Spine MRI Dataset – Fracture Detection & Segmentation

This dataset consists of .dcm files containing MRI scans of the spine from individuals with various dystrophic changes, including: osteophytes, dorsal disc extrusions, dorsal disc protrusions, and spondyloarthrosis. The images are labeled by doctors and include accompanying PDF-format medical reports.

The dataset contains 5 studies captured from multiple angles, offering comprehensive insight into spinal abnormalities. Each scan includes detailed imaging of vertebrae, intervertebral discs, nerves, and surrounding soft tissues, making it suitable for training spine anomaly classification and segmentation models.

MRI Study Angles in the Dataset

Multiple MRI views are included to support diverse learning tasks in spinal imaging.

Conditions Included in the Full Dataset:

• Degeneration of discs
• Osteophytes
• Osteochondrosis
• Hemangioma
• Disk extrusion
• Spondylitis
• ...and many other conditions

Use Cases

Researchers and healthcare professionals can use this dataset to study spinal disorders such as herniated discs, spinal stenosis, scoliosis, and fractures. It can also be used to develop and evaluate imaging techniques, AI models for automated diagnosis, and computer vision algorithms for image segmentation.

Content

• ST000001: Includes subfolders with 5 studies, each containing MRI scans in .dcm and .jpg formats.
• DICOMDIR: Metadata file including links and patient condition information.
• Spine_MRI_3.pdf: Medical report provided by a radiologist.
• .csv file: Lists study IDs and the number of images per study.

Medical Reports Contain:

• Patient demographic information
• Description of the case
• Preliminary diagnosis
• Recommendations for further action

All patients have provided informed consent for publication, and all data has been anonymized.

Data Services

Medical data can be collected according to your specific requirements. TrainingData offers high-quality annotation services tailored to AI/ML needs in healthcare.

Keywords

visual, label, positive, negative, symptoms, clinically, sensory, syndrome, predictors, rsna cervical, segmentations meta, mri spine scans, spinal imaging, radiology dataset, neuroimaging, image segmentation, lumbar spine mri, thoracic spine mri, cervical spine mri, spinal cord mri, orthopedic imaging, spine abnormalities, spine mri classification, artificial intelligence in radiology, spine pathology analysis, mri feature extraction, deep learning medical imaging, tomography