Bag of visual words is common technique used in by the computer vision community in visual recognition problems. In this paper, the authors explored it in a mid-level image representation for real-time heart view plane classification of 2D echocardiogram ultrasound images.
Throughout an extensive set of experiments, the proposed approach proved to be effective and efficient for the target problem, making it suitable for use in real-time setups, which are based on real-time constraints, noise filtering, and different machine learning classifiers. Proposed approach is very fast to compute and consistently achieves accuracy above 90%.
The paper also includes an in-depth review of the literature in the area of automatic echocardiogram view classification giving the reader a through comprehension of this field of study.
Penatti, Otávio; Werneck, Rafael; de Almeida, Waldir; Stein, Bernardo; Pazinato, Daniel; Mendes Júnior, Pedro; Torres, Ricardo; Rocha, Anderson. “Mid-level image representations for real-time heart view plane classification of echocardiograms”. Computers in Biology and Medicine. Elsevier, Vol. 66 , 66 – 81. 2015. doi: 10.1016/j.compbiomed.2015.08.004