Skip to main content

Datasets

Standard Dataset

3DLSC-COVID

Citation Author(s):
Xiaofei Wang
Submitted by:
Xiaofei Wang
Last updated:
DOI:
10.21227/mxb3-7j48
Data Format:
Research Article Link:
2501 views
Categories:
Keywords:
Average: 5 (1 vote)

Abstract

The 3DLSC-COVID datset  includes a total of  1,805 3D chest CT scans with more than 570,000 CT slices were collected from 2 standard CT scanners of Liyuan Hospital, i.e.,  UIH uCT 510 and GE Optima CT600.  Among all CT scans, there were 794 positive cases of COVID-19, which were further confirmed by clinical symptoms and RT-PCR from January 16 to April 16, 2020. Additionally, 540 positive cases of CAP,  471 non-pneumonia cases were randomly selected from the same hospital between November 5, 2016 and April 28, 2020.  Part of the CT scans (including 100 COVID-19 and 32 CAP cases) along with the corresponding lesion annotations have already passed the open source approval process and are now available. Other CT images and the correspoding lesion annotations wil be open sourced soon. If you have any questions, please contact xfwang@buaa.edu.cn. Note that the data can only used for academic purpose.

 

Instructions:

You can refer to the TMI 2021 paper entitled ''Joint learning of 3D leasion segmentation and classification for explainable COVID-19 diagnosis''.