D. Cavouras

Improving brain tumor characterization on MRI by probabilistic neural networks and non-linear transformation of textural features

The aim of the present study was to design, implement and evaluate a software system for discriminating between metastatic and primary brain tumors (gliomas and meningiomas) on MRI, employing textural features from routinely taken T1 post-contrast images. The proposed classifier is a modified probabilistic neural network (PNN), incorporating a non-linear least squares features transformation (LSFT) into the PNN classifier. Thirty-six textural features were extracted from each one of 67 T1-weighted post-contrast MR images (21 metastases, 19 meningiomas and 27 gliomas). LSFT enhanced the performance of the PNN, achieving classification accuracies of 95.24% for discriminating between metastatic and primary tumors and 93.48% for distinguishing gliomas from meningiomas. To improve the generalization of the proposed classification system, the external cross-validation method was also used, resulting in 71.43% and 81.25% accuracies in distinguishing metastatic from primary tumors and gliomas from meningiomas, respectively. LSFT improved PNN performance, increased class separability and resulted in dimensionality reduction.

Evaluating x-ray detectors for radiographic applications: A comparison of ZnSCdS:Ag with and screens

ZnSCdS:Ag was evaluated as a radiographic image receptor and was compared with and phosphors often used in radiography. The evaluation of a radiographic receptor was modelled as a three-step process: (i) determination of light output intensity as related to the input radiation dose, (ii) determination of visible light characteristics with respect to radiographic optical detectors, and (iii) determination of image information transfer efficiency. The light intensity emitted per unit of x-ray exposure rate was measured and theoretically calculated for laboratory prepared screens with coating thicknesses from 20 to and tube voltages from 50 to 250 kVp. ZnSCdS:Ag light intensity was higher than that of or for tube voltages less than 70 and 80 kVp respectively. ZnSCdS:Ag displayed the highest x-ray to light conversion efficiency (0.207) and had optical properties close to those of and , and its emission spectrum was well matched to optical detectors. The image information transfer properties described by the modulation transfer function, the quantum noise transfer function, and the detective quantum efficiency were calculated for both general radiographic and mammographic conditions and were found to be intermediate between those of and screens. Conclusively, ZnSCdS:Ag is an efficient phosphor well suited to radiography.

Modeling granular phosphor screens by Monte Carlo methods

The intrinsic phosphor properties are of significant importance for the performance of phosphor screens used in medical imaging systems. In previous analytical-theoretical and Monte Carlo studies on granular phosphor materials, values of optical properties, and light interaction cross sections were found by fitting to experimental data. These values were then employed for the assessment of phosphor screen imaging performance. However, it was found that, depending on the experimental technique and fitting methodology, the optical parameters of a specific phosphor material varied within a wide range of values, i.e., variations of light scattering with respect to light absorption coefficients were often observed for the same phosphor material. In this study, x-ray and light transport within granular phosphor materials was studied by developing a computational model using Monte Carlo methods. The model was based on the intrinsic physical characteristics of the phosphor. Input values required to feed the model can be easily obtained from tabulated data. The complex refractive index was introduced and microscopic probabilities for light interactions were produced, using Mie scattering theory. Model validation was carried out by comparing model results on x-ray and light parameters (x-ray absorption, statistical fluctuations in the x-ray to light conversion process, number of emitted light photons, output light spatial distribution) with previous published experimental data on Gd2O2S: Tb phosphor material (Kodak Min-R screen). Results showed the dependence of the modulation transfer function (MTF) on phosphor grain size and material packing density. It was predicted that granular Gd2O2S: Tb screens of high packing density and small grain size may exhibit considerably better resolution and light emission properties than the conventional Gd2O2S: Tb screens, under similar conditions (x-ray incident energy, screen thickness).

