Datasets
Models
About Dataset
CardioScans Dataset Sample Description
The CardioScans dataset is a detailed collection of high-quality CT and MRI scans of the heart, acquired from anonymized patients. This sample includes a single DICOM (.dcm) file to give an overview of the full dataset, which consists of over 20,000 DICOM files covering 150 patients and totaling 1 TB of data. If you are interested in accessing the complete dataset, please contact us via the provided email address.
Dataset Summary
File Type: DICOM (.dcm)
Content: Heart imaging scan (CT or MRI)
Purpose: Demonstration of the larger dataset’s scope and quality
Full Dataset Details
Total Size: 1 TB
Number of Files: Over 20,000 DICOM files
Number of Patients: 150
Imaging Types: Comprehensive CT and MRI heart scans
Metadata Included:
Gender
Age
Anonymized birth dates
Data Quality and Utility
High-Resolution Scans: Optimized for cardiac imaging research.
Diverse Imaging Protocols: Multiple series per patient, providing varied perspectives.
Rich Dataset for:
Deep learning applications (e.g., segmentation, anomaly classification)
3D reconstruction and volumetric analysis of cardiac structures
Exploratory cardiology and radiology research
Applications of the Full Dataset
Cardiac Imaging Research: Understanding heart anatomy and pathology.
Deep Learning Training: Developing models for segmentation and disease detection.
3D Modeling: Creating volumetric representations of the heart.
Algorithm Development: Image registration, noise reduction, and feature extraction.
Medical Education: Training and educational use in cardiac imaging.
Anonymization and Compliance
To ensure privacy and compliance with regulations (e.g., GDPR, HIPAA), all personal identifiers have been removed. Patients are labeled with anonymized identifiers, and only essential metadata (e.g., gender, age) is provided.
How to Access the Full Dataset
If you are interested in the complete CardioScans dataset or have collaboration inquiries, please contact us via email. contact@human-ai-ze.com