The aim of this study was to design a pattern recognition system for assisting the diagnosis of breast lesions, using image information from Ultrasound (US) and Digital Mammography (DM) imaging modalities. State-of-art computer technology was employed based on commercial Graphics Processing Unit (GPU) cards and parallel programming. An experienced radiologist outlined breast lesions on both US and DM images from 59 patients employing a custom designed computer software application. Textural features were extracted from each lesion and were used to design the pattern recognition system. Several classifiers were tested for highest performance in discriminating benign from malignant lesions. Classifiers were also combined into ensemble schemes for further improvement of the system's classification accuracy. Following the pattern recognition system optimization, the final system was designed employing the Probabilistic Neural Network classifier (PNN) on the GPU card (GeForce 580GTX) using CUDA programming framework and C++ programming language. The use of such state-of-art technology renders the system capable of redesigning itself on site once additional verified US and DM data are collected. Mixture of US and

The frequency histogram of connected elements (FHCE) is a recently proposed algorithm that has successfully been applied in various medical image segmentation tasks. The FHCE is based on the idea that most pixels belong to the same class as their neighbouring pixels. However, the FHCE performance relies to a great extent on the optimal selection of a threshold parameter. Since evaluating segmentation results is a highly subjective process, a collection of threshold values must typically be examined. No algorithm has been proposed to automate the determination of the threshold parameter value of the FHCE. This study presents a method based on the fuzzy C-means clustering algorithm, designed to automatically generate optimal threshold values for the FHCE. This new approach was applied as a part of a structured sequence of image processing steps in order to facilitate segmentation of microcalcifications in digitized mammograms. A unique threshold value was generated for each mammogram, taking into account the different grey-level patterns based on different compositions of various breast tissues in it. The segmentation algorithm was tested on 100 mammograms (50 collected from the Mammographic Image Analysis Society and 50 normal mammograms onto which a number of simulated microcalcifications were generated). The algorithm was able to detect subtle microcalcifications with sensitivity ranging from 93 to 98%, False alarm ratio from 3 to 5% and false negatives variability from 2 to 3%.

Breast cancer is one of the most frequently diagnosed cancers among women. Several techniques have been developed to help in the early detection of breast cancer such as conventional and digital x-ray mammography, positron and single-photon emission mammography, etc. A key advantage in digital mammography is that images can be manipulated as simple computer image files. Thus non-dedicated commercially available image manipulation software can be employed to process and store the images. The image processing tools of the Photoshop (CS 2) software usually incorporate digital filters which may be used to reduce image noise, enhance contrast and increase spatial resolution. However, improving an image quality parameter may result in degradation of another. The aim of this work was to investigate the influence of three sharpening filters, named hereafter sharpen, sharpen more and sharpen edges on image resolution and noise. Image resolu- tion was assessed by means of the Modulation Transfer Function (MTF).In conclusion it was found that the correct use of commercial non-dedicated software on digital mammograms may improve some aspects of image quality.

Accurate detection and segmentation of Intracranial Aneurysms in Computed Tomography Angiography is of major importance for radiologists. In this study, a pixel-based classification algorithm was designed for localization of aneurysms in CTA images. A series of thirty-eight DICOM CTA brain images, was collected from the Department of Radiology of the University Hospital of Patras, Greece. Pixel classification was performed as a twolevel hierarchical decision tree. In the first level, the bone-class was discriminated from the vesselparenchyma class. In the second level, the vesselclass was discriminated from the parenchyma-class. Segmented images were evaluated by an expert radiologist regarding missed, false, or correctly segmented brain blood vessels. The method proved accurate enough for brain blood vessel detection to be used in 3-D blood vessel reconstruction.

This study presents a novel method for automatic measurement of the anteroposterior diameter of the thyroid gland, in ultrasound images. Anteroposterior diameter is widely considered a basic parameter for goiter assessment. Goiter refers to abnormal enlargement of the thyroid gland and it is associated with iodine deficiency and various pathologic conditions including Hashimoto’s thyroiditis, and growing nodules. The proposed method is based on processing local histogram information extracted from vertically sampled horizontal image stripes. The anteroposterior diameter can be derived by measuring the distance between the hyperechoic lines that bound the thyroid lobes as depicted in longitudinal ultrasound images. Extensive experiments were carried out on a set of B-mode thyroid ultrasound images, in order to evaluate the effectiveness of the proposed method. The results advocate to the feasibility of its use for objective measurement of the anteroposterior diameter of the thyroid gland.

