Zhenhua Sang
Research Subjects
Zhenhua Sang has a total of 12 co-authors. Here are the top co-authors by number of shared publications:
- Guijin Wang
Department of Electronic Engineering, Tsinghua University, Beijing, China
2 shared publications - Huazhong Yang
Tsinghua University
2 shared publications - Ming Chen
Department of Electronic Engineering, Tsinghua University, Beijing, China
2 shared publications - Pengwei Xie
Tsinghua University
2 shared publications - Ping Zhang
Division of Toxicology, Bloomberg School
2 shared publications - Tingting Lv
Department of Cardiology, Beijing Tsinghua Changgung Hospital, School of Clinical Medicine, Tsinghua University, 102218 Beijing, China
2 shared publications
Latest Publications
- 24 June 2025
- Published by Elsevier in Biomedical Signal Processing and Control
- Vol. 110
Subject-adaptive SSVEP decoding based on time–frequency information
- 24 June 2025
- Published by Elsevier in Biomedical Signal Processing and Control
- Vol. 110
Subject-adaptive SSVEP decoding based on time–frequency information
- 1 December 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1306-1311
Approximate Quality Assessment with Sampling Approaches
Data is useful to the extent that it can be quickly analyzed to reveal valuable information. With high-quality data, we can increase revenue, reduce cost, and reduce risk. On the other hand, the consequences of poor-quality data can be severe. It has been estimated that poor quality customer data costs U.S. businesses $611 billion annually in postage, printing, and staff overhead. These issues make data quality assessment a necessary and critical step in any data-related systems. Big data brings new challenges to data quality assessment due to the scale of data, streaming data, and different forms of data. Therefore, we proposed a sampling-based approximate quality assessment model on large data. Sampling large datasets can make all quality assessment processes cheaper and more feasible because of data reduction. The protocol of this work: First, the sample size is determined for estimating a large dataset. Next, sampling techniques are applied to collect samples. Then, these samples are used to estimate the quality of the large dataset. The objective of quality assessment in this work is to evaluate the completeness, accuracy, and timeliness of data and to return fast and approximate scores. Using different sample sizes and different sampling methods, we obtained 72 sets of data and compared them. These results show that the proposed approach is efficient and provides some insight into the quality assessment with samples.
- 1 December 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1306-1311
Approximate Quality Assessment with Sampling Approaches
Data is useful to the extent that it can be quickly analyzed to reveal valuable information. With high-quality data, we can increase revenue, reduce cost, and reduce risk. On the other hand, the consequences of poor-quality data can be severe. It has been estimated that poor quality customer data costs U.S. businesses $611 billion annually in postage, printing, and staff overhead. These issues make data quality assessment a necessary and critical step in any data-related systems. Big data brings new challenges to data quality assessment due to the scale of data, streaming data, and different forms of data. Therefore, we proposed a sampling-based approximate quality assessment model on large data. Sampling large datasets can make all quality assessment processes cheaper and more feasible because of data reduction. The protocol of this work: First, the sample size is determined for estimating a large dataset. Next, sampling techniques are applied to collect samples. Then, these samples are used to estimate the quality of the large dataset. The objective of quality assessment in this work is to evaluate the completeness, accuracy, and timeliness of data and to return fast and approximate scores. Using different sample sizes and different sampling methods, we obtained 72 sets of data and compared them. These results show that the proposed approach is efficient and provides some insight into the quality assessment with samples.
- 1 July 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 2018, 2559-2562
Region Aggregation Network: Improving Convolutional Neural Network for ECG Characteristic Detection
Detection of ECG characteristic points serves as the first step in automated ECG analysis techniques. We propose a novel end-to-end deep learning scheme called Region Aggregation Network (RAN) for ECG characteristic points de- tection. A 1D Convolutional Neural Network (CNN) is adopted to automatically process ECG signals. A novel strategy of Region Aggregation is proposed to replace the conventional fully connected layer as regressor. Our work provides robust and accurate detection performance on public ECG database. The evaluation results of our method on QT database show comparable detection accuracy compared with state-of-the-art works.
- 1 July 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 2018, 2559-2562
Region Aggregation Network: Improving Convolutional Neural Network for ECG Characteristic Detection
Detection of ECG characteristic points serves as the first step in automated ECG analysis techniques. We propose a novel end-to-end deep learning scheme called Region Aggregation Network (RAN) for ECG characteristic points de- tection. A 1D Convolutional Neural Network (CNN) is adopted to automatically process ECG signals. A novel strategy of Region Aggregation is proposed to replace the conventional fully connected layer as regressor. Our work provides robust and accurate detection performance on public ECG database. The evaluation results of our method on QT database show comparable detection accuracy compared with state-of-the-art works.
- 1 July 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 2018, 2555-2558
Bidirectional Recurrent Neural Network And Convolutional Neural Network (BiRCNN) For ECG Beat Classification
We propose a novel electrocardiogram (ECG) beat classification algorithm using a combination of Bidirectional Recurrent Neural Network (BiRNN) and Convolutional Neural Network (CNN) named as BiRCNN. Our model is an end-to-end model. The morphological features of each ECG beat is extracted by CNN. Then the features of each beat are considered in the context via BiRNN. The assessment on MIT-BIH Arrhythmia Database (MITDB) resulted in a sensitivity of 98.7% and a positive predictivity of 96.4% on average for the VEB class. For the SVEB class, the sensitivity was 92.8%, which was an over 6% promotion compared with the state-of-the-art method, and the positive predictivity was 81.9% on average. The results demonstrate the superior classification performance of our method.
- 1 July 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 2018, 2555-2558
Bidirectional Recurrent Neural Network And Convolutional Neural Network (BiRCNN) For ECG Beat Classification
We propose a novel electrocardiogram (ECG) beat classification algorithm using a combination of Bidirectional Recurrent Neural Network (BiRNN) and Convolutional Neural Network (CNN) named as BiRCNN. Our model is an end-to-end model. The morphological features of each ECG beat is extracted by CNN. Then the features of each beat are considered in the context via BiRNN. The assessment on MIT-BIH Arrhythmia Database (MITDB) resulted in a sensitivity of 98.7% and a positive predictivity of 96.4% on average for the VEB class. For the SVEB class, the sensitivity was 92.8%, which was an over 6% promotion compared with the state-of-the-art method, and the positive predictivity was 81.9% on average. The results demonstrate the superior classification performance of our method.