M3T-LM: A multi-modal multi-task learning model for jointly predicting patient length of stay and mortality

被引:0
作者
Chen, Junde [1 ]
Li, Qing [2 ]
Liu, Feng [3 ]
Wen, Yuxin [1 ]
机构
[1] Dale E. and Sarah Ann Fowler School of Engineering, Chapman University, Orange, 92866, CA
[2] Department of Industrial and Manufacturing Systems Engineering, Iowa State University, Ames, 50011, IA
[3] School of Systems and Enterprises, Stevens Institute of Technology, Hoboken, 07030, NJ
基金
美国国家科学基金会;
关键词
Data-fusion model; Deep learning; Length of stay prediction; Multi-task learning;
D O I
10.1016/j.compbiomed.2024.109237
中图分类号
学科分类号
摘要
Ensuring accurate predictions of inpatient length of stay (LoS) and mortality rates is essential for enhancing hospital service efficiency, particularly in light of the constraints posed by limited healthcare resources. Integrative analysis of heterogeneous clinic record data from different sources can hold great promise for improving the prognosis and diagnosis level of LoS and mortality. Currently, most existing studies solely focus on single data modality or tend to single-task learning, i.e., training LoS and mortality tasks separately. This limits the utilization of available multi-modal data and prevents the sharing of feature representations that could capture correlations between different tasks, ultimately hindering the model's performance. To address the challenge, this study proposes a novel Multi-Modal Multi-Task learning model, termed as M3T-LM, to integrate clinic records to predict inpatients’ LoS and mortality simultaneously. The M3T-LM framework incorporates multiple data modalities by constructing sub-models tailored to each modality. Specifically, a novel attention-embedded one-dimensional (1D) convolutional neural network (CNN) is designed to handle numerical data. For clinical notes, they are converted into sequence data, and then two long short-term memory (LSTM) networks are exploited to model on textual sequence data. A two-dimensional (2D) CNN architecture, noted as CRXMDL, is designed to extract high-level features from chest X-ray (CXR) images. Subsequently, multiple sub-models are integrated to formulate the M3T-LM to capture the correlations between patient LoS and modality prediction tasks. The efficiency of the proposed method is validated on the MIMIC-IV dataset. The proposed method attained a test MAE of 5.54 for LoS prediction and a test F1 of 0.876 for mortality prediction. The experimental results demonstrate that our approach outperforms state-of-the-art (SOTA) methods in tackling mixed regression and classification tasks. © 2024 Elsevier Ltd
引用
收藏
相关论文
共 50 条
[41]   M2GSNet: Multi-Modal Multi-Task Graph Spatiotemporal Network for Ultra-Short-Term Wind Farm Cluster Power Prediction [J].
Fan, Hang ;
Zhang, Xuemin ;
Mei, Shengwei ;
Chen, Kunjin ;
Chen, Xinyang .
APPLIED SCIENCES-BASEL, 2020, 10 (21) :1-15
[42]   Multi-task deep learning based on T2-Weighted Images for predicting Muscular-Invasive Bladder Cancer [J].
Zou, Yuan ;
Cai, Lingkai ;
Chen, Chunxiao ;
Shao, Qiang ;
Fu, Xue ;
Yu, Jie ;
Wang, Liang ;
Chen, Zhiying ;
Yang, Xiao ;
Yuan, Baorui ;
Liu, Peikun ;
Lu, Qiang .
COMPUTERS IN BIOLOGY AND MEDICINE, 2022, 151
[43]   T3S: Improving Multi-Task Reinforcement Learning with Task-Specific Feature Selector and Scheduler [J].
Yu, Yuanqiang ;
Yang, Tianpei ;
Lv, Yongliang ;
Zheng, Yan ;
Hao, Jianye .
2023 INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS, IJCNN, 2023,
[44]   PS-MTL-LUCAS: A partially shared multi-task learning model for simultaneously predicting multiple soil properties [J].
Zhai, Zhaoyu ;
Chen, Fuji ;
Yu, Hongfeng ;
Hu, Jun ;
Zhou, Xinfei ;
Xu, Huanliang .
ECOLOGICAL INFORMATICS, 2024, 82
[45]   M3LA: A Novel Approach Based on Encoder-Decoder with Attention Framework for Multi-modal Multi-label Learning [J].
Zhu, Yinlong ;
Zhang, Yi .
2020 INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS (IJCNN), 2020,
[46]   Inter-organ correlation based multi-task deep learning model for dynamically predicting functional deterioration in multiple organ systems of ICU patients [J].
Zeng, Zhixuan ;
Liu, Yang ;
Yao, Shuo ;
Lin, Minjie ;
Cai, Xu ;
Nan, Wenbin ;
Xie, Yiyang ;
Gong, Xun .
BIODATA MINING, 2025, 18 (01)
[47]   A 3D end-to-end multi-task learning network for predicting lymph node metastasis at multiple nodal stations in gastric cancer [J].
Zhu, Hao ;
Yang, Zhi ;
Zheng, Chang ;
Jiang, Ping ;
Fang, Yi ;
Xu, Yuejie ;
Xiang, Ying ;
Xu, En ;
Wang, Lei ;
Bao, Shanhua ;
Guan, Wenxian ;
Zou, Xiaoping .
BIOMEDICAL SIGNAL PROCESSING AND CONTROL, 2025, 108
[48]   A transformer-based multi-task deep learning model for simultaneous T-stage identification and segmentation of nasopharyngeal carcinoma [J].
Yang, Kaifan ;
Dong, Xiuyu ;
Tang, Fan ;
Ye, Feng ;
Chen, Bei ;
Liang, Shujun ;
Zhang, Yu ;
Xu, Yikai .
FRONTIERS IN ONCOLOGY, 2024, 14
[49]   A MULTI-TASK DEEP LEARNING MODEL FOR POPULATION AND LULC (M2PL-NET) PREDICTION WITH SCALING TO A PEOPLE FLOW GRID [J].
Vinayaraj, Poliyapram ;
Anderson, Jeremiah Luke ;
Mayank, Bansal .
2022 IEEE INTERNATIONAL GEOSCIENCE AND REMOTE SENSING SYMPOSIUM (IGARSS 2022), 2022, :135-138
[50]   Fully automated 3D multi-modal deep learning model for preoperative T-stage prediction of colorectal cancer using 18F-FDG PET/CT [J].
Zhang, Mobei ;
Li, Yufan ;
Zheng, Chunhong ;
Xie, Fei ;
Zhao, Zhenwei ;
Dai, Fahao ;
Wang, Jiarou ;
Wu, Hubing ;
Zhu, Zhaohui ;
Liu, Qingxing ;
Li, Yinfeng .
EUROPEAN JOURNAL OF NUCLEAR MEDICINE AND MOLECULAR IMAGING, 2025,