The aim of the present study was to examine the radiation absorption efficiency and the light emission efficiency of Lu2SiO5:Ce (LSO) powder scintillator under x-ray mammographic imaging conditions. LSO is an efficient and extremely fast scintillator, employed in positron emission tomography, which however has never been used in X-ray imaging. For the purposes of the present study a 25 mg/cm thick scintillating screen was prepared in our laboratory, by sedimentation of Lu2SiO5:Ce powder. Absolute luminescence efficiency measurements were performed within the range of X-ray tube voltages (22-40 kVp) used in mammographic applications. Parameters related to X-ray detection, i.e. the energy absorption efficiency and the quantum detective efficiency were calculated. A theoretical model, describing radiation and light transfer, was employed to fit experimental data and to estimate values of the intrinsic conversion efficiency and the light attenuation coefficients of the screen. The Lu2SiO5:Ce powder scintillator material showed to exhibit high radiation absorption properties in comparison with Gd2O2S:Tb and CsI:Tl and considering its very short decay time, it may successfully be employed in digital mammography.

This paper investigates a novel computational approach to thyroid tissue characterization in ultrasound images. It is based on the hypothesis that tissues in thyroid ultrasound images may be differentiated by directionality patterns. These patterns may not be always distinguishable by the human eye because of the dominant image noise. The encoding of the directional patterns in the thyroid ultrasound images is realized by means of radon transform features. A representative set of ultrasound images, acquired from 66 patients was constructed to perform experiments that test the validity of the initial hypothesis. Supervised classification experiments showed that the proposed approach is capable of discriminating normal and nodular thyroid tissues, whereas nodular tissues can be further characterized as of high or low malignancy risk.

A wavelet-based method for speckle suppression in ultrasound images of the thyroid gland is introduced. The classification of image pixels as speckle or part of important image structures is accomplished within the framework of back-propagation tracking and singularity detection of wavelet transform modulus maxima, derived from inter-scale analysis. A comparative study with other de-speckling techniques, employing quantitative indices, demonstrated that our method achieved superior speckle reduction performance and edge preservation properties. Moreover, a questionnaire regarding qualitative imaging parameters, emanating from various visual observations, was employed by two experienced physicians in order to evaluate the algorithm's speckle suppression efficiency.

The aim of the present study was to evaluate the imaging performance of YAP:Ce powder scintillating screens under exposure conditions employed in diagnostic radiology (50–140kV). Various screens were prepared in our laboratory from YAP: Ce powder (Phosphor Technology, Ltd.), with coating thickness ranging from 53 to 110mg/cm2. The imaging performance of the screens was assessed by experimental determination of the modulation transfer function (MTF) and the noise transfer function (NTF). MTF was determined by the edge spread function (ESF) method while NTF was estimated by noise power spectrum (NPS) measurements after uniform screen irradiation. In addition, parameters related to overall image quality, such as the signal-to-noise ratio transfer (MTF/NTF), were estimated. MTF curves were affected by the beam hardening effects caused by the patient simulating 20mm thick aluminum phantom. Under these conditions MTF values were found to increase with the mean X-ray photon energy. A similar effect was observed for NTF curves. Results were compared with data obtained on CsI:Tl scintillator. Taking into consideration the very fast response of YAP:Ce, these data may be of interest in designing X-ray imaging detectors.

The aim of this study was to examine the angular distribution of the light emitted from radiation-excited scintillators in medical imaging detectors. This distribution diverges from Lambert's cosine law and affects the light emission efficiency of scintillators, hence it also affects the dose burden to the patient. In the present study, the angular distribution was theoretically modeled and was used to fit experimental data on various scintillator materials. Results of calculations revealed that the angular distribution is more directional than that predicted by Lambert's law. Divergence from this law is more pronounced for high values of light attenuation coefficient and thick scintillator layers (screens). This type of divergence reduces light emission efficiency and hence it increases the incident X-ray flux required for a given level of image brightness.

The purpose of the present study was to evaluate, under X-ray medical imaging conditions, the X-ray luminescence efficiency (XLE) and the optical quantum gain (OQG) of the Gd2SiO5:Ce scintillator in single crystal form, suitable for tomographic applications. Intrinsic physical properties and light emission characteristics of the Gd2SiO5:Ce scintillator, were also studied. Both experimental and Monte Carlo techniques were used. Various X-ray tube voltages (40–140kV), currently employed in X-ray imaging applications, were used. XLE was found to vary slowly with X-ray tube voltage from (0.021±0.003) to (0.017±0.003). OQG varied from (317±18) to (466±23) light photons per incident X-ray. These values were adequately high for imaging applications using the particular energy range. Additionally, it was found by Monte Carlo simulations that for crystal thicknesses higher than 0.5cm both XLE and OQG reached saturation levels, indicating that higher thickness crystals are of no practical use in X-ray medical imaging.

