PT Unknown
AU Francesco Ciompi
   Oriol Pujol
   Simone Balocco
   Xavier Carrillo
   J. Mauri
   Petia Radeva
TI Automatic Key Frames Detection in Intravascular Ultrasound Sequences
BT In MICCAI 2011 Workshop on Computing and Visualization for Intra Vascular Imaging
PY 2011
AB We present a method for the automatic detection of key frames in Intravascular Ultrasound (IVUS) sequences. The key frames are markers delimiting morphological changes along the vessel. The aim of defining key frames is two-fold: (1) they allow to summarize the content of the pullback into few representative frames; (2) they represent the basis for the automatic detection of clinical events in IVUS. The proposed approach achieved a compression ratio of 0.016 with respect to the original sequence and an average inter-frame distance of 61.76 frame, minimizing the number of missed clinical events.
ER