A5100378810- AI in cancer detection
- Radiomics and Machine Learning in Medical Imaging
- Natural Language Processing Techniques
- Topic Modeling
- Thyroid Cancer Diagnosis and Treatment
- Digital Imaging for Blood Diseases
- Face recognition and analysis
- Imbalanced Data Classification Techniques
- Advanced Computational Techniques and Applications
- Teratomas and Epidermoid Cysts
- COVID-19 diagnosis using AI
- Brain Tumor Detection and Classification
- Service-Oriented Architecture and Web Services
- Domain Adaptation and Few-Shot Learning
- Text and Document Classification Technologies
- Machine Learning and Algorithms
- Algorithms and Data Compression
- Advanced Vision and Imaging
- Tracheal and airway disorders
- Congenital Diaphragmatic Hernia Studies
- 3D Shape Modeling and Analysis
- Translation Studies and Practices
- Web Data Mining and Analysis
- Robotics and Sensor-Based Localization
Hong Kong University of Science and Technology
2020-2024
University of Hong Kong
2020-2024
Chenzhou First People's Hospital
2008-2022
Peking University
2019-2021
Peking University International Hospital
2019
China United Network Communications Group (China)
2008
IBM (United States)
2000
With the increasing cases of thyroid malignant tumors, diagnosis nodule has attracted more and attention. Deep learning achieved promising results in computer-aided due to advantages obtaining high-dimensional features. In this paper, we proposed a hybrid multi-branch convolutional neural network based on feature cropping method for extraction classification ultrasound images. Firstly, designed backbone extract shared maps as global branch. Next, added branch perform multi-cropping batch...
Article Free Access Share on Active learning using adaptive resampling Authors: Vijay S. Iyengar IBM Research Division, T.J. Watson Center, P.O. Box 218, Yorktown Heights, NY NYView Profile , Chidanand Apte Tong Zhang Authors Info & Claims KDD '00: Proceedings of the sixth ACM SIGKDD international conference Knowledge discovery and data miningAugust 2000 Pages 91–98https://doi.org/10.1145/347090.347110Online:01 August 2000Publication History 48citation1,216DownloadsMetricsTotal...
In this paper we study the problem of finding most topical named entities among all in a document, which refer to as focused entity recognition. We show that these are useful for many natural language processing applications, such document summarization, search result ranking, and detection tracking. propose statistical model recognition by converting it into classification problem. then impact various linguistic features compare number algorithms. From experiments on an annotated Chinese...
Tong Zhang, Long Wei Ye, Bo Li, Jinan Sun, Xiaoyu Zhu, Wen Zhao, Shikun Zhang. Proceedings of the 59th Annual Meeting Association for Computational Linguistics and 11th International Joint Conference on Natural Language Processing (Volume 1: Papers). 2021.
As deep learning technologies advance, increasingly more data is necessary to generate general and robust models for various tasks. In the medical domain, however, large-scale multi-parties training analyses are infeasible due privacy security concerns. this paper, we propose an extendable elastic framework preserve while enabling collaborative with efficient communication. The proposed named distributed Asynchronized Discriminator Generative Adversarial Networks (AsynDGAN), which consists...
Tracheal agenesis is a rare and uniformly lethal congenital malformation. We recently had the opportunity to diagnose prenatally case of type I tracheal without fistula in 22-week-old fetus, which hyperechogenicity enlarged lungs was associated with ascites, dilatation lower half. report ultrasonographic autopsy findings review literature on this condition contribute understanding 1 tracheoesophageal fistula.
In this paper, we propose a data privacy-preserving and communication efficient distributed GAN learning framework named Distributed Asynchronized Discriminator (AsynDGAN). Our proposed aims to train central generator learns from discriminator, use the generated synthetic image solely segmentation model.We validate on application of health entities problem which is known be privacy sensitive. experiments show that our approach: 1) could learn real image's distribution multiple datasets...