Skip to Main Content

An official website of the United States government

Find Unique Patterns from Pathology Images

Principal Investigator
Name
Tian Xia
Degrees
PhD
Institution
School of Electronic Information and Communications, Huazhong University of Sci. & Tech.
Position Title
Associate Dean, Professor
Email
About this CDAS Project
Study
NLST (Learn more about this study)
Project ID
NLST-466
Initial CDAS Request Approval
Jan 7, 2019
Title
Find Unique Patterns from Pathology Images
Summary
We assume we can find some unique image patterns from pathology images. In complex image analysis tasks , DNN is a robust and durable tools to solve problems. So, we proposed a Deep Neural Network (DNN) model to process pathology images. And we build an unsupervised learning architecture in order to discover unique patterns which cannot be recognized by humans’ vision. These patterns may very complex so we can’t connect the valid part in our brains directly. We must use computers to discover the law.

We need more data to train our model and tune the hyper parameters. NLST pathology images may help us to improve our DNN architecture.
Aims

1. Develop an unsupervised learning architecture to find some unique image patterns from pathology images.
2. Analyze the meanings of these patterns and do advanced research on them.

Collaborators

1. Zitong He, School of Electronic Information and Communications, Huazhong University of Sci. and Tech.
2. Tong Wang, School of Electronic Information and Communications, Huazhong University of Sci. and Tech.
3. Zhenbang Li, School of Electronic Information and Communications, Huazhong University of Sci. and Tech.
4. Yufeng Liu, School of Electronic Information and Communications, Huazhong University of Sci. and Tech.
5. Yaobing Chen, Institute of Pathology, Tongji Hospital, Tongji Medical College, Huazhong University of Sci. and Tech.
6. Ke Ma, School of Electronic Information and Communications, Huazhong University of Sci. and Tech.
7. Wenjuan Zhang, School of Electronic Information and Communications, Huazhong University of Sci. and Tech.
8. Xinyu Zou, School of Electronic Information and Communications, Huazhong University of Sci. and Tech.