Manifold learning for image-based breathing gating in MRI

Study of organ motion due to respiration is an important issue for image-guided procedures in the abdominal region. We propose a novel image-based breathing gating method to recover the respiratory signal from the images. The gating is accomplished using Laplacian eigenmaps, a manifold learning technique, by determining the low-dimensional manifold embedded in the high-dimensional image space. We perform the manifold learning on MR slices acquired from a fixed location. Experiments are conducted on liver data by projecting the data on different dimensions. The results are evaluated in terms of suitability for usage as a similarity criterion for 4D reconstruction methods.     «

Study of organ motion due to respiration is an important issue for image-guided procedures in the abdominal region. We propose a novel image-based breathing gating method to recover the respiratory signal from the images. The gating is accomplished using Laplacian eigenmaps, a manifold learning technique, by determining the low-dimensional manifold embedded in the high-dimensional image space. We perform the manifold learning on MR slices acquired from a fixed location. Experiments are conducted...     »