Pantelis A. Asvestas

Multimodal registration of retinal images using self organizing maps

In this paper, an automatic method for registering multimodal retinal images is presented. The method consists of three steps: the vessel centerline detection and extraction of bifurcation points only in the reference image, the automatic correspondence of bifurcation points in the two images using a novel implementation of the self organizing maps and the extraction of the parameters of the affine transform using the previously obtained correspondences. The proposed registration algorithm was tested on 24 multimodal retinal pairs and the obtained results show an advantageous performance in terms of accuracy with respect to the manual registration.

Contrast enhancement of images using Partitioned Iterated Function Systems

A new algorithm for the contrast enhancement of images, based on the theory of Partitioned Iterated Function System (PIFS), is presented. A PIFS consists of contractive transformations, such that the original image is the fixed point of the union of these transformations. Each transformation involves the contractive affine spatial transform of a square block, as well as the linear transform of the gray levels of its pixels. The transformation of the gray levels is determined by two parameters which adjust the brightness and the contrast of the transformed block. The PIFS is used in order to create a lowpass version of the original image. The contrast-enhanced image is obtained by adding the difference of the original image with its lowpass version, to the original image itself. The proposed algorithm uses a predefined constant value for the contrast parameter, whereas, the parameters of the affine spatial transform, as well as the parameter adjusting the brightness, are calculated using k-dimensional trees. The lowpass version of the original image is obtained applying the PIFS on the original image repeatedly while using a value for the contrast parameter that is lower than the predefined one. Quantitative and qualitative results stress the superior performance of the proposed contrast enhancement algorithm against four other widely used contrast enhancement methods; namely, linear and nonlinear unsharp masking, Contrast Limited Adaptive Histogram Equalization and Local Range Modification.

A comparative study of surface‐ and volume‐based techniques for the automatic registration between CT and SPECT brain images

Image registration of multimodality images is an essential task in numerous applications in three-dimensional medical image processing. Medical diagnosis can benefit from the complementary information in different modality images. Surface-based registration techniques, while still widely used, were succeeded by volume-based registration algorithms that appear to be theoretically advantageous in terms of reliability and accuracy. Several applications of such algorithms for the registration of CT-MRI, CT-PET, MRI-PET, and SPECT-MRI images have emerged in the literature, using local optimization techniques for the matching of images. Our purpose in this work is the development of automatic techniques for the registration of real CT and SPECT images, based on either surface- or volume-based algorithms. Optimization is achieved using genetic algorithms that are known for their robustness. The two techniques are compared against a well-established method, the Iterative Closest Point-ICP. The correlation coefficient was employed as an independent measure of spatial match, to produce unbiased results. The repeated measures ANOVA indicates the significant impact of the choice of registration method on the magnitude of the correlation (F = 4.968, p = 0.0396). The volume-based method achieves an average correlation coefficient value of 0.454 with a standard deviation of 0.0395, as opposed to an average of 0.380 with a standard deviation of 0.0603 achieved by the surface-based method and an average of 0.396 with a standard deviation equal to 0.0353 achieved by ICP. The volume-based technique performs significantly better compared to both ICP (p<0.05, Neuman Keuls test) and the surface-based technique (p<0.05, Neuman-Keuls test). Surface-based registration and ICP do not differ significantly in performance.

Thoracic non-rigid registration combining self-organizing maps and radial basis functions

An automatic three-dimensional non-rigid registration scheme is proposed in this paper and applied to thoracic computed tomography (CT) data of patients with stage III non-small cell lung cancer (NSCLC). According to the registration scheme, initially anatomical set of points such as the vertebral spine, the ribs, and shoulder blades are automatically segmented slice by slice from the two CT scans of the same patient in order to serve as interpolant points. Based on these extracted features, a rigid-body transformation is then applied to provide a pre-registration of the data. To establish correspondence between the feature points, the novel application of the self-organizing maps (SOMs) is adopted. In particular, the automatic correspondence of the interpolant points is based on the initialization of the Kohonen neural network model capable to identify 500 corresponding pairs of points approximately in the two CT sets. Then, radial basis functions (RBFs) using the shifted log function is subsequently employed for elastic warping of the image volume, using the correspondence between the interpolant points, as obtained in the previous phase. Quantitative and qualitative results are also presented to validate the performance of the proposed elastic registration scheme resulting in an alignment error of 6 mm, on average, over 15 CT paired datasets. Finally, changes of the tumor volume in respect to each reference dataset are estimated for all patients, which indicate inspiration and/or movement of the patient during acquisition of the data. Thus, the practical implementation of this scheme could provide estimations of lung tumor volumes during radiotherapy treatment planning.

Fractal dimension estimation of carotid atherosclerotic plaques from B-mode ultrasound: a pilot study

In this paper, a pilot study regarding carotid atherosclerotic plaque instability using B-mode ultrasound (US) images was carried out. The fractal dimension of plaques obtained from ten symptomatic subjects (i.e., subjects having experienced neurological symptoms) as well as from nine asymptomatic subjects, was estimated using a novel method, called the kth nearest neighbour (KNN) method. The results indicated a significant difference, as per the fractal dimension, between the two groups, providing a significantly lower value for the asymptomatic group. Moreover, the phase of the cardiac cycle (systole/diastole) during which the fractal dimension was estimated had no systematic effect on the calculations. The fractal dimension of the plaques was also estimated using a well-known method, namely the box-counting (BC) method. No significant differences between the two groups, as per the fractal dimension, were observed using the BC method. The presented pilot study suggests that the fractal dimension, estimated by the proposed method, could be used as a single determinant for the discrimination of symptomatic and asymptomatic subjects.