The aim of the present study was to examine the light emission efficiency of Lu2SiO5:Ce (LSO) powder scintillator under X-ray mammographic imaging conditions. Powder LSO scintillator has never been used in X-ray imaging. For the purposes of the present study, a 25mg/cm2 thick scintillating screen was prepared in our laboratory, by sedimentation of Lu2SiO5:Ce powder. Absolute luminescence efficiency measurements were performed within the range of X-ray tube voltages (22–49kVp) used in mammographic applications. Parameters related to X-ray detection, i.e. the energy absorption efficiency (EAE) and the quantum detection efficiency (QDE) were calculated. A theoretical model, describing radiation and light transfer, was employed to fit experimental data and to estimate values of the intrinsic conversion efficiency and the light attenuation coefficients of the screen. The spectral compatibility of the LSO powder scintillator to mammographic X-ray films and to various electronic optical detectors was determined by performing light emission spectrum measurements and by taking into account the spectral sensitivity of the optical detectors. Results in the voltage range used in mammography showed that Lu2SiO5:Ce powder scintillator has approximately 10% higher values of QDE and 4.5% higher values of EAE than Gd2O2S:Tb.

A hybrid model for thyroid nodule boundary detection on ultrasound images is introduced. The segmentation model combines the advantages of the “á trous” wavelet transform to detect sharp gray-level variations and the efficiency of the Hough transform to discriminate the region of interest within an environment with excessive structural noise. The proposed method comprise three major steps: a wavelet edge detection procedure for speckle reduction and edge map estimation, based on local maxima representation. Subsequently, a multiscale structure model is utilised in order to acquire a contour representation by means of local maxima chaining with similar attributes to form significant structures. Finally, the Hough transform is employed with ‘a priori’ knowledge related to the nodule's shape in order to distinguish the nodule's contour from adjacent structures. The comparative study between our automatic method and manual delineations demonstrated that the boundaries extracted by the hybrid model are closely correlated with that of the physicians. The proposed hybrid method can be of value to thyroid nodules’ shape-based classification and as an educational tool for inexperienced radiologists.

Nodular thyroid disease is a frequent occurrence in clinical practice and it is associated with increased risk of thyroid cancer and hyperfunction. In this paper we propose a novel method for computer-aided detection of thyroid nodules in ultrasound (US) images. The proposed method is based on a level-set image segmentation approach that takes into account the inhomogeneity of the US images. This novel method was experimentally evaluated using US images acquired from 35 patients. The results show that the proposed method achieves more accurate delineation of the thyroid nodules in the US images and faster convergence than other relevant methods.

The aim of the present study was to evaluate ZnS:Cu phosphor for use in X-ray mammographic detectors. This phosphor has never been used in medical imaging probably due to its moderate scintillation decay time. However, it may be suitable for non-dynamic medical imaging, due to its “green” emission spectrum, which is compatible with the sensitivity of many currently used photodetectors, and its high X-ray to light intrinsic conversion efficiency. ZnS:Cu phosphor powder was used to prepare several test screens in laboratory. Parameters related to light emission and image quality properties were experimentally as well as theoretically evaluated and compared to those of other known ZnS-based phosphor materials. Results showed that ZnS:Cu performed adequately well in the mammographic energy range.

Thirteen (13) patients with liver neuroendocrine carcinomas metastases, positive for somatostatin receptors, confirmed by scintigraphy were infused with 4070-7030 MBq per session of In-111-octreotide after selective hepatic catheterization, exploiting the catastrophic activity of Indium Auger and Internal Conversion electron emission on cell DNA. Evaluation of the treatment was assessed by ultrasonography (US) as well as by computed tomography and/or magnetic resonance imaging scans. US appears to be the imaging procedure of choice because the examination is sensitive for evaluating lesions' edema and cystic components, provides precise measurement of tumor size, and is inexpensive. Degeneration US signs were classified in stage I (an echolucent rim in the periphery of the lesion), stage IIa (lesion with large cystic spaces), stage IIb (tiny cystic spaces) and stage III (absorption of the cystic component or stable cystic remnants).

The aim of this study was to examine the angular distribution of the light emitted from radiation excited scintillators in medical imaging detectors. This distribution diverges from Lambert’s cosine law and affects the light emission efficiency of scintillators. Hence it also affects the dose burden to the patient. In the present study the angular distribution was theoretically modeled and was used to fit experimental data on various scintillator materials. Results of calculations revealed that the angular distribution is more directional than that predicted by Lambert’s law. Divergence from this law is more pronounced for high values of light attenuation coefficient and thick scintillator layers (screens). This type of divergence reduces light emission efficiency and hence it increases the incident x-ray flux required for a given level of image brightness.

This paper presents a computer-aided approach for nodule delineation in thyroid ultrasound (US) images. The developed algorithm is based on a novel active contour model, named variable background active contour (VBAC), and incorporates the advantages of the level set region-based active contour without edges (ACWE) model, offering noise robustness and the ability to delineate multiple nodules. Unlike the classic active contour models that are sensitive in the presence of intensity inhomogeneities, the proposed VBAC model considers information of variable background regions. VBAC has been evaluated on synthetic images, as well as on real thyroid US images. From the quantification of the results, two major impacts have been derived: 1) higher average accuracy in the delineation of hypoechoic thyroid nodules, which exceeds 91%; and 2) faster convergence when compared with the ACWE model.