Email : tangyuxing87 [AT] gmail [DOT] com
| CV | Google Scholar | |
Yuxing Tang currently holds the position of Staff Algorithm Engineer at Alibaba DAMO Academy, USA. Prior to this role, he served as a Senior Research Scientist at PAII Inc. before being promoted to Staff. He also worked as a Postdoctoral Visiting Fellow at the National Institutes of Health (NIH) between 2017 and 2020. Yuxing earned his Ph.D. degree in Computer Science from Ecole Centrale de Lyon, France, in 2016. He completed his bachelor's and master's degrees at Beijing Jiaotong University, China. His research expertise lies in the fields of computer vision and deep learning, with a particular focus on their applications in medical imaging, computer-aided diagnosis, and weakly supervised object/lesion detection. |
|
| Journal Articles | Conference Proceedings | |
|
We found that our mammographic CAD system could assist mid-level radiologists in effectively reducing unnecessary BI-RADS 0 mammogram recalls and the benign biopsy rate without missing highly malignant tumors. |
|
MommiNet-v2 aggregates information from the high-resolution representations of all mammographic views and incorporates the malignancy information from both biopsy and BI-RADS categories. |
|
A novel disentangled generative deep model for chest X-ray decomposition that can interpret chest X-rays by outputting disease residue or saliency maps, to improve radiology workflow and patient care. |
|
A two-step solution for few-shot learning for disease classification in chest X-rays. |
|
We develop a framework for rapidly constructing a chest X-ray/CT database containing Covid-19 images from PubMed Central® (PMC) full-text articles. |
|
AI algorithms can distinguish normal and abnormal chest X-rays with accuracy comparable to that of experienced radiologists, allowing these studies to be triaged for priority review. |
|
LENS (Lesion ENSemble) is a universal lesion detection framework that can effectively learn with multiple heterogeneous datasets and mine missing annotations from partially-labeled datasets. |
|
Can knowledge about visual and semantic similarities of object categories help improve the performance of detectors trained in a semi- or weakly- supervised setting? |
|
Using region proposals to improve the weakly supervised deformable part-based model. |
|
A sparse global/local affinity graph over superpixels of an input image to capture both short and long range grouping cues. |
|
A novel self-training framework for semi-supervised mass detection with soft image-level labels generated from diagnosis reports by a RoBERTa-based NLP model. |
|
A deep learning-based BI-RADS classification for individual calcification in mammograms. A new evaluation metric for BI-RADS classification which considers the severity of malignancy. |
|
A framework of supervised contrastive pre-training followed by supervised fine-tuning to improve mammographic triage screening models. |
|
A two-stage framework (edge enhanced network) for 2D liver and tumor segmentation on CT scans. |
|
Elsevier MedIA/ PR / PRL / NEUCOM Nature Publication Group npj Digital Medicine / Scientific Reports RSNA Radiology / Radiology: AI / Cardiothoracic Imaging IET Computer Vision / Signal Processing / Image Processing / Electronics Letters MDPI Sensors / Remote Sensing / Applied Sciences / Algorithms SPIE Journal of Electronic Imaging Medical Physics, PLOS ONE |
|