Predicting coronavirus disease 2019 severity using explainable artificial intelligence techniques.

Journal: Scientific reports
PMID: 40108236

Abstract

Predictive models for determining coronavirus disease 2019 (COVID-19) severity have been established; however, the complexity of the interactions among factors limits the use of conventional statistical methods. This study aimed to establish a simple and accurate predictive model for COVID-19 severity using an explainable machine learning approach. A total of 3,301 patients ≥ 18 years diagnosed with COVID-19 between February 2020 and October 2022 were included. The discovery cohort comprised patients whose disease onset fell before October 1, 2020 (N = 1,023), and the validation cohort comprised the remaining patients (N = 2,278). Pointwise linear and logistic regression models were used to extract 41 features. Reinforcement learning was used to generate a simple model with high predictive accuracy. The primary evaluation was the area under the receiver operating characteristic curve (AUC). The predictive model achieved an AUC of ≥ 0.905 using four features: serum albumin levels, lactate dehydrogenase levels, age, and neutrophil count. The highest AUC value was 0.906 (sensitivity, 0.842; specificity, 0.811) in the discovery cohort and 0.861 (sensitivity, 0.804; specificity, 0.675) in the validation cohort. Simple and well-structured predictive models were established, which may aid in patient management and the selection of therapeutic interventions.

Authors

  • Takuya Ozawa
  • Shotaro Chubachi
    Division of Pulmonary Medicine, Department of Internal Medicine, Keio University School of Medicine, Tokyo, Japan. bachibachi472000@z6.keio.jp.
  • Ho Namkoong
    Department of Infectious Diseases, Keio University School of Medicine, 35 Shinanomachi, Shinjuku-ku, Tokyo, 160-8582, Japan. hounamugun@gmail.com.
  • Shota Nemoto
    Industrial & Digital Business Unit, Hitachi Ltd, Chiyoda-ku, Tokyo, Japan.
  • Ryo Ikegami
    Industrial and Digital Business Unit, Hitachi, Ltd, Tokyo, Japan.
  • Takanori Asakura
    Division of Pulmonary Medicine, Department of Internal Medicine, Keio University School of Medicine, Tokyo, Japan.
  • Hiromu Tanaka
    Division of Pulmonary Medicine, Department of Internal Medicine, Keio University School of Medicine, Tokyo, Japan.
  • Ho Lee
    Department of Nuclear Engineering, Hanyang University, Seoul, 02841, Republic of Korea.
  • Takahiro Fukushima
    Division of Pulmonary Medicine, Department of Internal Medicine, Keio University School of Medicine, Tokyo, Japan.
  • Shuhei Azekawa
    Division of Pulmonary Medicine, Department of Internal Medicine, Keio University School of Medicine, Tokyo, Japan.
  • Shiro Otake
    Division of Pulmonary Medicine, Department of Internal Medicine, Keio University School of Medicine, Tokyo, Japan.
  • Kensuke Nakagawara
    Division of Pulmonary Medicine, Department of Internal Medicine, Keio University School of Medicine, Tokyo, Japan.
  • Mayuko Watase
    Division of Pulmonary Medicine, Department of Internal Medicine, Keio University School of Medicine, Tokyo, Japan.
  • Katsunori Masaki
    Division of Pulmonary Medicine, Department of Internal Medicine, Keio University School of Medicine, Tokyo, Japan.
  • Hirofumi Kamata
    Division of Pulmonary Medicine, Department of Internal Medicine, Keio University School of Medicine, Tokyo, Japan.
  • Norihiro Harada
    Department of Respiratory Medicine, Faculty of Medicine, Graduate School of Medicine, Juntendo University, Tokyo, Japan.
  • Tetsuya Ueda
    Graduate School of Engineering, Chiba University, Chiba, Japan.
  • Soichiro Ueda
    JCHO (Japan Community Health Care Organization, Internal Medicine, Saitama Medical Center, Saitama, Japan.
  • Takashi Ishiguro
    Department of Respiratory Medicine, Saitama Cardiovascular and Respiratory Center, Saitama, Japan.
  • Ken Arimura
    Department of Respiratory Medicine, Tokyo Women's Medical University, Tokyo, Japan.
  • Fukuki Saito
    Department of Emergency and Critical Care Medicine, Kansai Medical University General Medical Center, Osaka, Japan.
  • Takashi Yoshiyama
    Respiratory Disease Center, Fukujuji Hospital, Tokyo, Japan.
  • Yasushi Nakano
    Department of Internal Medicine, Kawasaki Municipal Ida Hospital, Kawasaki, Kanagawa, Japan.
  • Yoshikazu Muto
    Department of Infectious Diseases, Tosei General Hospital, Aichi, Japan.
  • Yusuke Suzuki
    Department of Radiology, Fujita Health University Hospital.
  • Ryuya Edahiro
    Department of Statistical Genetics, Osaka University Graduate School of Medicine, Osaka, Japan.
  • Koji Murakami
    Department of Respiratory Medicine, Tohoku University Graduate School of Medicine, Miyagi, Japan.
  • Yasunori Sato
    Graduate School of Health Management, Keio University, Tokyo, Japan.
  • Yukinori Okada
    Department of Statistical Genetics, Osaka University Graduate School of Medicine, Suita, 565-0871, Japan. yokada@sg.med.osaksa-u.ac.jp.
  • Ryuji Koike
    Health Science Research and Development Center, Tokyo Medical and Dental University, Tokyo, Japan.
  • Makoto Ishii
    Department of Respiratory Medicine, Nagoya University Graduate School of Medicine, 65 Tsurumai-cho, Showa-ku, Nagoya, Aichi, 4668550, Japan.
  • Naoki Hasegawa
    Department of Pharmaceutics, Hoshi University, 2-4-41 Ebara, Shinagawa, Tokyo 142-8501 Japan.
  • Yuko Kitagawa
    Department of Surgery Keio University School of Medicine Tokyo Japan.
  • Katsushi Tokunaga
    Genome Medical Science Project (Toyama), National Center for Global Health and Medicine, Tokyo, Japan.
  • Akinori Kimura
    Institute of Research, Tokyo Medical and Dental University, Tokyo, Japan.
  • Satoru Miyano
    The Institute of Medical Science, The University of Tokyo, Shirokanedai 4-6-1, Minato-ku, Tokyo, 108-8639, Japan. miyano@ims.u-tokyo.ac.jp.
  • Seishi Ogawa
    Department of Pathology and Tumor Biology, Kyoto University, Kyoto, Japan.
  • Takanori Kanai
    Division of Gastroenterology and Hepatology, Department of Internal Medicine, Keio University School of Medicine, Tokyo, Japan.
  • Koichi Fukunaga
    Pulmonary Division, Department of Medicine, Keio University School of Medicine, 35 Shinanomachi Shinjuku-ku, Tokyo 160-8582, Japan. km-fuku@cpnet.med.keio.ac.jp.
  • Seiya Imoto
    The Institute of Medical Science, The University of Tokyo, Shirokanedai 4-6-1, Minato-ku, Tokyo, 108-8639, Japan.

Keywords

External Resources

Popular Topics
Recent Journals