Search

A novel automatic registration-fusion scheme based on similarity measures is proposed in this paper. It utilizes the geometrical configuration between the X-ray source, the CCD sensor and the target tissue using the appropriate geometric transformations, as well as the calculation of similarity measures between two dental radiographic images to be registered. Moreover, a fusion process has been developed to combine information from registered dental images. Result on clinical data reveals the advantageous performance of the proposed automatic registration method compared to the manual one using an independent evaluation criterion such as the mutual information criterion. Furthermore, the automatic registration approach outperforms despite the fuzzy dental boundaries and the lack of characteristic edges of the radiographic images. These preliminary findings may be used in order to establish the use of the proposed automatic method relieving the expert from the tedious and time-consuming task of defining analogous markers between the two radiographic images., Presented at Annual International Conference of the IEEE Engineering in Medicine and Biology Society (23rd) held in Istanbul, Turkey on 25-28 Oct 2001. See also ADM001351 for entire conference on cd-rom.

In this paper, we investigate the possibility of using the fractal dimension to characterize carotid atheromatous plaques from B-mode ultrasound images. The images were obtained from ten symptomatic and nine symptomatic subjects, Symptomatic subjects included patients with previous history, of cerebral events, whereas asymptomatic ones had no evidence of any cerebral symptoms prior to the time of the investigation, For each subject, a sequence of images was collected corresponding to 2-3 cardiac cycles. The boundary, of the atheromatous plaque was defined by an expert in three different images of each sequence, corresponding to systole, diastole and a random phase of the cardiac cycle. The fractal dimension of each plaque was estimated using a new method, namely the k-th nearest neighbors method. The results showed that the values of the fractal dimension in the symptomatic group were significantly higher than those in the asymptomatic group., Presented at Annual International Conference of the IEEE Engineering in Medicine and Biology Society (23rd) held in Istanbul, Turkey on 25-28 Oct 2001. See also ADM001351 for entire conference on cd-rom.

In this paper a computer-aided diagnostic system for the classification of hepatic lesions from Computed Tomography (CT) images is presented. Regions of Interest (ROI's) taken from non-enhanced CT images of normal liver, hepatic cysts, hemangiomas, and hepatocellular carcinomas (a total of 147 samples), have been used as input to the system. The system consists of two levels: the feature extraction and the classification levels. The feature extraction level calculates the average grey scale and 48 texture characteristics, which are derived from the spatial grey-level co-occurrence matrices, obtained from the ROI's. The classifier level consists of three sequentially placed feed-forward Neural Networks (NN's), which are activated sequentially. The first NN classifies into normal or pathological liver regions. The pathological liver regions are classified by the second NN into cysts or "other disease". The third NN classifies "other disease" into hemangiomas and hepatocellular carcinomas. In order to enhance the performance of the classifier and improve the execution time, the dimensionality of the initial feature vector has been reduced using the sequential forward floating selection method for each individual NN input vector. A total classification rate of 98% has been achieved., Presented at Annual International Conference of the IEEE Engineering in Medicine and Biology Society (23rd) held in Istanbul, Turkey on 25-28 Oct 2001. See also ADM 001351 for entire conference on CD-ROM. The original document contains color images.

A quantitative study for the discrimination of different hepatic lesions is presented in this paper. The study is based on the fractal analysis of CT liver images in order to estimate their fractal dimension and to differentiate normal liver parenchyma from hepatocellular carcinoma. Four fractal dimension estimators have been implemented throughout this work; three well-established methods and a novel implementation of a method. Analytically, these methods correspond to the power spectrum method, the box counting method, the morphological fractal estimator and the novel modification of the kth-nearest neighbor method. The Fuzzy C-Means algorithm is finally applied revealing that the k-th nearest neighbor method outperforms the other methods; thus discriminating up to 93% of the normal parenchyma and up to 82% of the hepatocellular carcinoma, correctly., Presented at Annual International Conference of the IEEE Engineering in Medicine and Biology Society (23rd) held in Istanbul, Turkey on 25-28 Oct 2001. See also ADM001351 for entire conference on cd-rom.

A collaborative environment for virtual simulation in radiotherapy treatment planning is presented, The environment architecture is based on both off-line and on-line communication of data under a secure framework and can be directly integrated into the infrastructure of a radiotherapy department The on-line collaboration is based on the simultaneous execution of all actions at both collaborating sites, and prerequisites the off-line communication of the data set on which the collaboration will be performed A technical pilot study has been carried out on clinical sites, monitoring the performance of the implementation, and revealed high interactivity during an on-line collaboration session, Papers from 23rd Annual International Conference of the IEEE Engineering in Medicine and Biology Society, October 25-26, 2001 held in Istanbul, Turkey. See also ADM001351 for entire conference on cd-rom.