Duke Liver Dataset (MRI) v2
Creators
- 1. Duke University, School of Medicine
Description
The Duke Liver DataSet (DLDS) provides over 2000 anonymized MRI image series acquired in routine liver MRI protocols across 105 subjects that can be used to train algorithms for two applications:
(1) Series identification
(2) Liver segmentation
The series identification grouping (Series_Classification.zip) includes 2146 unique image series across all 105 subjects. The image series encompass seventeen different contrast types including different pulse sequences, contrast bolus phases, and imaging orientations. A randomized, alphabetical series label corresponding to one of these seventeen contrast types is assigned to each contrast type. The file SeriesClassificationKey.csv includes a list of all image series included in the dataset and their corresponding label. The file SequenceTypes.csv shows the mapping of these labels to contrast type. It is intended that these labels can be used to train an automated series classification model for abdominal MRI.
The liver segmentation sub-grouping (Segmentation.zip) includes 310 unique image series across a subset of 95 subjects. The image series encompass four main contrast types: axial in-phase, axial opposed, axial precontrast fat-suppressed T1w, and contrast-enhanced portal venous t1w. It is intended that this data can be used to train an automated liver segmentation model. The file SegmentationKey.csv includes a list of the cases with included masks files. Since there is overlap in these series with the series identification dataset, corresponding series labels are also included.
In both datasets, cases are organized in directories by DLDS ID -> Series number. The segmentation cases are further divided into image and mask sub-directories.
Please note that a previous version of this dataset (version 1.0.0) withheld roughly 25% of series labels and segmentations. These have been restored to the dataset for completeness. As reference for users wishing to know which data was added (or looking for a list of subjects to remove for a convenient test set), labels and masks were previously withheld for the following DLDS id's: [0002, 0003, 0004, 0017, 0020, 0023, 0031, 0032, 0033, 0035, 0038, 0042, 0046, 0050, 0063, 0074, 0075, 0078, 0083, 0084, 0096, 0097, 0098, 0100, 0103, 0104].
Files
Restricted
The record is publicly accessible, but files are restricted to users with access.
Request access
If you would like to request access to these files, please fill out the form below.
You need to satisfy these conditions in order for this request to be accepted:
This dataset, and any copyrights therein, are owned by Duke University. In order to receive the dataset, you must choose one of the following two licenses:
1. An open license under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0) license: https://creativecommons.org/licenses/by-nc-nd/4.0/
To enter a license agreement with the CC BY-NC-ND 4.0 restrictions, please submit a request here and provide your academic affiliation (if any) and a brief description of why you would like to use this data.
2. A custom license with Duke University, for use without the CC-BY-NC-ND-4.0 restrictions, which can include commercial uses.
To enter a license agreement without the CC-BY-NC-ND-4.0 restrictions, please contact the Digital Innovations department at Duke Office for Translation & Commercialization (OTC) (https://otc.duke.edu/software/) at otcquestions@duke.edu with reference to "OTC File DOI 10.5281/zenodo.6328446." in your email.
Please include in your email your affiliation (if applicable) and a brief description of your research topics and why you would like to use this dataset. Duke University will make use of this information to evaluate approval of your request.
Outside contributions to the Duke-owned dataset cannot be accepted unless the contributor assigns copyrights to any modifications, changes, and/or derivatives over to Duke University.
Please note that this dataset is distributed AS IS, WITHOUT ANY WARRANTY; and without the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
You are currently not logged in. Do you have an account? Log in here