ;
Hisakazu Ohtani
1, 5, 6
, PhD
1
Graduate School of Pharmaceutical Sciences, Keio University, Tokyo, Japan
2
Faculty of Pharmacy, Takasaki University of Health and Welfare, Takasaki, Japan
3
Nara Institute of Science and Technology, Ikoma, Japan
4
Faculty of Pharmacy, Keio University, Tokyo, Japan
5
Department of Pharmacy, Keio University Hospital, Tokyo, Japan
6
School of Medicine, Keio University, Tokyo, Japan
Corresponding Author:
-
Hisakazu Ohtani, PhD
-
Graduate School of Pharmaceutical Sciences
-
Keio University
-
1-5-30, Shibakouen, Minato
-
Tokyo, 105-8512
-
Japan
-
Phone:
81 3-5400-2486
-
Email: ohtani@keio.jp