Dynamic predict in-hospital mortality risk in intensive care unit with a new deep learning of artificial intelligence
Abstract Background Dynamic prediction of patients’ mortality risk in ICU with time series data is limited due to the high dimensionality, uncertainty with sampling intervals, and other issues. New deep learning method, temporal convolution network (TCN), makes it possible to deal with complex clinical time series data in ICU. We aimed to develop and validate it to predict mortality risk using time series data from MIMIC III dataset. Methods Finally, 21139 records of ICU stays were analyzed and in total 17 physiological variables from the MIMIC III dataset were used to predict mortality risk. Then we compared the model performances of attention-based TCN with traditional artificial intelligence (AI) method. Results The Area Under Receiver Operating Characteristic (AUCROC) and Area Under Precision-Recall curve (AUC-PR) of attention-based TCN for predicting the mortality risk 48 h after ICU admission were 0.837(0.824–0.850) and 0.454. The sensitivity and specificity of attention-based TCN were 67.1% and 82.6%, compared to the traditional AI method yield low sensitivity (< 50%). Conclusions Attention-based TCN model achieved better performance in prediction of mortality risk with time series data than traditional AI methods and conventional score-based models. Attention-based TCN mortality risk model has the potential for helping decision-making in critical patients..
| Medienart: |
|
|---|
Preprint| Erscheinungsjahr: |
2021 |
|---|---|
| Erschienen: |
2021 |
| Enthalten in: |
ResearchSquare.com - (2021) vom: 17. März Zur Gesamtaufnahme - year:2021 |
|---|
| Sprache: |
Englisch |
|---|
| Beteiligte Personen: |
Chen, Yu-wen [VerfasserIn] |
|---|
| Links: |
Volltext [kostenfrei] |
|---|
| doi: |
10.21203/rs.3.rs-44310/v1 |
|---|
| funding: |
|
|---|---|
| Förderinstitution / Projekttitel: |
|
| PPN (Katalog-ID): |
XRA034137289 |
|---|
| LEADER | 01000caa a22002652 4500 | ||
|---|---|---|---|
| 001 | XRA034137289 | ||
| 003 | DE-627 | ||
| 005 | 20230429181516.0 | ||
| 007 | cr uuu---uuuuu | ||
| 008 | 220113s2021 xx |||||o 00| ||eng c | ||
| 024 | 7 | |a 10.21203/rs.3.rs-44310/v1 |2 doi | |
| 035 | |a (DE-627)XRA034137289 | ||
| 035 | |a (ResearchSquare)10.21203/rs.3.rs-44310/v1 | ||
| 040 | |a DE-627 |b ger |c DE-627 |e rakwb | ||
| 041 | |a eng | ||
| 082 | 0 | |a 570 |q DE-84 | |
| 100 | 1 | |a Chen, Yu-wen |e verfasserin |4 aut | |
| 245 | 1 | 0 | |a Dynamic predict in-hospital mortality risk in intensive care unit with a new deep learning of artificial intelligence |
| 264 | 1 | |c 2021 | |
| 336 | |a Text |b txt |2 rdacontent | ||
| 337 | |a Computermedien |b c |2 rdamedia | ||
| 338 | |a Online-Ressource |b cr |2 rdacarrier | ||
| 520 | |a Abstract Background Dynamic prediction of patients’ mortality risk in ICU with time series data is limited due to the high dimensionality, uncertainty with sampling intervals, and other issues. New deep learning method, temporal convolution network (TCN), makes it possible to deal with complex clinical time series data in ICU. We aimed to develop and validate it to predict mortality risk using time series data from MIMIC III dataset. Methods Finally, 21139 records of ICU stays were analyzed and in total 17 physiological variables from the MIMIC III dataset were used to predict mortality risk. Then we compared the model performances of attention-based TCN with traditional artificial intelligence (AI) method. Results The Area Under Receiver Operating Characteristic (AUCROC) and Area Under Precision-Recall curve (AUC-PR) of attention-based TCN for predicting the mortality risk 48 h after ICU admission were 0.837(0.824–0.850) and 0.454. The sensitivity and specificity of attention-based TCN were 67.1% and 82.6%, compared to the traditional AI method yield low sensitivity (< 50%). Conclusions Attention-based TCN model achieved better performance in prediction of mortality risk with time series data than traditional AI methods and conventional score-based models. Attention-based TCN mortality risk model has the potential for helping decision-making in critical patients. | ||
| 700 | 1 | |a Li, Yu-jie |e verfasserin |4 aut | |
| 700 | 1 | |a Yang, Zhi-yong |e verfasserin |4 aut | |
| 700 | 1 | |a Zhong, Kun-hua |e verfasserin |4 aut | |
| 700 | 1 | |a Zhang, Li-ge |e verfasserin |4 aut | |
| 700 | 1 | |a Chen, Yang |e verfasserin |4 aut | |
| 700 | 1 | |a Zhi, Hong-yu |e verfasserin |4 aut | |
| 700 | 1 | |a Deng, Peng |e verfasserin |4 aut | |
| 700 | 1 | |a Wang, Dan-dan |e verfasserin |4 aut | |
| 700 | 1 | |a Gu, Jian-teng |e verfasserin |4 aut | |
| 700 | 1 | |a Ning, Jiao-lin |e verfasserin |4 aut | |
| 700 | 1 | |a Lu, Kai-zhi |e verfasserin |4 aut | |
| 700 | 1 | |a Zhang, Ju |e verfasserin |4 aut | |
| 700 | 1 | |a Xia, Zheng-yuan |e verfasserin |4 aut | |
| 700 | 1 | |a Yi, Bin |e verfasserin |4 aut | |
| 773 | 0 | 8 | |i Enthalten in |t ResearchSquare.com |g (2021) vom: 17. März |
| 773 | 1 | 8 | |g year:2021 |g day:17 |g month:03 |
| 856 | 4 | 0 | |u http://dx.doi.org/10.21203/rs.3.rs-44310/v1 |z kostenfrei |3 Volltext |
| 912 | |a GBV_XRA | ||
| 912 | |a SSG-OLC-PHA | ||
| 951 | |a AR | ||
| 952 | |j 2021 |b 17 |c 03 | ||
| 953 | |2 045F |a 570 | ||