A Power Differentiation Method of Fractal Dimension Estimation for 2-D Signals

Fractal dimension has been used for texture analysis as it is highly correlated with the human perception of surface roughness. Several methods have been proposed for the estimation of the fractal dimension of an image. One of the most popular is via its power spectrum density, provided that it is modeled as a fractional Brownian function. In this paper, a new method, called the power differentiation method (PDM), for estimating the fractal dimension of a two-variable signal from its power spectrum density is presented. The method is first applied to noise-free data of known fractal dimension. It is also tested with noise-corrupted and quantized data. Particularly, in the case of noise-corrupted data, the modified power differentiation method (MPDM) is developed, resulting in more accurate estimation of the fractal dimension. The results obtained by the PDM and the MPDM are compared directly to those obtained using four other well-known methods of fractal dimension. Finally, preliminary results for the classification of ultrasonic liver images, obtained by applying the new method, are presented.

Abnormal P600 in heroin addicts with prolonged abstinence elicited during a working memory test.

The P600 component of event-related potentials, believed to be generated by anterior cingulate gyrus and basal ganglia, is considered as an index of aspects of second-pass parsing processes of information processing, having much in common with working memory (WM) systems. Moreover, dysfunction of these brain structures as well as WM deficits have been implicated in the pathophysiology of opioid addicts. The present study is focused on P600 elicited during a WM test in twenty heroin addicts with prolonged abstinence compared with an equal number of healthy controls. The results showed significantly prolonged latencies at right hemisphere, specifically at Fp2 abduction. Moreover, memory performance of patients did not differ from that of normal controls. These findings may indicate that abstinent heroin addicts manifest abnormal aspects of second-pass parsing processes as are reflected by the P600 latencies, elicited during a WM test. Additionally, the P600 might serve as a valuable investigative tool for a more comprehensive understanding of the neurobiological substrate of drug abuse.

Registration of electronic portal images for patient set-up verification

Images acquired from an electronic portal imaging device are aligned with digitally reconstructed radiographs (DRRs) or other portal images to verify patient positioning during radiation therapy. Most of the currently available computer aided registration methods are based on the manual placement of corresponding landmarks. The purpose of the paper is twofold: (a) the establishment of a methodology for patient set-up verification during radiotherapy based on the registration of electronic portal images, and (b) the evaluation of the proposed methodology in a clinical environment. The estimation of set-up errors, using the proposed methodology, can be accomplished by matching the portal image of the current fraction of the treatment with the portal image of the baseline treatment (reference portal image) using a nearly automated technique. The proposed registration method is tested on a number of phantom data as well as on data from four patients. The phantom data included portal images that corresponded to various positions of the phantom on the treatment couch. For each patient, a set of 30 portal images was used. For the phantom data (for both transverse and lateral portal images), the maximum absolute deviations of the translational shifts were within 1.5 mm, whereas the in-plane rotation angle error was less than 0.5 degrees. The two-way Anova revealed no statistical significant variability both within observer and between-observer measurements (P > 0.05). For the patient data, the mean values obtained with manual and the proposed registration methods were within 0.5 mm. In conclusion, the proposed registration method has been incorporated within a system, called ESTERR-PRO. Its image registration capability achieves high accuracy and both intra- and inter-user reproducibility. The system is fully operational within the Radiotherapy Department of HYGEIA Hospital in Athens and it could be easily installed in any other clinical environment since it requires standardized hardware specifications and minimal human intervention.

Detection of glaucomatous change based on vessel shape analysis

Glaucoma, a leading cause of blindness worldwide, is a progressive optic neuropathy with characteristic structural changes in the optic nerve head and concomitant visual field defects. Ocular hypertension (i.e. elevated intraocular pressure without glaucoma) is the most important risk factor to develop glaucoma. Even though a number of variables, including various optic disc and visual field parameters, have been used in order to identify early glaucomatous damage, there is a need for computer-based methods that can detect early glaucomatous progression so that treatment to prevent further progression can be initiated. This paper is focused on the description of a system based on image processing and classification techniques for the estimation of quantitative parameters to define vessel deformation and the classification of image data into two classes: patients with ocular hypertension who develop glaucomatous damage and patients with ocular hypertension who remain stable. The proposed system consists of the retinal image preprocessing module for vessel central axis segmentation, the automatic retinal image registration module based on a novel application of self organizing maps (SOMs) to define automatic point correspondence, the retinal vessel attributes calculation module to select the vessel shape attributes and the data classification module, using an artificial neural network classifier, to perform the necessary subject classification. Implementation of the system to optic disc data from 127 subjects obtained by a fundus camera at regular intervals provided a classification rate of 87.5%, underscoring the value of the proposed system to assist in the detection of early glaucomatous change.

Application of Kohonen network for automatic point correspondence in 2D medical images

In this paper, a generalized application of Kohonen Network for automatic point correspondence of unimodal medical images is presented. Given a pair of two-dimensional medical images of the same anatomical region and a set of interest points in one of the images, the algorithm detects effectively the set of corresponding points in the second image, by exploiting the properties of the Kohonen self organizing maps (SOMs) and embedding them in a stochastic optimization framework. The correspondences are established by determining the parameters of local transformations that map the interest points of the first image to their corresponding points in the second image. The parameters of each transformation are computed in an iterative way, using a modification of the competitive learning, as implemented by SOMs. The proposed algorithm was tested on medical imaging data from three different modalities (CT, MR and red-free retinal images) subject to known and unknown transformations. The quantitative results in all cases exhibited sub-pixel accuracy. The algorithm also proved to work efficiently in the case of noise corrupted data. Finally, in comparison to a previously published algorithm that was also based on SOMs, as well as two widely used techniques for detection of point correspondences (template matching and iterative closest point), the proposed algorithm exhibits an improved performance in terms of accuracy and robustness.