Overview

Schema Version

https://atlas.rsna.org/schemas/2025-11/dataset.json

Name

Intramedullary Spinal Cord Tumors and Inflammatory Demyelinating Lesions study dataset

Link

https://dx.doi.org/10.1148/ryai.210292

Indexing

Keywords: Spinal cord, Intramedullary, Astrocytoma, Ependymoma, Multiple sclerosis, NMOSD, Deep learning, Segmentation, Classification, T2-weighted MRI

Content: NR, MR

RadLex: RID4066, RID4033, RID19284, RID35613, RID7395

SNOMED: 1157043006, 1186904005, 443643007, 24700007

Author(s)

Zhizheng Zhuo

Jie Zhang

Yunyun Duan

Liying Qu

Chenlu Feng

Xufang Huang

Dan Cheng

Xiaolu Xu

Ting Sun

Zhaohui Li

Xiaopeng Guo

Xiaodong Gong

Yongzhi Wang

Wenqing Jia

Decai Tian

Xinghu Zhang

Fudong Shi

Sven Haller

Frederik Barkhof

Chuyang Ye

Yaou Liu

Organization(s)

Beijing Tiantan Hospital, Capital Medical University

BioMind

Neusoft Group, Department of Medical Imaging Product

China National Clinical Research Center for Neurologic Diseases

Tianjin Neurologic Institute, Tianjin Medical University General Hospital

University Hospitals of Geneva and University of Geneva, Department of Imaging and Medical Informatics

UCL Institutes of Neurology and Healthcare Engineering

Amsterdam University Medical Centers, Department of Radiology and Nuclear Medicine

Beijing Institute of Technology, School of Information and Electronics

License

Contact

Corresponding author: Yaou Liu (contact email provided in article text)

Funding

National Science Foundation of China (81870958, 81571631); Beijing Municipal Natural Science Foundation for Distinguished Young Scholars (JQ20035); Special Fund of the Pediatric Medical Coordinated Development Center of Beijing Hospitals Authority (XTYB201831); ECTRIMS-MAGNMIS Fellowship from ECTRIMS (Y.L.).

Ethical review

Study performed in accordance with the Declaration of Helsinki and approved by the animal and human ethics committee of the local institution; written informed consent was obtained from all patients.

Comments

Retrospective development cohort (n=490) and prospective test cohort (n=157) of patients with intramedullary spinal cord lesions used to develop and evaluate a DL pipeline for segmentation and classification using sagittal T2-weighted MRI.

Date

Published: 2022-09-07

References

[1] Zhuo Z, Zhang J, Duan Y, et al.. "Automated Classification of Intramedullary Spinal Cord Tumors and Inflammatory Demyelinating Lesions Using Deep Learning". Radiology: Artificial Intelligence. 2022-11-01. doi:10.1148/ryai.210292. PMID: 36523644. PMCID: PMC9745442.

Dataset

Motivation

To develop a deep learning pipeline for integrated segmentation and differential diagnosis of intramedullary spinal cord tumors and inflammatory demyelinating lesions using widely available T2-weighted MRI.

Sampling

Retrospective identification of consecutive patients prior to treatment (2012–2018) and prospective consecutive enrollment (2019–2020) according to inclusion/exclusion criteria.

Partitioning scheme

Retrospective cohort (2012–2018) split into training (80%) and validation (20%); independent prospective test cohort (2019–2020).

Missing information

Exact imaging file formats, acquisition parameters, and per-partition class distributions for the retrospective training and validation splits are not reported in the main text.

Confidentiality

Clinical MRI with patient health information; data sharing by request from the corresponding author.

Sensitive data

Clinical imaging data of patients with spinal cord diseases.