Skip to main content
Fig. 5 | Journal of Cardiovascular Magnetic Resonance

Fig. 5

From: MRXCAT2.0: Synthesis of realistic numerical phantoms by combining left-ventricular shape learning, biophysical simulations and tissue texture generation

Fig. 5

Visualization of the latent space variables of the shape model for the Automated Cardiac Diagnosis Challenge (ACDC) dataset. Anatomies are color coded according to the clinical label: healthy (NOR, green squares), hypertrophic cardiomyopathy (HCM, grey triangles) and dilated cardiomyopathy (DCM, orange circles). Black symbols show the clusters' positions calculated using the k-means algorithm on the latent-space variable vectors. Note that due to the non-linear nature of the embedding, the cluster means do not necessarily match the barycenter of the point clouds

Back to article page