Practical insights into digital radiology

Digital radiography detectors—based on different technological solutions—are currently available for clinical applications and widespread in clinical practice. Computed radiography (CR) and digital radiology systems have been available for clinical applications and the trend over the last few years has become digital. Radiology departments have been changing from traditional screen–film technology to digital technology. This chapter is intended to give the reader a practical understanding about the key aspects concerning digital systems, related to the performance of different technologies, image quality, and dose and patient safety/protection. The discussion around an optimization framework for digital systems is provided.

Image enhancement for digital radiography

Once in a digital form, a radiographic image may be processed in several ways in order to turn the visualization an act of improved diagnostic value. Practitioners should be aware that, depending on each clinical context, digital image processing techniques are available to help to unveil visual information that is, in fact, carried by the bare digital radiograph and may be otherwise neglected. The range of visual enhancement procedures includes simple techniques that deal with the usual brightness and contrast manipulation up to much more elaborate multi-scale processing that provides customized control over the emphasis given to the relevant finer anatomical details. This chapter is intended to give the reader a practical understanding of image enhancement techniques that might be helpful to improve the visual quality of the digital radiographs and thus to contribute to a more reliable and assertive reporting.

Digital radiology and Picture Archiving and Communication System (PACS)

Plain radiography still accounts for the vast majority of imaging studies that are performed at multiple clinical instances. Digital detectors are now prominent in many imaging facilities and they are the main driving force towards filmless environments. There has been a working paradigm shift due to the functional separation of acquisition, visualization, and storage with deep impact in the imaging workflows. Moreover with direct digital detectors images are made available almost immediately. Digital radiology is now completely integrated in Picture Archiving and Communication System (PACS) environments governed by the Digital Imaging and Communications in Medicine (DICOM) standard. In this chapter a brief overview of PACS architectures and components is presented together with a necessarily brief account of the DICOM standard. Special focus is given to the DICOM digital radiology objects and how specific attributes may now be used to improve and increase the metadata repository associated with image data. Regular scrutiny of the metadata repository may serve as a valuable tool for improved, cost-effective, and multidimensional quality control procedures.

Vertical CNT-Si Photodiode Array

Agências financiadoras: National Natural Science Foundation of China - 61204077; Shenzhen Science and Technology Innovation Commission - JCYJ20120614150521967

PINPIN a-Si: H based structures for X-ray image detection using the laser scanning technique

Conventional film based X-ray imaging systems are being replaced by their digital equivalents. Different approaches are being followed by considering direct or indirect conversion, with the later technique dominating. The typical, indirect conversion, X-ray panel detector uses a phosphor for X-ray conversion coupled to a large area array of amorphous silicon based optical sensors and a couple of switching thin film transistors (TFT). The pixel information can then be readout by switching the correspondent line and column transistors, routing the signal to an external amplifier. In this work we follow an alternative approach, where the electrical switching performed by the TFT is replaced by optical scanning using a low power laser beam and a sensing/switching PINPIN structure, thus resulting in a simpler device. The optically active device is a PINPIN array, sharing both front and back electrical contacts, deposited over a glass substrate. During X-ray exposure, each sensing side photodiode collects photons generated by the scintillator screen (560 nm), charging its internal capacitance. Subsequently a laser beam (445 nm) scans the switching diodes (back side) retrieving the stored charge in a sequential way, reconstructing the image. In this paper we present recent work on the optoelectronic characterization of the PINPIN structure to be incorporated in the X-ray image sensor. The results from the optoelectronic characterization of the device and the dependence on scanning beam parameters are presented and discussed. Preliminary results of line scans are also presented. (C) 2014 Elsevier B.V. All rights reserved.

Development of human central nervous system 3D in vitro models for preclinical research

