Xuyang Cao

Senior Engineer, Ph.D.

JD.com, China
No. 18 Kechuang 11 Street, Tongzhou District, Beijing
#A JD Building

Email: caoxuyang [at] bjtu [dot] edu [dot] cn
newxuyangcao [at] gmail [dot] com


Biography

Dr. Xuyang Cao is a senior engineer at JD.com. His research interests include machine learning, computer vision, and artificial intelligence. Specially, his research mainly focuses on image segmentation, image super-resolution, semi-supervised learning, medical image analysis, etc.

Before joining JD.com, he obtained his Ph.D degree in Beijing Jiaotong University in April 2022, supervised by professor Houjin Chen and professor Yanfen Li. He also worked closely with professor Yahui Peng.

Education

2017-2022 Ph.D,   School of Electronic and Information Engineering,   Beijing Jiaotong University
2016-2017 Master Candidate,   School of Electronic and Information Engineering,   Beijing Jiaotong University
2012-2016 Bachelor of Engineering,   School of Electronic and Information Engineering,   Beijing Jiaotong University

Projects

2019-2022 Research on Deep Learning based Breast Mass (Ultrasound) and Lung Mass (MRI) Segmentation ,  The National Natural Science Fund,   Participant
  • Research on deep learning based semantic segmentation algorithms on breast ultrasound images as well as lung MRI images.
  • Dig deep into semantic segmentation algorithms, such as fully supervised learning, semi-supervised learning, neural network architecture search, etc.
  • Proposed lightweight dilated densely connected network for 3D breast rumor segmentation. The performance improved over 5% compared with classical segmentation networks, while network parameters were over 20 times smaller than classical networks.
  • Designed an uncertainty-aware temporal-ensembling model for semi-supervised segmentation. The performance of semi-supervised method is able to achieve 94.4% of that in supervised segmentation with only 1.1% labeled data.
  • Suggested an NAS-based 3D medical image segmentation framework, and achieved an improvement of 4.2% compared with the state-of-art human-designed segmentation network.
  • Related journal and conference papers have been published.
2017-2018 High Speed Train Gear Defect Detection Based on Computer Vision,   Horizontal Scientific Research Project,   Participant
  • Provide solution for detection and quantitative analysis of surface pitting in high-speed train gears, such as getting defect status of gear tooth surface, reporting statistic results, etc.
  • Developed novel detection and segmentation algorithms as well as software platform.
  • Related patents, journal and conference papers have been published.
2016-2017 Geometric Parameters Measurement of an Overhead Line System,   Participant
  • Provide solution for measuring the geometric parameters of an overhead line system using scale factors and frame differences.
  • I was responsible for the previous algorithm simulation and participated in the hardware structure design work.
  • Related patents have been granted.

Publications

2022 X. Cao, H. Chen, Y. Li, Y. Peng, Y. Zhou, L. Cheng, T. Liu, D. Shen. Auto-DenseUnet: Searchable Neural Network Architecture for Tumor Segmentation in 3D Automated Breast Ultrasound. Medical Image Analysis, 2022, 82: 102589. [paper]

Y. Zhou, H. Chen, Y. Li, X. Cao, S. Wang, D. Shen. Cross-Model Attention-Guided Tumor Segmentation for 3D Automated Breast Ultrasound (ABUS) Images. IEEE Journal of Biomedical and Health Informatics, 2022, 26(1): 301-311. [paper]

2021 X. Cao, H. Chen, Y. Li, Y. Peng, S. Wang, L. Cheng. Uncertainty Aware Temporal-Ensembling Model for Semi-supervised ABUS Mass Segmentation. IEEE Transactions on Medical Imaging, 2021, 40(1):431-443. [paper]

X. Cao, H. Chen, Y. Li, Y. Peng, S. Wang, L. Cheng. Dilated Densely Connected U-Net with Uncertainty Focus Loss for 3D ABUS Mass Segmentation. Computer Methods and Programs in Biomedicine, 2021, 209: 106313. [paper]

J. Li, H. Chen, Y. Li, Y. Peng, N. Cai, X. Cao. AMRSegNet: Adaptive Modality Recalibration Network for Lung Tumor Segmentation on Multi-Modal MR Images. Multimedia Tools and Applications, 2021, 80: 33779–33797. [paper]

2020 X. Cao, H. Chen, Y. Li, Y. Peng, Y. Zhou, L. Cheng. Boundary Loss with Non-Euclidean Distance Constraint for ABUS Mass Segmentation. 2020 CISP-BMEI, Chengdu, China, 2020, pp: 645-650. [paper]

2019 Y. Peng, X. Cao, H. Chen, Y. Li, J. Li, X. Wang. Preliminary Study on Noise and Artifact Reduction in Phase-Contrast CT Image of Tristructural-Isotropic Coated Fuel Particle (in Chinese). Acta Electronica Sinica, 2019, 47(2): 448-453. [paper]

2018 C. Wang, F. Li, Y. Li, H. Chen and X. Cao. A Defect Status Detecting Method for External Gear in Railway. 2018 IEEE 3rd International Conference on Image, Vision and Computing (ICIVC), Chongqing, 2018, pp: 123-127. [paper]

Y. Li, X. Cao, H. Chen, L. Zhang, N. Yang. Defect Status Detection Method Based on Machine Vision for External Gear in Train (in Chinese). Journal of The China Railway Society, 2018, 40(12):33-41. [paper]

2017 J. Wei, X. Cao, H. chen, Y. li. Research on benign and malignant masses classification in mammogram (in Chinese). Journal of Beijing Jiaotong University, 2017, 41(5): 73-. [paper]

Patents

2018 Y. Peng, W. Jiang, Z. Zhu, H. Yang, X. Cao, H. Chen. A method of Measuring the Geometric Parameters of an Overhead Line System by using Geometric Magnification and Monocular Vision. China, CN201810182553.1, 2018-11-13. [Link]

2017 Y. Peng, C. Zhang, B. Zheng, J. Yin, X. Cao, H. Chen. A method and a Device for Measuring the Geometric Parameters of an Overhead Line System by using Scale Factors and Frame Differences. China, CN201710464403.5, 2017-06-19. [Link]

Translated Book

2022 Y. Zhou, X. Cao. Neural Networks with TensorFlow 2, Apress, 2020. [Link]

Language Skills

English IELTS (Band 7), CET-6
Chinese Mother Tongue

Computer Skills

Advanced Python, Pytorch, Vim
Intermediate Linux, C++, OpenCV, LaTex