Ahmet M. Bagci

Thickness profiles of retinal layers by optical coherence tomography image segmentation.

PURPOSE To report an image segmentation algorithm that was developed to provide quantitative thickness measurement of six retinal layers in optical coherence tomography (OCT) images. DESIGN Prospective cross-sectional study. METHODS Imaging was performed with time- and spectral-domain OCT instruments in 15 and 10 normal healthy subjects, respectively. A dedicated software algorithm was developed for boundary detection based on a 2-dimensional edge detection scheme, enhancing edges along the retinal depth while suppressing speckle noise. Automated boundary detection and quantitative thickness measurements derived by the algorithm were compared with measurements obtained from boundaries manually marked by three observers. Thickness profiles for six retinal layers were generated in normal subjects. RESULTS The algorithm identified seven boundaries and measured thickness of six retinal layers: nerve fiber layer, inner plexiform layer and ganglion cell layer, inner nuclear layer, outer plexiform layer, outer nuclear layer and photoreceptor inner segments (ONL+PIS), and photoreceptor outer segments (POS). The root mean squared error between the manual and automatic boundary detection ranged between 4 and 9 mum. The mean absolute values of differences between automated and manual thickness measurements were between 3 and 4 mum, and comparable to interobserver differences. Inner retinal thickness profiles demonstrated minimum thickness at the fovea, corresponding to normal anatomy. The OPL and ONL+PIS thickness profiles respectively displayed a minimum and maximum thickness at the fovea. The POS thickness profile was relatively constant along the scan through the fovea. CONCLUSIONS The application of this image segmentation technique is promising for investigating thickness changes of retinal layers attributable to disease progression and therapeutic intervention.

A method for detection of retinal layers by optical coherence tomography image segmentation

A novel algorithm is reported for automatically segmenting optical coherence tomography (OCT) images to identify six layers within the retina. The boundary detection technique presented here is based on a 2-D edge detection scheme, in which a filter with wedge-shaped pass band is introduced. The proposed filter enhances edges along the vertical boundaries while suppressing speckle noise. The detected contours are labeled based on a retina model. The layer thickness measurements derived by the algorithm are compared with thickness measurements from manually marked boundaries. The automated thickness measurements derived by the algorithm differ from manual segmentation results by less than 5 microns on average.

Low-complexity implementation of non-subsampled directional filter banks using polyphase representations and generalized separable processing

In this paper we present efficient filtering methods for the directional decomposition of images and extraction of directional cues. The design and implementation framework proposed in our paper is based on a non-subsampled tree structure, which allows for redundancy capturing some image characteristics for detection. We examine structures for analysis using 2M filters, M > 3, with narrow wedge-shaped passbands. Our structures are derived from the decimation of two-dimensional impulse responses composed of configurations of one-dimensional building blocks. In the implementation, we reconfigure transfer functions of the filters with decimated impulse responses by exploiting appropriate polyphase structure of one-dimensional building blocks to retain the property of generalized separability along directions determined by the passband orientations of the directional filters. The reconfiguration depends on the value of M, and in this paper we provide general rules for choosing the integer factors for polyphase decomposition that help re-cast the expressions for the transfer functions of the analysis bank of filters into forms that allow efficient implementation. The proposed implementation is shown to provide significant improvement compared with conventional two-dimensional implementation. Simulations show that the filtering structure provides good frequency selectivity on airborne surveillance images.

Magnetic resonance imaging–based measures predictive of short-term surgical outcome in patients with Chiari malformation Type I: a pilot study

