Jie Wang

Jie Wang

Auto-generated profile, based on publicly available information

Jie Wang

Fujitsu Research & Development Center Co.,LTD., Beijing, China

Research Focus & Metrics

Academic Career Length

32Years (1995-Present)

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

Representative Publications

View All Publications
9 November 2022 in IEEE Transactions on Instrumentation and Measurement
Verified
Journal Article
Closed access
JWJie Wang
Jie Wang
CZChaoliang Zhong
Chaoliang Zhong
CFCheng Feng
Cheng Feng
YZYing Zhang
Ying Zhang
JJJun JUN
Jun JUN
YYYasuto Yokota
Yasuto Yokota
Image segmentation is a long-standing problem in medical image analysis to facilitate the clinical diagnosis and intervention. Progress has been made owing to deep learning via supervised training with elaborate human labelling, however, the segmentation models trained by the labeled source domain cannot perform well in the target domain, making existing approaches lack robustness and generalization ability. Considering the acquisition of medical image labels is quite expensive and time-consuming, we propose a novel feature disentanglement-based unsupervised domain adaptation (UDA) method to improve the robustness of the trained model in the target domain. A segmentation network is designed to learn disentangled features with two parts: I. content-related features, which are responsible for the segmentation task and invariant across domains; II. style-related features, which elucidate the discrepancy between different domains. Feature disentanglement (FD) is achieved by multi-task learning and image translation. Meanwhile, knowledge distillation is introduced to improve the performance on fine-grained segmentations. And for objects with regular shape, we incorporate the adversarial training to predict shape-invariant segmentation masks across domains. Comprehensive experiments are conducted on retina vessel segmentation and sinus surgical instrument segmentation to validate the effectiveness of the proposed method. The average Dice of twenty regular transfer directions achieves 79.26% on five public benchmarks of retina vessel segmentation, the average Dice of two transfer directions from regular to UWF attains 72.63%, and the Dice from cadaveric images to live images reaches 68.1% on sinus surgical instrument segmentation. The results demonstrate that the proposed method achieves the state-of-the-art segmentation performance in the UDA setting.

15 September 2016 in BioMedical Engineering OnLine
Verified
Journal Article
Open access
PDF file
Yongming LiYongming Li
Yongming Li
XZXueru Zhu
Xueru Zhu
PWPin Wang
Pin Wang
JWJie Wang
Jie Wang
SLShujun Liu
Shujun Liu
MQMingguo Qiu
Mingguo Qiu
Amyloid β-protein (Aβ) plaque deposition is an important prevention and treatment target for Alzheimer’s disease (AD). As a noninvasive, nonradioactive and highly cost-effective clinical imaging method, magnetic resonance imaging (MRI) is the perfect imaging technology for the clinical diagnosis of AD, but it cannot display the plaque deposition directly. This paper resolves this problem based on pixel feature selection algorithms at the image level. Firstly, the brain region was segmented from mouse model brain MR images. Secondly, the pixels in the segmented brain region were extracted as a feature vector (features). Thirdly, feature selection was conducted on the extracted features, and the optimal feature subset was obtained. Fourthly, the various optimal feature subsets were obtained by repeating the same processing above. Fifthly, based on the optimal feature subsets, the final optimal feature subset was obtained by voting mechanism. Finally, using the final optimal selected features, the corresponding pixels on the MR images could be found and marked to show the information about Aβ plaque deposition. The MR images and brain histological image slices of twenty-two model mice were used in the experiments. Four feature selection algorithms were used on the MR images and six kinds of classification experiments are conducted, thereby choosing a pixel feature selection algorithm for further study. The experimental results showed that by using the pixel features selected by the algorithms in this paper, the best classification accuracy between early AD and control slides could be as high as 80 %. The selected and marked MR pixels could show information of Aβ plaque deposition without missing most of the Aβ plaque deposition compared with brain histological slice images. The hit rate is over than 90 %. According to the experimental results, the proposed detection algorithm of the Aβ plaque deposition based on MR pixel feature selection algorithm is effective. The proposed algorithm can detect the information of the Aβ plaque deposition on MR images and the information can be useful for improving the classification accuracy as assistant MR biomarker. Besides, these findings firstly show the feasibility of detection of the Aβ plaque deposition on MR images and provide reference method for interested relevant researchers in public.

10 November 2025
Verified
Conference Paper
Open access
JWJie Wang
Jie Wang
ZZZhongguang Zheng
Zhongguang Zheng
CZChaoliang Zhong
Chaoliang Zhong
JJJun JUN
Jun JUN

15 December 2023
Verified
Conference Paper
Closed access
WJWang Jie
Wang Jie
CXChen Xianzhi
Chen Xianzhi
WMWang Mingxin
Wang Mingxin
LXLiu Xinyue
Liu Xinyue
Object detection is a fundamental task for complex visual applications, with wide-ranging uses in various fields. However, detecting small objects in cluttered scenes is challenging, due to factors such as limited feature information, occlusion, and interference. To address the difficulties of small object detection, we propose an improved version of the YOLOv5 algorithm, which incorporates a sophisticated attention mechanism called YOLOv5-Sim. This method uses a three-dimensional simple parameter-free attention module (Simam) to reduce the influence of environmental noise on image targets. Simam considers both spatial and channel dimensions, and refines features to selectively enhance those related to small targets. Moreover, we modify the backbone network structure of YOLOv5, and use the ReLU function as the loss function to boost the expression of small targets and speed up their detection. We conduct experiments on the Flow-Img small target dataset from Okahublot Company, and compare our model with the original YOLOv5x. The results show that YOLOv5-Sim achieves an average mean accuracy map@0.5 of 89.4%, surpassing YOLOv5x by 3.1%, without adding any extra parameters. Our model balances accuracy and real-time performance, and is suitable for various small target application scenarios.


1 March 2002 in JACC
Verified
Journal Article
Closed access
EFEduard Fiehn
Eduard Fiehn
WRWilfried Roethy
Wilfried Roethy
AGAnguo Gu
Anguo Gu
KSKotaro Suehiro
Kotaro Suehiro
GYGeng-Hua Yi
Geng-Hua Yi
DBDaniel Burkhoff
Daniel Burkhoff