1,721,095 research outputs found
Application of an expectation maximization method to the reconstruction of X-ray-tube spectra from transmission data
No full textAdvantages of quasi-monochromatic X-ray sources in absorption mammography
No full textMammography is a challenging field of medical imaging. Early detection of breast cancer requires identifying small contrast details. The choice of the appropriate monochromatic energy enhances the visibility of such details. Thomson scattering source can provide tunable quasi-monochromatic X-ray beams. In this work, we investigate by Monte Carlo simulations the optimal monochromatic energy to image mammographic phantoms. In order to mimic a Thomson scattering source, we consider the effect on image quality of the presence of an energy spread and of the presence of higher-order harmonics
Compact x-ray sources for mammographic applications: Monte Carlo simulations of image quality
No full textThomson scattering x-ray sources can provide spectral distributions that are ideally suited for mammography with sufficient fluence rates. In this article, the authors investigate the effects of different spectral distributions on the image quality in simulated images of a breast mammographic phantom containing details of different compositions and thicknesses. They simulated monochromatic, quasimonochromatic, and polychromatic x-ray sources in order to define the energy for maximum figure of merit (signal-difference-to-noise ratio squared/mean glandular dose), the effect of an energy spread, and the effect of the presence of higher-order harmonics. The advantages of these sources with respect to conventional polychromatic sources as a function of phantom and detail thickness were also investigated. The results show that the energy for the figure of merit peak is between 16 and 27.4 keV, depending on the phantom thickness and detail composition and thickness. An energy spread of about 1 keV standard deviation, easily achievable with compact x-ray sources, does not appreciably affect the image quality
CMOS APS detector characterization for quantitative X-ray imaging
No full textAn X-ray Imaging detector based on CMOS Active Pixel Sensor and structured scintillator is characterized for quantitative X-ray imaging in the energy range 11-30 keV. Linearity, dark noise, spatial resolution and flat-field correction are the characteristics of the detector subject of investigation. The detector response, in terms of mean Analog-to-Digital Unit and noise, is modeled as a function of the energy and intensity of the X-rays. The model is directly tested using monochromatic X-ray beams and it is also indirectly validated by means of polychromatic X-ray-tube spectra. Such a characterization is suitable for quantitative X-ray imaging and the model can be used in simulation studies that take into account the actual performance of the detector
Start-to-end simulation of a Thomson source for mammography
No full textThomson scattering X-ray sources have many features which are of relevance for several applications: the main one is the capability to produce intense, quasi-monochromatic, tunable X-ray beams, after collimation, still with a reasonably small size apparatus. Applications to medical physics are straightforward, in particular in mammography where dose control in screening programs is the main relevant issue. An optimal choice of the X-ray energy to image the breast will result in a best image quality and hence will lead to a dose reduction. A Thomson scattering source is presently under development at the Frascati National Laboratories (LNF) of INFN (Istituto Nazionale di Fisica Nucleare). A complete simulation of the source including electron beam, laser beam, Thomson interaction and X-ray imaging is presented. The X-rays are generated in the energy range suitable for mammography and used to generate images of a mammographic phantom. Image quality is evaluated in terms of dose efficiency and compared to those obtained by monochromatic beams and conventional X-ray tubes. (C) 2010 Elsevier B.V. All rights reserved. RI Oliva, Piernicola/E-5839-201
Multi-Site MRI Data Harmonization with an Adversarial Learning Approach: Implementation to the Study of Brain Connectivity in Autism Spectrum Disorders
Full text linkMagnetic resonance imaging (MRI) nowadays plays an important role in the identification of brain underpinnings in a wide range of neuropsychiatric disorders, including Autism Spectrum Disorders (ASD). Characterizing the hallmarks in these pathologies is not a straightforward task and machine learning (ML) is certainly one of the most promising tools for addressing complex and non-linear problems. ML algorithms and, in particular, deep neural networks (DNNs), need large datasets in order to be properly trained and thus ensure generalization capabilities on new data. Large datasets can be obtained by collecting images from different centers, thus bringing unavoidable biases in the analysis due to differences in hardware and scanning protocols between different centers. In this work, we dealt with the issue of multicenter MRI data harmonization by comparing two different approaches: the analytical ComBat-GAM procedure, whose effectiveness is already documented in the literature, and an originally developed site-adversarial deep neural network (ad-DNN). The latter aims to perform a classification task while simultaneously searching for site-relevant patterns in order to make predictions free from site-related biases. As a case study, we implemented DNN and ad-DNN classifiers to distinguish subjects with ASD with respect to typical developing controls based on functional connectivity measures derived from data of the multicenter ABIDE collection. The classification performance of the proposed ad-DNN, measured in terms of the area under the ROC curve (AUC), achieved the value of AUC = 0.70±0.03, which is comparable to that obtained by a DNN on data harmonized according to the analytical procedure (AUC = 0.71±0.01). The relevant functional connectivity alterations identified by both procedures showed an agreement between each other and with the patterns of neuroanatomical alterations previously detected in the same cohort of subjects
Exploring Autism Spectrum Disorder: A Comparative Study of Traditional Classifiers and Deep Learning Classifiers to Analyze Functional Connectivity Measures from a Multicenter Dataset
Full text linkThe investigation of functional magnetic resonance imaging (fMRI) data with traditional machine learning (ML) and deep learning (DL) classifiers has been widely used to study autism spectrum disorders (ASDs). This condition is characterized by symptoms that affect the individual’s behavioral aspects and social relationships. Early diagnosis is crucial for intervention, but the complexity of ASD poses challenges for the development of effective treatments. This study compares traditional ML and DL classifiers in the analysis of tabular data, in particular, functional connectivity measures obtained from the time series of a public multicenter dataset, and evaluates whether the features that contribute most to the classification task vary depending on the classifier used. Specifically, Support Vector Machine (SVM) classifiers, with both linear and radial basis function (RBF) kernels, and Extreme Gradient Boosting (XGBoost) classifiers are compared against the TabNet classifier (a DL architecture customized for tabular data analysis) and a Multi Layer Perceptron (MLP). The findings suggest that DL classifiers may not be optimal for the type of data analyzed, as their performance trails behind that of standard classifiers. Among the latter, SVMs outperform the other classifiers with an AUC of around 75%, whereas the best performances of TabNet and MLP reach 65% and 71% at most, respectively. Furthermore, the analysis of the feature importance showed that the brain regions that contribute the most to the classification task are those primarily responsible for sensory and spatial perception, as well as attention modulation, which is known to be altered in ASDs
- …
