D. Kelekis

Bio: D. Kelekis is an academic researcher. The author has contributed to research in topics: Prophylactic cranial irradiation & Conventional PCI. The author has an hindex of 5, co-authored 9 publications receiving 338 citations.

A computer-aided diagnostic system to characterize CT focal liver lesions: design and optimization of a neural network classifier

Abstract: In this paper, a computer-aided diagnostic (CAD) system for the classification of hepatic lesions from computed tomography (CT) images is presented. Regions of interest (ROIs) taken from nonenhanced CT images of normal liver, hepatic cysts, hemangiomas, and hepatocellular carcinomas have been used as input to the system. The proposed system consists of two modules: the feature extraction and the classification modules. The feature extraction module calculates the average gray level and 48 texture characteristics, which are derived from the spatial gray-level co-occurrence matrices, obtained from the ROIs. The classifier module consists of three sequentially placed feed-forward neural networks (NNs). The first NN classifies into normal or pathological liver regions. The pathological liver regions are characterized by the second NN as cyst or "other disease". The third NN classifies "other disease" into hemangioma or hepatocellular carcinoma. Three feature selection techniques have been applied to each individual NN: the sequential forward selection, the sequential floating forward selection, and a genetic algorithm for feature selection. The comparative study of the above dimensionality reduction methods shows that genetic algorithms result in lower dimension feature vectors and improved classification performance.

Characterization of CT liver lesions based on texture features and a multiple neural network classification scheme

Abstract: In this paper, a Computer Aided Diagnosis (CAD) system for the characterization of hepatic tissue from Computed Tomography (CT) images is presented. Regions of Interest (ROI's) corresponding to normal liver, cyst, hemangioma, and hepatocellular carcinoma, are drawn by an experienced radiologist on abdominal nonenhanced CT images. For each ROI, five distinct sets of texture features are extracted using the following methods: first order statistics, spatial gray level dependence matrix, gray level difference method, Laws' texture energy measures, and fractal dimension measurements. If the dimensionality of a feature set is greater than a predefined threshold, feature selection based on a Genetic Algorithm (GA) is applied. Classification of the ROI is then carried out by a system of five neural networks (NNs), each using as input one of the above feature sets. The members of the NN system (primary classifiers) are 4-class NNs trained by the backpropagation algorithm with adaptive learning rate and momentum. The final decision of the CAD system is based on the application of a voting scheme across the outputs of the individual NNs. The multiple classification scheme using the five sets of texture features results in significantly enhanced performance, as compared to the classification performance of the individual primary classifiers.

Classification of hepatic lesions from CT images using texture features and neural networks

Abstract: 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.

A fractal analysis of CT liver images for the discrimination of hepatic lesions: a comparative study

Abstract: 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 neighbour method. The Fuzzy C-Means algorithm is finally applied revealing that the k-th nearest neighbour method outperforms the other methods; thus discriminating up to 93% of the normal parenchyma and up to 82% of the hepatocellular carcinoma, correctly.

Analysis of resting state and task-related fMRI data in small cell lung cancer patients before undertaking PCI

Abstract: As the use of prophylactic cranial irradiation (PCI) increases life expectancy in certain types of cancer, there is evidence that this technique may contribute to neurocognitive deficits. The aim of this research is to investigate functional connectivity in patients before and after being administered brain radiotherapy as well as their neurocognitive functionality, depending on the course of disease and the phase of treatment and time of clinical examination. In the current study we attempt to identify potential alterations in brain function during resting state and task-related functional magnetic resonance imaging (fMRI) in small cell lung cancer patients (SCLC) patients before PCI compared to healthy subjects. The brain regions that were activated for both groups during a sequential tapping task are consistent with previous studies. However, the patient group showed a lower level of activation in the primary somatosensory cortex bilaterally. During rest, the Default Mode Network (DMN) was identified in both groups. The results presented are subject to further investigation with larger patient and control group.

Computer and Robot Vision

Abstract: Computer and Robot Vision Vol. 1, by R.M. Haralick and Linda G. Shapiro, Addison-Wesley, 1992, ISBN 0-201-10887-1.

Comparison and Evaluation of Methods for Liver Segmentation From CT Datasets

Abstract: This paper presents a comparison study between 10 automatic and six interactive methods for liver segmentation from contrast-enhanced CT images. It is based on results from the "MICCAI 2007 Grand Challenge" workshop, where 16 teams evaluated their algorithms on a common database. A collection of 20 clinical images with reference segmentations was provided to train and tune algorithms in advance. Participants were also allowed to use additional proprietary training data for that purpose. All teams then had to apply their methods to 10 test datasets and submit the obtained results. Employed algorithms include statistical shape models, atlas registration, level-sets, graph-cuts and rule-based systems. All results were compared to reference segmentations five error measures that highlight different aspects of segmentation accuracy. All measures were combined according to a specific scoring system relating the obtained values to human expert variability. In general, interactive methods reached higher average scores than automatic approaches and featured a better consistency of segmentation quality. However, the best automatic methods (mainly based on statistical shape models with some additional free deformation) could compete well on the majority of test images. The study provides an insight in performance of different segmentation approaches under real-world conditions and highlights achievements and limitations of current image analysis techniques.

Automated medical image segmentation techniques

Abstract: Accurate segmentation of medical images is a key step in contouring during radiotherapy planning. Computed topography (CT) and Magnetic resonance (MR) imaging are the most widely used radiographic techniques in diagnosis, clinical studies and treatment planning. This review provides details of automated segmentation methods, specifically discussed in the context of CT and MR images. The motive is to discuss the problems encountered in segmentation of CT and MR images, and the relative merits and limitations of methods currently available for segmentation of medical images.

Synthetic data augmentation using GAN for improved liver lesion classification

Abstract: In this paper, we present a data augmentation method that generates synthetic medical images using Generative Adversarial Networks (GANs). We propose a training scheme that first uses classical data augmentation to enlarge the training set and then further enlarges the data size and its diversity by applying GAN techniques for synthetic data augmentation. Our method is demonstrated on a limited dataset of computed tomography (CT) images of 182 liver lesions (53 cysts, 64 metastases and 65 hemangiomas). The classification performance using only classic data augmentation yielded 78.6% sensitivity and 88.4% specificity. By adding the synthetic data augmentation the results significantly increased to 85.7% sensitivity and 92.4% specificity.