|
Time Series Multi-task Learning for Prognosis of MICU and SICU
Yen-Jung Chiu1, Szu-Hsien Wu2, Ping-Feng Wu3, Chao-Chun Chuang1, Ming-Liang Hsiao4, Mei-Jung Chen5, Pei-Ru Chen5 and Shih-Tsang Tang5*
1National Center for High-Performance Computing, Taiwan
2Department of Surgery, Taipei Veterans General Hospital, Taiwan
3Division of Infectious Diseases, Department of Medicine, Taipei Veterans General Hospital, Taiwan
4Division of Experiment Surgery, Department of Surgery, Taipei Veterans General Hospital, Taiwan
5Department of Biomedical Engineering, Ming Chuan University, Taiwan
|
Abstract:
The prognostic assessment of an ICU patient involves assessing the severity of their condition, interventions, and length of ICU stay. Over the past 30 years, researchers have proposed numerous predictive models and severity assessment scales for ICU patients in specific regions, including APACHE II and SAPS II. However, most existing methods rely heavily on curve fitting which do not account for misclassifications caused by false negatives and positives. Specificity and sensitivity must be provided as an indicator of model performance. The primary aim in this study is to develop a machine-learning model to formulate a prognosis for MICU and SICU patients by using data from the MIMIC-IV for training. The predictive models developed in this study facilitate the prediction of mortality and other outcomes across various treatment regimens.
|
|
Keywords: MICU, SICU, Prognostic assessment, and Predictive model
|
Download PDF
Received:May 10, 2022; Revised:June 01, 2022; Accepted:June 10, 2022; Published:June 30, 2022
|
|
*Corresponding author; e-mail: sttang@mail.mcu.edu.tw;
|
|
Citation:Chiu, Y.J.; Wu, S.H.; Wu, P.F.; Chuang, C.C.; Hsiao, M.L.; Chen, M.J.; Chen, P.R.; Tang, S.T.Time Series Multi-task Learning for Prognosis of MICU and SICU. International Journal of Clinical Medicine and Bioengineering 2022, 2, 55-62. https://doi.org/10.35745/ijcmb2022v02.02.0006
87 Views 97 Downloads |
Copyright: ©
2022
The Author(s). Published with license by IIKII, Singapore. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (CC BY), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
|