Database access

The overall ACDC dataset was created from real clinical exams acquired at the University Hospital of Dijon. Acquired data were fully anonymized and handled within the regulations set by the local ethical committee of the Hospital of Dijon (France). Our dataset covers several well-defined pathologies with enough cases to (1) properly train machine learning methods and (2) clearly assess the variations of the main physiological parameters obtained from cine-MRI (in particular diastolic volume and ejection fraction). The dataset is composed of 150 exams (all from different patients) divided into 5 evenly distributed subgroups (4 pathological plus 1 healthy subject groups) as described below. Furthermore, each patient comes with the following additional information : weight, height, as well as the diastolic and systolic phase instants.

Although the challenge is now closed, the data and the groundtruth are still publicly available via the following link.

Study population

The targeted population for the study is composed of 150 patients divided into 5 subgroups as follows:

  • 30 normal subjects - NOR

  • 30 patients with previous myocardial infarction (ejection fraction of the left ventricle lower than 40% and several myocardial segments with abnormal contraction) - MINF

  • 30 patients with dilated cardiomyopathy (diastolic left ventricular volume >100 mL/m2 and an ejection fraction of the left ventricle lower than 40%) - DCM

  • 30 patients with hypertrophic cardiomyopathy (left ventricular cardiac mass high than 110 g/m2, several myocardial segments with a thickness higher than 15 mm in diastole and a normal ejecetion fraction) - HCM

  • 30 patients with abnormal right ventricle (volume of the right ventricular cavity higher than 110 mL/m2 or ejection fraction of the rigth ventricle lower than 40%) - RV

Each group was clearly defined according to physiological parameter, such as the left or right diastolic volume or ejection fraction, the local contraction of the LV, the LV mass and the maximum thickness of the myocardium. More details can be found on the Classification rules tab.

Involved systems

The acquisitions were obtained over a 6 year period using two MRI scanners of different magnetic strengths (1.5 T (Siemens Area, Siemens Medical Solutions, Germany) and 3.0 T (Siemens Trio Tim, Siemens Medical Solutions, Germany)). Cine MR images were acquired in breath hold with a retrospective or prospective gating and with a SSFP sequence in short axis orientation. Particularly, a series of short axis slices cover the LV from the base to the apex, with a thickness of 5 mm (or sometimes 8 mm) and sometimes an interslice gap of 5 mm (then one image every 5 or 10 mm, according to the examination). The spatial resolution goes from 1.37 to 1.68 mm2/pixel and 28 to 40 images cover completely or partially the cardiac cycle (in the second case, with prospective gating, only 5 to 10 % of the end of the cardiac cycle was omitted), all depending on the patient.

You must refer to this citation for any use of the ACDC database.

  • O. Bernard, A. Lalande, C. Zotti, F. Cervenansky, et al.
    "Deep Learning Techniques for Automatic MRI Cardiac Multi-structures Segmentation and
    Diagnosis: Is the Problem Solved ?" in IEEE Transactions on Medical Imaging,
    vol. 37, no. 11, pp. 2514-2525, Nov. 2018

    doi: 10.1109/TMI.2018.2837502