Jie Gao

Jie Gao

Auto-generated profile, based on publicly available information

Jie Gao

Key Laboratory of Intelligent Perception and Image Understanding of the Ministry of Education of China, International Research Center of Intelligent Perception and Computation, School of Artificial Intelligence, Xidian University, Xian, China

Research Focus & Metrics

Academic Career Length

15Years (2012-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
28 November 2022 in Computers in Biology and Medicine
Verified
Journal Article
Closed access
XLXuewei Li
Xuewei Li
SPShuo Pang
Shuo Pang
RZRuixuan Zhang
Ruixuan Zhang
JZJialin Zhu
Jialin Zhu
XFXuzhou Fu
Xuzhou Fu
JGJie Gao
Jie Gao

14 December 2021 in IEEE Transactions on Artificial Intelligence
Verified
Journal Article
Closed access
LJLicheng Jiao
Licheng Jiao
JGJie Gao
Jie Gao
XLXu Liu
Xu Liu
LFLiu Fang
Liu Fang
SYShuyuan Yang
Shuyuan Yang
BHBiao Hou
Biao Hou
Feature representation has been widely used and developed recently. Multiscale features have led to remarkable breakthroughs for representation learning process in many computer vision tasks. This paper aims to provide a comprehensive survey of the recent multiscale representation learning achievements in classification tasks. Multiscale representation learning methods can be divided into two broad categories (multiscale geometric analysis and multiscale networks). Eleven kinds of multiscale geometric tools and seven kinds of multiscale networks are introduced. Some corresponding fundamental subproblems of these two broad categories are also described, including some concepts in representation process, specific representation methods with multiscale geometric analysis, and multiscale representation design strategies for networks. Then, the correlation between these two broad categories is illustrated, including respective characteristics, combination strategies, and characteristics of optimal representation. Some datasets and evaluation results are included to verify the effectiveness of the multiscale representation learning. Eventually, conclusion and future work are given, covering four directions [a) Choice and fusion; b) Self-adaption; c) Structure; and d) Generalization and proof].

22 April 2022 in IEEE Access
Verified
Journal Article
Open access
LWLiang-Hung Wang
Liang-Hung Wang
YYYan-Ting Yu
Yan-Ting Yu
WLWei Liu
Wei Liu
LXLu Xu
Lu Xu
CXChao-Xin Xie
Chao-Xin Xie
PAPatricia Angela Abu
Patricia Angela Abu
Electrocardiogram (ECG) is the primary basis for the diagnosis of cardiovascular diseases. However, the amount of ECG data of patients makes manual interpretation time-consuming and onerous. Therefore, the intelligent ECG recognition technology is an important means to decrease the shortage of medical resources. This study proposes a novel classification method for arrhythmia that uses for the very first time a three-heartbeat multi-lead (THML) ECG data in which each fragment contains three complete heartbeat processes of multiple ECG leads. The THML ECG data pre-processing method is formulated which makes use of the MIT-BIH arrhythmia database as training samples. Four arrhythmia classification models are constructed based on one-dimensional convolutional neural network (1D-CNN) combined with a priority model integrated voting method to optimize the integrated classification effect. The experiments followed the recommended inter-patient scheme of the Association for the Advancement of Medical Instrumentation (AAMI) recommendations, and the practicability and effectiveness of THML ECG data are proved with ablation experiments. Results show that the average accuracy of the N, V, S, F, and Q classes is 94.82%, 98.10%, 97.28%, 98.70%, and 99.97%, respectively, with the positive predictive value of the N, V, S, and F classes being 97.0%, 90.5%, 71.9%, and 80.4%, respectively. Compared with current studies, the THML ECG data can effectively improve the morphological integrity and time continuity of ECG information and the 1D-CNN model of ECG sequence has a higher accuracy for arrhythmia classification. The proposed method alleviates the problem of insufficient samples, meets the needs of medical ECG interpretation and contributes to the intelligent dynamic research of cardiac disease.

15 October 2021 in IEEE Transactions on Neural Networks and Learning Systems
Verified
Journal Article
Closed access
JGJie Gao
Jie Gao
LJLicheng Jiao
Licheng Jiao
LFLiu Fang
Liu Fang
SYShuyuan Yang
Shuyuan Yang
BHBiao Hou
Biao Hou
XLXu Liu
Xu Liu
Feature representation has received more and more attention in image classification. Existing methods always directly extract features via convolutional neural networks (CNNs). Recent studies have shown the potential of CNNs when dealing with images' edges and textures, and some methods have been explored to further improve the representation process of CNNs. In this article, we propose a novel classification framework called the multiscale curvelet scattering network (MSCCN). Using the multiscale curvelet-scattering module (CCM), image features can be effectively represented. There are two parts in MSCCN, which are the multiresolution scattering process and the multiscale curvelet module. According to multiscale geometric analysis, curvelet features are utilized to improve the scattering process with more effective multiscale directional information. Specifically, the scattering process and curvelet features are effectively formulated into a unified optimization structure, with features from different scale levels being efficiently aggregated and learned. Furthermore, a one-level CCM, which can essentially improve the quality of feature representation, is constructed to be embedded into other existing networks. Extensive experimental results illustrate that MSCCN achieves better classification accuracy when compared with state-of-the-art techniques. Eventually, the convergence, insight, and adaptability are evaluated by calculating the trend of loss function's values, visualizing some feature maps, and performing generalization analysis.

25 November 2022 in IEEE Transactions on Neural Networks and Learning Systems
Verified
Journal Article
Closed access
JGJie Gao
Jie Gao
LJLicheng Jiao
Licheng Jiao
XLXu Liu
Xu Liu
LLLingling Li
Lingling Li
PCPuhua Chen
Puhua Chen
SYShuyuan Yang
Shuyuan Yang
The feature representation learning process greatly determines the performance of networks in classification tasks. By combining multiscale geometric tools and networks, better representation and learning can be achieved. However, relatively fixed geometric features and multiscale structures are always used. In this article, we propose a more flexible framework called the multiscale dynamic curvelet scattering network (MSDCCN). This data-driven dynamic network is based on multiscale geometric prior knowledge. First, multiresolution scattering and multiscale curvelet features are efficiently aggregated in different levels. Then, these features can be reused in networks flexibly and dynamically, depending on the multiscale intervention flag. The initial value of this flag is based on the complexity assessment, and it is updated according to feature sparsity statistics on the pretrained model. With the multiscale dynamic reuse structure, the feature representation learning process can be improved in the following training process. Also, multistage fine-tuning can be performed to further improve the classification accuracy. Furthermore, a novel multiscale dynamic curvelet scattering module, which is more flexible, is developed to be further embedded into other networks. Extensive experimental results show that better classification accuracies can be achieved by MSDCCN. In addition, necessary evaluation experiments have been performed, including convergence analysis, insight analysis, and adaptability analysis.

1 December 2012 in Applied Mechanics and Materials
Verified
Journal Article
Closed access
PDF file
SZSheng Zhang
Sheng Zhang
JGJie Gao
Jie Gao
JYJie Yang
Jie Yang
SYShun Yu
Shun Yu
A Mallat based wavelet de-noising algorithm in ECG analysis is studied. We use bior3.7 wavelet based on Mallat algorithm for ECG decomposition. Then we choose composite threshold and wavelet reconfiguration algorithm for signal de-noising to achieve an effective result. Data get from MIT/BIH is examined using the method. The result shows that it can not only remove the power frequency disturbance, EMG interference and baseline drift emerging in ECG, but also preserve the ECG characteristics.