Personalized Heart Disease Detection via ECG Digital Twin Generation
Authors: Yaojun Hu, Jintai Chen, Lianting Hu, Dantong Li, Jiahuan Yan, Haochao Ying, Huiying Liang, Jian Wu
IJCAI 2024 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Experiments demonstrate that our approach not only excels in generating high-fidelity ECG signals but also improves personalized heart disease detection. Moreover, our approach ensures robust privacy protection, safeguarding patient data in model development. |
| Researcher Affiliation | Academia | Yaojun Hu1 , Jintai Chen2, , Lianting Hu3,4,5 , Dantong Li3,4,5 , Jiahuan Yan1 , Haochao Ying6 , Huiying Liang3,4,5 , Jian Wu6,7 1College of Computer Science and Technology, Zhejiang University 2Computer Science Department, University of Illinois at Urbana-Champaign 3Medical Big Data Center, Guangdong Provincial People s Hospital, Southern Medical University 4Guangdong Cardiovascular Institute, Guangdong Provincial People s Hospital, Guangdong Academy of Medical Sciences 5Guangdong Provincial Key Laboratory of Artificial Intelligence in Medical Image Analysis and Application, Guangdong Provincial People s Hospital (Guangdong Academy of Medical Sciences) 6School of Public Health, Zhejiang University 7State Key Laboratory of Transvascular Implantation Devices of The Second Affiliated Hospital, Zhejiang University School of Medicine |
| Pseudocode | No | The paper describes the model architecture and procedures in text and diagrams (Figure 1, Figure 2), but does not include structured pseudocode or algorithm blocks. |
| Open Source Code | Yes | The code can be found at https://github.com/huyjj/LAVQ-Editor. |
| Open Datasets | Yes | In our experiments, we utilize the PTB-XL dataset, a comprehensive and publicly accessible collection of electrocardiograms, which contains 21,837 clinical 12-lead ECGs from 18,885 individuals. |
| Dataset Splits | Yes | To ensure a robust test of our model s capabilities, we selected 291 patients who have both normal and disease ECG records for testing, while allocating the remaining patients data for training and validation purposes. |
| Hardware Specification | Yes | All subsequent experiments are executed using the Py Torch 1.9 on an NVIDIA Ge Force RTX-3090 GPU. |
| Software Dependencies | Yes | All subsequent experiments are executed using the Py Torch 1.9 on an NVIDIA Ge Force RTX-3090 GPU. |
| Experiment Setup | Yes | For the training process of LAVQ-Editor, we employ the Adam optimization algorithm, with a learning rate set at 0.00001 and a weight decay of 0.001. The model undergo a rigorous training regime spanning 500 epochs and is processed with a batch size of 128. The threshold l is set as 0.5 in the following experiments and we provide ablation study of the threshold in the supplements. All subsequent experiments are executed using the Py Torch 1.9 on an NVIDIA Ge Force RTX-3090 GPU. |