Design and implementation of an SVM-based computer classification system for discriminating depressive patients from healthy controls using the P600 component of ERP signals

A computer-based classification system has been designed capable of distinguishing patients with depression from normal controls by event-related potential (ERP) signals using the P600 component. Clinical material comprised 25 patients with depression and an equal number of gender and aged-matched healthy controls. All subjects were evaluated by a computerized version of the digit span Wechsler test. EEG activity was recorded and digitized from 15 scalp electrodes (leads). Seventeen features related to the shape of the waveform were generated and were employed in the design of an optimum support vector machine (SVM) classifier at each lead. The outcomes of those SVM classifiers were selected by a majority-vote engine (MVE), which assigned each subject to either the normal or depressive classes. MVE classification accuracy was 94% when using all leads and 92% or 82% when using only the right or left scalp leads, respectively. These findings support the hypothesis that depression is associated with dysfunction of right hemisphere mechanisms mediating the processing of information that assigns a specific response to a specific stimulus, as those mechanisms are reflected by the P600 component of ERPs. Our method may aid the further understanding of the neurophysiology underlying depression, due to its potentiality to integrate theories of depression and psychophysiology.

Using handheld devices for real-time wireless teleconsultation

Recent advances in the hardware of handheld devices, opened up the way for newer applications in the healthcare sector, and more specifically, in the teleconsultation field. Out of these devices, this paper focuses on the services that personal digital assistants and smartphones can provide to improve the speed, quality and ease of delivering a medical opinion from a distance and laying the ground for an all-wireless hospital. In that manner, PDAs were used to wirelessly support the viewing of digital imaging and communication in medicine (DICOM) images and to allow for mobile videoconferencing while within the hospital. Smartphones were also used to carry still images, multiframes and live video outside the hospital. Both of these applications aimed at increasing the mobility of the consultant while improving the healthcare service.

An evaluation of the Y2O3:Eu3+ scintillator for application in medical x-ray detectors and image receptors.

The suitability off a Y2O3:Eu3+ scintillator for use in radiation detectors and medical image receptors was studied. Y2O3:Eu3+ was used in the form of laboratory prepared screens of different coating thicknesses. The x-ray luminescence efficiency of the screens was measured for tube voltages between 50-200 kVp and in both transmission and reflection modes of observation. The intrinsic x ray to light conversation efficiency (nc) and other parameters of the Y2O3:Eu3+ phosphor material related to optical scattering, absorption, and reflection were determined. These were used in the calculation of the image transfer characteristics, MTF and zero frequency DQE, for various screen coating thicknesses and x-ray tube voltages. The light emission spectrum of Y2O3:Eu3+ was measured (narrow band peak at 613 nm) and its spectral compatibility to the spectral sensitivity of several commonly employed optical photon detectors was determined. The x-ray luminescence efficiency varied with x-ray tube voltage, attaining maximum value at about 80 kVp for all screen thicknesses. It also varied with coating thickness reaching 25 microW m(-2)/mR s(-1) and 18 microW m(-2)/mR s(-1) at 175 mg/cm2 for reflection and transmission modes, respectively. The intrinsic x ray to light conversion efficiency and the image transfer characteristics were found to be comparable to several commercially used phosphors: nc = 0.095, MTF0.05 ranged between 10 and 25 line pairs per mm and peal values of DQE(0) varied between 0.33 and 0.14 in the coating thickness and kVp ranges useful for x-ray imaging. Spectral compatibility to some red sensitive optical photon detectors was excellent (0.9 or better). Results indicated that Y2O3:Eu3+ is a medium to high overall performance material that could be used in medical x-ray detectors and image receptors.

The size of the intra- and extraventricular cerebrospinal fluid compartments in children with idiopathic benign widening of the frontal subarachnoid space

The aim of this study was to quantify the intra- and extraventricular cerebrospinal fluid (CSF) spaces in children with benign enlargement of the frontal subarachnoid space (BE). The infra-and supratentorial CSF compartments were measured in 61 CT examinations of children with BE, 3–27 months old, and compared with those of 96 CT examinations considered normal. Measurements of the ventricular system, and the pontine and chiasmatic cisterns were related to cranial size. In all children with BE the lateral and third ventricles were dilated and the chiasmatic cistern was widened. The subarachnoid space was wider than the upper limits in the control group, in the frontal region (4mm), and the anterior interhemispheric (4mm) and Sylvian (3 mm) fissures. The infratentorial CSF compartments, the occipital subarachnoid space, the posterior part of the interhemispheric fissure and, in most cases, the cortical sulci were normal in size in children with BE. The majority were macrocephalic or had rapid head growth but there were also normocephalic children with normal head growth. The size of the posterior fossa was within the normal range in all children with BE. Idiopathic BE is not uncommon in children up to about 3 years old who are healthy or have minimal neurological disturbance and is characterised by a specific pattern of widening of the supratentorial CSF compartments.

