Re-curated Breast Imaging Subset DDSM Dataset (RBIS-DDSM)

Citation Author(s):
Arunava
Chakravarty
Department of Electrical Engineering, Indian Institute of Technology Kharagpur, Kharagpur, India
Tandra
Sarkar
Apollo Gleneagles Hospital, Kolkata, India
Rakshith
Sathish
Advanced Technology Development Center, Indian Institute of Technology Kharagpur, Kharagpur, India
Ramanathan
Sethuraman
Intel Technology India Pvt. Ltd., Bangalore, India.
Debdoot
Sheet
Department of Electrical Engineering, Indian Institute of Technology Kharagpur, Kharagpur, India
Submitted by:
RAKSHITH SATHISH
Last updated:
Thu, 03/03/2022 - 10:44
DOI:
10.21227/nqp1-sp19
License:
Creative Commons Attribution
758 Views
Categories:
Artificial Intelligence
Machine Learning
Image Processing
Medical Imaging
Computer Vision
Keywords:
Mammograms, Screening, annotations
0
0 ratings - Please login to submit your rating.
ACCESS DATASET CITE SHARE/EMBED

Abstract 

Re-curated Breast Imaging Subset DDSM Dataset (RBIS-DDSM) is a curated version of 849 images from the CBIS-DDSM dataset available online with a permissive copyright license (CC-BY-SA 3.0). The  CBIS-DDSM dataset is an improved version of the DDSM dataset. The authors of the CBIS-DDSM dataset attempted to improve the ground truth by applying simple image processing based methods to enhance the edges without any manual intervention from medical experts in order to segment and annotate masses. However, these annotations (segmentation maps) are inaccurate in most of the images. 

RBIS-DDSM provides more accurate annotations marked manually under the supervision of an experienced senior Radiologist. These annotations can be used for localisation tasks and to evaluate the inter-rater agreement when used with annotations provided in the CBIS-DDSM dataset. The ROI was manually marked for the images using an in-house Matlab GUI tool. 

 

Note: The local annotations for mass and micro-calcifications were marked as ellipses around the pathologies and are not an exact semantic segmentation. This dataset should not be used to evaluate models/algorithms designed for semantic segmentation tasks.

 

License: The dataset is available on a CC BY 4.0 license.

Instructions: 

The dataset is organized as follows.

 

RBIS-DDSM

|

- images

- roi_mask

- cbis_annotation

- annotations

|   |

|   - calc_ben_mask

|   - calc_mlgn_mask

|   - mass_mask

|   |    |

|   |    - mass_ben_mask

|   |    - mass_mlgn_mask

- ReadMe.pdf

x 

The ‘RBIS-DDSM’ directory contains four subdirectories: ' images’, ‘roi_mask’, ‘cbis_annotation’, and ‘annotation’. The ‘images’ directory contains 849 mammogram images selected from the CBIS-DDSM dataset, ‘roi_mask’ contains the binary breast tissue mask for the images, and ‘cbis_annotation’ contains annotations provided in the CBIS-DDSM dataset. The ‘annotation’ directory has four subdirectories: 'calc_ben_mask’, ‘calc_mlgn_mask’, ‘mass_ben_mask’ and ‘mass_mlgn_mask’. The ‘calc_ben_mask’ directory contains annotations for benign micro-calcifications, ‘calc_mlgn_mask’ contains annotations for malignant micro-calcifications, ‘mass_ben_mask’ contains annotations for benign masses, and ‘mass_mlgn_mask’ contains annotations for malignant masses.

Comments

I require access to the dataset. Kindly permit.

Submitted by Ramanathan Seth... on Thu, 03/03/2022 - 09:02

Dataset Files

LOGIN TO ACCESS DATASET FILES

Documentation

AttachmentSize
File Readme.txt2.71 KB

QUESTIONS?