No abstract

Optical coherence tomography (OCT) has proven to be an essential imaging modality for ophthalmology and is proving to be very important in neurology. OCT enables high resolution imaging of the retina, both at the optic nerve head and the macula. Macular retinal layer thicknesses provide useful diagnostic information and have been shown to correlate well with measures of disease severity in several diseases. Since manual segmentation of these layers is time consuming and prone to bias, automatic segmentation methods are critical for full utilization of this technology. In this work, we build a random forest classifier to segment eight retinal layers in macular cube images acquired by OCT. The random forest classifier learns the boundary pixels between layers, producing an accurate probability map for each boundary, which is then processed to finalize the boundaries. Using this algorithm, we can accurately segment the entire retina contained in the macular cube to an accuracy of at least 4.3 microns for any of the nine boundaries. Experiments were carried out on both healthy and multiple sclerosis subjects, with no difference in the accuracy of our algorithm found between the groups.

This article presents a new method for measuring longitudinal strain in a short-axis section of the heart using harmonic phase magnetic resonance imaging (HARP-MRI). The heart is tagged using 1-1 SPAMM at end-diastole with tag surfaces parallel to a short-axis imaging plane. Two or more images are acquired such that the images have different phase encodings in a direction orthogonal to the image plane. A dense map of the longitudinal strain can be computed from these images using a simple, fast computation.

This article introduces a new image processing technique for rapid analysis of tagged cardiac magnetic resonance image sequences. The method uses isolated spectral peaks in SPAMM-tagged magnetic resonance images, which contain information about cardiac motion. The inverse Fourier transform of a spectral peak is a complex image whose calculated angle is called a harmonic phase (HARP) image. It is shown how two HARP image sequences can be used to automatically and accurately track material points through time. A rapid, semiautomated procedure to calculate circumferential and radial Lagrangian strain from tracked points is described. This new computational approach permits rapid analysis and visualization of myocardial strain within 5-10 min after the scan is complete. Its performance is demonstrated on MR image sequences reflecting both normal and abnormal cardiac motion. Results from the new method are shown to compare very well with a previously validated tracking algorithm. Keywords cardiac motion; harmonic phase; magnetic resonance tagging; myocardial strain Major developments over the past decade in tagged cardiac magnetic resonance imaging (1-6) have made it possible to measure the detailed strain patterns of the myocardium in vivo (7-11). MR tagging uses a special pulse sequence to spatially modulate the longitudinal magnetization of the subject to create temporary features, called tags, in the myocardium. Fast spoiled gradient echo imaging techniques are used to create CINE sequences that show the motion of both the anatomy of the heart and the tag features that move with the heart. Analysis of the motion of the tag features in many images taken from different orientations and at different times can be used to track material points in 3D, leading to detailed maps of the strain patterns within the myocardium (11,12). Tagged MRI has figured prominently in many recent medical research and scientific investigations. It has been used to develop and refine models of normal and abnormal myocardial motion (7,8,12-14) to better understand the correlation of coronary artery disease with myocardial motion abnormalities (15), to analyze cardiac activation patterns using pacemakers (16), to understand the effects of treatment after myocardial infarction (17), and in combination with stress testing for the early detection of myocardial ischemia (18). Despite

The proposed methodology utilizes the full resolution of the data, rather than relying on global descriptions such as area measurements. The application of this methodology to an elderly group indicated sex-related differences in the callosal shape and size.