Histology-based Digital Staining of Pathology Images (HD-Staining) is a newly developed deep-learning algorithm. We trained a Mask-RCNN model to segment tumor nuclei, stroma nuclei, lymphocyte nuclei, macrophage nuclei, karyorrhexis, and red blood cells in pathological Hematoxylin & Eosin (HE) stained images. Around 10,000 cells are covered in our training dataset from the National Lung Screening Trial (NLST) cohort. This tool aims to dissect tumor microenvironment in cell level.
Prepare your pathological image file.
Analyze user's image using HD-Staining Online Tool.
Visualize the results and allow the user to download.
The design of this segmentation online tool aims to provide a simply and user-friendly usage environment. The user only needs to prepare a pathological image to submit. We suggest the user input the email to receive job completion alert because the job execution might take several minutes. Once the job is completed, the original image and the segmented results will be displayed together on this website.
| # | Original | Processed | |
|---|---|---|---|
| Original Image | Mask Image | Overlapped Image | |
| 1 |  |
 |
 |
| 2 |  |
 |
 |
| 3 |  |
 |
 |
| 4 |  |
 |
 |
| 5 |  |
 |
 |
| 6 |  |
 |
 |
| 7 |  |
 |
 |
| 8 |  |
 |
 |
Please cite our work if you find this HDStaining tool helpful:
Wang, Shidan, Ruichen Rong, Donghan M. Yang, Junya Fujimoto, Shirley Yan, Ling Cai, Lin Yang et al.
"Computational staining of pathology images to study the tumor microenvironment in lung cancer." Cancer
Research 80, no. 10 (2020): 2056-2066. [Find the article]