OBJECTIVE This study identifies quantitative imaging-based measures in patients with Chiari malformation Type I (CM-I) that are associated with positive outcomes after suboccipital decompression with duraplasty. METHODS Fifteen patients in whom CM-I was newly diagnosed underwent MRI preoperatively and 3 months postoperatively. More than 20 previously described morphological and physiological parameters were derived to assess quantitatively the impact of surgery. Postsurgical clinical outcomes were assessed in 2 ways, based on resolution of the patients chief complaint and using a modified Chicago Chiari Outcome Scale (CCOS). Statistical analyses were performed to identify measures that were different between the unfavorable- and favorable-outcome cohorts. Multivariate analysis was used to identify the strongest predictors of outcome. RESULTS The strongest physiological parameter predictive of outcome was the preoperative maximal cord displacement in the upper cervical region during the cardiac cycle, which was significantly larger in the favorable-outcome subcohorts for both outcome types (p < 0.05). Several hydrodynamic measures revealed significantly larger preoperative-to-postoperative changes in the favorable-outcome subcohort. Predictor sets for the chief-complaint classification included the cord displacement, percent venous drainage through the jugular veins, and normalized cerebral blood flow with 93.3% accuracy. Maximal cord displacement combined with intracranial volume change predicted outcome based on the modified CCOS classification with similar accuracy. CONCLUSIONS Tested physiological measures were stronger predictors of outcome than the morphological measures in patients with CM-I. Maximal cord displacement and intracranial volume change during the cardiac cycle together with a measure that reflects the cerebral venous drainage pathway emerged as likely predictors of decompression outcome in patients with CM-I.

Computer-aided method for automated selection of optimal imaging plane for measurement of total cerebral blood flow by MRI

A computer-aided method for finding an optimal imaging plane for simultaneous measurement of the arterial blood inflow through the 4 vessels leading blood to the brain by phase contrast magnetic resonance imaging is presented. The method performance is compared with manual selection by two observers. The skeletons of the 4 vessels for which centerlines are generated are first extracted. Then, a global direction of the relatively less curved internal carotid arteries is calculated to determine the main flow direction. This is then used as a reference direction to identify segments of the vertebral arteries that strongly deviates from the main flow direction. These segments are then used to identify anatomical landmarks for improved consistency of the imaging plane selection. An optimal imaging plane is then identified by finding a plane with the smallest error value, which is defined as the sum of the angles between the planes normal and the vessel centerlines direction at the location of the intersections. Error values obtained using the automated and the manual methods were then compared using 9 magnetic resonance angiography (MRA) data sets. The automated method considerably outperformed the manual selection. The mean error value with the automated method was significantly lower than the manual method, 0.09±0.07 vs. 0.53±0.45, respectively (p<.0001, Students t-test). Reproducibility of repeated measurements was analyzed using Bland and Altmans test, the mean 95% limits of agreements for the automated and manual method were 0.01~0.02 and 0.43~0.55 respectively.

Efficient structures for image decomposition using directional filter banks

Image decomposition using directional filter banks is useful in discovering and extracting edge orientation cues for target detection in airborne surveillance images. Since images of interest are very large and the filtered images are not downsampled in the application of interest, conventional filtering can be computationally extremely demanding and there is a need to explore procedures to make the filtering efficient. In this paper a novel filter bank structure for directional filtering of images is proposed and its design described. The design is carried out by imposing structural constraints on the filters, which are implemented using a generalized notion of separable filtering. The structure uses one-dimensional (1-D) filters as building blocks, which are employed in novel configurations to obtain filters with narrow wedge-shaped passbands. Design procedures have been developed for constructing 16-band, 32-band, and 64- band partitions starting with either built-in or user-specified 1-D prototypes. Implementations of filters using the proposed method show significant improvement compared with conventional implementation, often more by an order of magnitude, which is also supported by a theoretical analysis of the filter complexity.

Automated Retinal Layer Segmentation in OCT Images Using Spatially Variant Filtering

We have developed a new method to segment and analyze retinal layers in optical coherence tomography (OCT) images with the intent of monitoring changes in thickness of retinal layers due to disease. OCT is an imaging modality that obtains cross-sectional images of the retina, which makes it possible to measure thickness of individual layers. In this paper we present a method that identifies six key layers in OCT images. OCT images present challenges to conventional edge detection algorithms, including that due to the presence of speckle noise which affects the sharpness of inter-layer boundaries significantly. We use a directional filter bank, which has a wedge shaped passband that helps reduce noise while maintaining edge sharpness, in contrast to previous methods that use Gaussian filter or median filter variants that reduce the edge sharpness resulting in poor edge-detection performance. This filter is utilized in a spatially variant setting which uses additional information from the intersecting scans. The validity of extracted edge cues is determined according to the amount of gray-level transition across the edge, strength, continuity, relative location and polarity. These cues are processed according to the retinal model that we have developed and the processing yields edge contours.