Datasets
Standard Dataset
Liver Vessel Segmentation
- Citation Author(s):
-
LiangZhao
Taihe Hospital - Submitted by:
- Liang Zhao
- Last updated:
- Mon, 10/16/2023 - 04:48
- DOI:
- 10.21227/rwys-mk84
- Data Format:
- Links:
- License:
3325 Views- Categories:
- Keywords:
Abstract
Liver vessels generated from computed tomography are usually pretty small, which poses big challenges to satisfactory vessel segmentation, including 1) scarcity of high-quality and large-volume vessel masks, 2) difficult to capture vessel-specific features and 3) heavily imbalanced distribution of vessels and liver tissues. To help the improvement of vessel segmentation, we build a high-quality and large-volume dataset having both liver and vessel masks available. Comprehensive experiments conducted on the dataset as well as existing well known datasets show that the averaged dice coefficient score is lifted by at least 12.8%. We believe that this dataset can be of great help for this area.
There are 532 files named "casexxxx.nii.gz", where "xxxx" ranges from "0001" to "0532", representing the 532 CT volumes. We randomly annotated 30 slices for each CT volume, and those liver slices with all "0" masks are not annotated. The dataset contains two folders: the original abdominal image containing the raw HU values clipped to [0, 400] and named "image_nii", and the vessel mask named "vessel_mask_nii".
For more details, please refer to our paper published with IEEE TMI entitled "Laplacian salience-gated feature pyramid network for accurate liver vessel segmentation" (DOI: 10.1109/TMI.2023.3273528).
Comments
Dear Liang Zhao
I am Rajan, a post-doc. and working on liver, and tumor segmentation for clinical 3D visualization. It will be a great help for my work if you provide the dataset.
With best regards,
N. V. Rajan
Hi Rajan,
The data will be released in case the manuscript is accepted, and I will inform you asap.
Best,
Liang
thanks
Sir, great respect for your research, it is helpful to my area. I am looking forward to the dataset so much. Thank you.
Dear Liang Zhao
I am Caixia, a. phd student and working on vessel segmentation for clinical 3D visualization. I am looking forward to the dataset so much. Thank you.
With best regards,
Caixia Dong
tanks
Dear Mr.zhao:
I am Yvchen Wang,a master student working on vessel segmentation . Your paper is amaing. I would be very grately if you would be willing to share the dataset. Thank you.
With best regards,
Yvchen Wang
Dear Mr.zhao:
I am Jun Li, a doctoral student working on liver vessel segmentation . Your work is so great. I would appreciate it if you could share the dataset with me. Thank you so much!
I look forward to your reply!
Yours sincerely,
Jun Li