Prediction of the duration needed to achieve culture negativity in patients with active pulmonary tuberculosis using convolutional neural networks and chest radiography.

Journal: Respiratory investigation
Published Date:

Abstract

BACKGROUND: We aimed to predict the duration needed to achieve culture negativity in patients with active pulmonary tuberculosis using convolutional neural networks (CNNs) and chest radiography.

Authors

  • Masayoshi Higashiguchi
    Department of Internal Medicine, Osaka Anti-Tuberculosis Association Osaka Hospital, Neyagawa, Osaka, Japan. Electronic address: dr-higashiguchi@osaka-hospital.jp.
  • Koji Nishioka
    Department of Internal Medicine, Osaka Anti-Tuberculosis Association Osaka Hospital, Neyagawa, Osaka, Japan. Electronic address: dr-nishioka@osaka-hospital.jp.
  • Hiromi Kimura
    Department of Internal Medicine, Osaka Anti-Tuberculosis Association Osaka Hospital, Neyagawa, Osaka, Japan. Electronic address: dr-kimura@osaka-hospital.jp.
  • Tomoshige Matsumoto
    Department of Internal Medicine, Osaka Anti-Tuberculosis Association Osaka Hospital, Neyagawa, Osaka, Japan. Electronic address: tomoshige.matsumoto@gmail.com.