Dataset SpineXR

logo-VinBigData-2020-02

VinDr-SpineXR: An open dataset for spinal lesions detection and classification from radiographs

Dataset description

Radiographs are used as the most important imaging tool for identifying spine anomalies in clinical practice. To the best of our knowledge, no existing studies have been devoted to the development and evaluation of a comprehensive system for classifying and localizing multiple spine lesions from X-ray scans. The lack of large datasets with high-quality images and human experts’ annotations is the key obstacle. To fill this gap,  Vingroup Big Data Institute (VinBigdata) has created and made freely available the VinDr-SpineXR: A large-scale X-ray dataset for spinal lesions detection and classification. The VinDr-SpineXR contains 10,469 images from 5,000 studies that are manually annotated with 13 types of abnormalities, each scan was annotated by an expert radiologist.

To the best of our knowledge, the VinDr-SpineXR is currently the largest dataset to date that provides radiologist’s bounding-box annotations for developing supervised-learning object detection algorithms. We believe that the dataset will serve as a benchmark dataset for accelerating the development and evaluation of new machine learning models for the spinal X-ray interpretation.

Table 1. Overview of publicly available MSK image datasets.

 

Dataset 

Year

Study
type

label

Number
of images

Digital Hand Atlas [1]

2007

Left hand 

Bone age 

1,390

Osteoarthritis Initiative [2]

2013

Knee

K&L Grade

8,892

MURA [3]

2017

Upper body

Abnormalities

40,561

RSNA Pediatric Bone Age [4]

2019

Hand

Bone age

14,236

Kuok et al. [5]

2018 

Spine 

Lumbar vertebrae mask 

60 

Kim et al. [6]

2020 

Spine 

Spine position 

797

Ours 

2021 

Spine

Multiple abnormalities 

10,469

Figure 1. Examples of spine X-ray scans with radiologist’s annotations. Abnormal findings (local labels) marked by radiologists are plotted on the original images for visualization purposes.

Dataset Statistics

Table 2. Characteristics of patients in the training and test datasets.

Download Dataset

The full version of the VinDr-SpineXR can be downloaded from PhysioNet (https://www.physionet.org/content/vindr-spinexr/1.0.0/).

Visualization

The image and annotation quality of the dataset can be via VinDr Laboratory – our hub for all public datasets. To access the data hub, users are required to complete a request access form.

Author List and Affiliations

Hieu T. Nguyen [1,2], Hieu H. Pham [1,3], Nghia T. Nguyen [1], Ha Q. Nguyen [1,3], Thang Q. Huynh [2], Minh Dao [1], and Van Vu [1,4]

[1] Medical Imaging Center, Vingroup Big Data Institute, Hanoi, Vietnam

[2] School of Information and Communication Technology, Hanoi University of Science and Technology, Hanoi, Vietnam

[3] College of Engineering & Computer Science, VinUniversity, Hanoi, Vietnam

[4]  Department of Mathematics, Yale University, New Heaven, USA

*  Corresponding author: Hieu H. Pham (hieuph4@vinbigdata.org)

 

Citation

For any publication that explores this resource, the authors must cite this original paper as follows:

 

@article{nguyen2021vindr,
   title={VinDr-SpineXR: A deep learning framework for spinal lesions detection and classification from radiographs},
author={Nguyen, Hieu T and Pham, Hieu H and Nguyen, Nghia T and Nguyen, Ha Q and Huynh, Thang Q and Dao, Minh and Vu, Van}, journal={International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI 2021)}, year={2021}
}

Contact

We welcome any comments, suggestions or feedback you have for us that help improve the dataset, correspondence should be addressed to: Hieu H. Pham (hieuph4@vinbigdata.org).

References

[1] Gertych, A., Zhang, A., Sayre, J., Pospiech-Kurkowska, S., Huang, H.: Bone age assessment of children using a digital hand atlas. Computerized Medical Imaging and Graphics 31(4-5), 322–331 (2007).

[2] Osteoarthritis initiative: A multi-center observational study of men and women. https://oai.epi-ucsf.org/datarelease/, accessed: 2021-02-22.

[3] Rajpurkar, P., Irvin, J., Bagul, A., Ding, D., Duan, T., Mehta, H., Yang, B., Zhu, K., Laird, D., Ball, R.L., et al.: MURA: Large dataset for abnormality detection in musculoskeletal radiographs. arXiv preprint arXiv:1712.06957 (2017).

[4] Halabi, S.S., Prevedello, L.M., Kalpathy-Cramer, J., Mamonov, A.B., Bilbily, A., Cicero, M., Pan, I., Pereira, L.A., Sousa, R.T., Abdala, N., et al.: The RSNA pediatric bone age machine learning challenge. Radiology 290(2), 498–503 (2019).

[5] Kuok, C.P., Fu, M.J., Lin, C.J., Horng, M.H., Sun, Y.N.: Vertebrae segmentation from X-ray images using convolutional neural network. In: International Conference on Information Hiding and Image Processing (IHIP). pp. 57–61 (2018).

en_USEN