ACDC: Automated Cardiac Diagnosis Challenge MRI Scans

Automated Cardiac Diagnosis Challenge (ACDC) materials were introduced during MICCAI 2017 by Bernard et al. The dataset contains cine cardiac MRI acquisitions from healthy subjects and patients with multiple cardiac pathologies, designed to advance research in cardiac MRI analysis and representation learning.

Use Cases

• Training machine learning models for automated cardiac disease classification based on cine MRI scans.
• Developing representation learning techniques for cardiac image analysis using data from healthy and pathological subjects.
• Benchmarking automated segmentation and diagnosis algorithms against a challenge-standardized dataset.
• Advancing research in understanding cardiac pathologies through structured MRI analysis.

Strengths

• Dataset is associated with a recognized challenge (Automated Cardiac Diagnosis Challenge) introduced at MICCAI 2017.
• Data includes both healthy subjects and patients with multiple cardiac pathologies, providing a basis for comparative analysis.

Limitations

• Description metadata is limited; actual data quality requires manual inspection after download.
• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.

Provenance

Source

chehablab on Hugging Face, originating from the Automated Cardiac Diagnosis Challenge (ACDC).

Collection Method

Likely collected as part of a medical research challenge for standardized benchmarking.

Freshness

Last updated 2026-05-22 15:15:58; freshness should be verified.