Multi-scaled morphological features for the characterization of mammographic masses using statistical classification schemes

OBJECTIVE A comprehensive signal analysis approach on the mammographic mass boundary morphology is presented in this article. The purpose of this study is to identify efficient sets of simple yet effective shape features, employed in the original and multi-scaled spectral representations of the boundary, for the characterization of the mammographic mass. These new methods of mass boundary representation and processing in more than one domain greatly improve the information content of the base data that is used for pattern classification purposes, introducing comprehensive spectral and multi-scale wavelet versions of the original boundary signals. The evaluation is conducted against morphological and diagnostic characterization of the mass, using statistical methods, fractal dimension analysis and a wide range of classifier architectures. METHODS AND MATERIALS This study consists of (a) the investigation of the original radial distance measurements under the complete spectrum of signal analysis, (b) the application of curve feature extractors of morphological characteristics and the evaluation of the discriminative power of each one of them, by means of statistical significance analysis and dataset fractal dimension, and (c) the application of a wide range of classifier architectures on these morphological datasets, in order to conduct a comparative evaluation of the efficiency and effectiveness of all architectures, for mammographic mass characterization. Radial distance signal was exploited using the discrete Fourier transform (DFT) and the discrete wavelet transform (DWT) as additional carrier signals. Seven uniresolution feature functions were applied over these carrier signals and multiple shape descriptors were created. Classification was conducted against mass shape type and clinical diagnosis, using a wide range of linear and non-linear classifiers, including linear discriminant analysis (LDA), least-squares minimum distance (LSMD), k-nearest neighbor (k-NN), radial basis function (RBF) and multi-layered perceptron (MLP) neural networks (NN), and support vector machines (SVM). Fractal analysis was employed as a dataset analysis tool in the feature selection phase. The discriminative power of the features produced by this composite analysis is subsequently analyzed by means of multivariate analysis of variance (MANOVA) and tested against two distinct classification targets, namely (a) the morphological shape type of the mass and (b) the histologically verified clinical diagnosis for each mammogram. RESULTS Statistical analysis and classification results have shown that the discrimination value of the features extracted from the DWT components and especially the DFT spectrum, are of great importance. Furthermore, much of the information content of the curve features in the case of DFT and DWT datasets is directly related to the texture and fine-scale details of the corresponding envelope signal of the spectral components. Neural classifiers outperformed all other methods (SVM not used because they are mainly two-class classifiers) with overall success rate of 72.3% for shape type identification, while SVM achieved the overall highest 91.54% for clinical diagnosis. Receiver operating characteristic (ROC) analysis has been employed to present the sensitivity and specificity of the results of this study.

Automated diagnosis of brain tumours astrocytomas using probabilistic neural network clustering and support vector machines

A computer-aided diagnosis system was developed for assisting brain astrocytomas malignancy grading. Microscopy images from 140 astrocytic biopsies were digitized and cell nuclei were automatically segmented using a Probabilistic Neural Network pixel-based clustering algorithm. A decision tree classification scheme was constructed to discriminate low, intermediate and high-grade tumours by analyzing nuclear features extracted from segmented nuclei with a Support Vector Machine classifier. Nuclei were segmented with an average accuracy of 86.5%. Low, intermediate, and high-grade tumours were identified with 95%, 88.3%, and 91% accuracies respectively. The proposed algorithm could be used as a second opinion tool for the histopathologists.

A study of the variation of colonic positioning in the pararenal space as shown by computed tomography

In a review of 1708 consecutive CT examinations of the abdomen the position of the ascending and descending colon in relation to the posterior and lateral edge of the kidney was studied. It was found that part of the colon was positioned posterior or posterolateral to the kidneys edge in percentages that varied between 14.2% and 0.9% in the different sex groups at the levels of upper, mid- and lower poles of the right and left kidney. It is concluded that this anatomical variation should be known if colon perforation is to be avoided during percutaneous nephrostomy or biopsy.