HolOrama is an open-source application for cardiac image analysis. Its name combines the Greek words holo, meaning whole, and orama, meaning vision, reflecting the aim of viewing the entire heart across different imaging methods within a single interface. It currently supports intravascular ultrasound, a catheter-based technique that images a coronary artery from the inside; optical coherence tomography, a light-based intravascular technique; and coronary computed tomography angiography, a non-invasive scan of the coronary arteries. Images are read in the DICOM and NIfTI file formats that are standard in medical imaging.

For the two intravascular methods, the application allows frame-by-frame inspection and the manual drawing of one or several contours, covering the vessel lumen, the external elastic membrane, calcium, lipid, macrophages and side branches, with the corresponding measurements calculated automatically. Lumen contours in intravascular ultrasound can also be segmented automatically across all frames. Frames belonging to the same phase of the heart cycle are identified by an automatic gating step, and breathing motion is detected from the lumen area signal, reviewed or corrected by hand, and then used to reorder the gated frames into a breathing-corrected pullback. Results can be exported as detailed reports, as coordinate tables, or as image and segmentation files suitable for training machine learning models.

For coronary computed tomography angiography, volumes are displayed as synchronized axial, coronal and sagittal slices alongside a three-dimensional rendering of the segmentation. A brush tool adds or erases labels on any of the two-dimensional views, and a lasso tool in the three-dimensional view removes the voxels of a chosen label that fall inside a drawn region, which makes it straightforward to clean up outliers. Cut lines drawn on the slices define the left ventricular outflow tract and the aortic root, after which the coronary arteries, aorta and left ventricle can be exported as a combined segmentation file or as a surface mesh. Data from computed tomography and from the intravascular methods can be fused into a single geometry.