Neurological disorders are a major concern in modern societies, with increasing prevalence mainly related with the higher life expectancy. Most of the current available therapeutic options can only control and ameliorate the patients’ symptoms, often be-coming refractory over time. Therapeutic breakthroughs and advances have been hampered by the lack of accurate central nervous system (CNS) models. The develop-ment of these models allows the study of the disease onset/progression mechanisms and the preclinical evaluation of novel therapeutics. This has traditionally relied on genetically engineered animal models that often diverge considerably from the human phenotype (developmentally, anatomically and physiologically) and 2D in vitro cell models, which fail to recapitulate the characteristics of the target tissue (cell-cell and cell-matrix interactions, cell polarity). The in vitro recapitulation of CNS phenotypic and functional features requires the implementation of advanced culture strategies that enable to mimic the in vivo struc-tural and molecular complexity. Models based on differentiation of human neural stem cells (hNSC) in 3D cultures have great potential as complementary tools in preclinical research, bridging the gap between human clinical studies and animal models. This thesis aimed at the development of novel human 3D in vitro CNS models by integrat-ing agitation-based culture systems and a wide array of characterization tools. Neural differentiation of hNSC as 3D neurospheres was explored in Chapter 2. Here, it was demonstrated that human midbrain-derived neural progenitor cells from fetal origin (hmNPC) can generate complex tissue-like structures containing functional dopaminergic neurons, as well as astrocytes and oligodendrocytes. Chapter 3 focused on the development of cellular characterization assays for cell aggregates based on light-sheet fluorescence imaging systems, which resulted in increased spatial resolu-tion both for fixed samples or live imaging. The applicability of the developed human 3D cell model for preclinical research was explored in Chapter 4, evaluating the poten-tial of a viral vector candidate for gene therapy. The efficacy and safety of helper-dependent CAV-2 (hd-CAV-2) for gene delivery in human neurons was evaluated, demonstrating increased neuronal tropism, efficient transgene expression and minimal toxicity. The potential of human 3D in vitro CNS models to mimic brain functions was further addressed in Chapter 5. Exploring the use of 13C-labeled substrates and Nucle-ar Magnetic Resonance (NMR) spectroscopy tools, neural metabolic signatures were evaluated showing lineage-specific metabolic specialization and establishment of neu-ron-astrocytic shuttles upon differentiation. Chapter 6 focused on transferring the knowledge and strategies described in the previous chapters for the implementation of a scalable and robust process for the 3D differentiation of hNSC derived from human induced pluripotent stem cells (hiPSC). Here, software-controlled perfusion stirred-tank bioreactors were used as technological system to sustain cell aggregation and dif-ferentiation. The work developed in this thesis provides practical and versatile new in vitro ap-proaches to model the human brain. Furthermore, the culture strategies described herein can be further extended to other sources of neural phenotypes, including pa-tient-derived hiPSC. The combination of this 3D culture strategy with the implemented characterization methods represents a powerful complementary tool applicable in the drug discovery, toxicology and disease modeling.

Towards Detecting Changes in Underwater Image Sequences

Detecting changes between images of the same scene taken at different times is of great interest for monitoring and understanding the environment. It is widely used for on-land application but suffers from different constraints. Unfortunately, Change detection algorithms require highly accurate geometric and photometric registration. This requirement has precluded their use in underwater imagery in the past. In this paper, the change detection techniques available nowadays for on-land application were analyzed and a method to automatically detect the changes in sequences of underwater images is proposed. Target application scenarios are habitat restoration sites, or area monitoring after sudden impacts from hurricanes or ship groundings. The method is based on the creation of a 3D terrain model from one image sequence over an area of interest. This model allows for synthesizing textured views that correspond to the same viewpoints of a second image sequence. The generated views are photometrically matched and corrected against the corresponding frames from the second sequence. Standard change detection techniques are then applied to find areas of difference. Additionally, the paper shows that it is possible to detect false positives, resulting from non-rigid objects, by applying the same change detection method to the first sequence exclusively. The developed method was able to correctly find the changes between two challenging sequences of images from a coral reef taken one year apart and acquired with two different cameras

Large-Area Photo-Mosaics Using Global Alignment and Navigation Data

Seafloor imagery is a rich source of data for the study of biological and geological processes. Among several applications, still images of the ocean floor can be used to build image composites referred to as photo-mosaics. Photo-mosaics provide a wide-area visual representation of the benthos, and enable applications as diverse as geological surveys, mapping and detection of temporal changes in the morphology of biodiversity. We present an approach for creating globally aligned photo-mosaics using 3D position estimates provided by navigation sensors available in deep water surveys. Without image registration, such navigation data does not provide enough accuracy to produce useful composite images. Results from a challenging data set of the Lucky Strike vent field at the Mid Atlantic Ridge are reported

Automatic classification of breast density

A recent trend in digital mammography is computer-aided diagnosis systems, which are computerised tools designed to assist radiologists. Most of these systems are used for the automatic detection of abnormalities. However, recent studies have shown that their sensitivity is significantly decreased as the density of the breast increases. This dependence is method specific. In this paper we propose a new approach to the classification of mammographic images according to their breast parenchymal density. Our classification uses information extracted from segmentation results and is based on the underlying breast tissue texture. Classification performance was based on a large set of digitised mammograms. Evaluation involves different classifiers and uses a leave-one-out methodology. Results demonstrate the feasibility of estimating breast density using image processing and analysis techniques

Medical image registration based on random line sampling

One of the key aspects in 3D-image registration is the computation of the joint intensity histogram. We propose a new approach to compute this histogram using uniformly distributed random lines to sample stochastically the overlapping volume between two 3D-images. The intensity values are captured from the lines at evenly spaced positions, taking an initial random offset different for each line. This method provides us with an accurate, robust and fast mutual information-based registration. The interpolation effects are drastically reduced, due to the stochastic nature of the line generation, and the alignment process is also accelerated. The results obtained show a better performance of the introduced method than the classic computation of the joint histogram

Mòdul de visualització 3D d’ ajuda al diagnòstic de les neoplàsies rectals

El diagnòstic mitjançant la imatge mèdica s’ha convertit en una eina fonamental en la pràctica clínica, permet entre altres coses, reconstruir a partir d’un conjunt d’imatges 2D, obtingudes a partir d’aparells de captació, qualsevol part de l’organisme d’un pacient i representar-lo en un model 3D. Sobre aquest model 3D poden realitzar-se diferents operacions que faciliten el diagnòstic i la presa de decisions als especialistes. El projecte que es presenta forma part del desenvolupament de la plataforma informàtica de visualització i tractament de dades mèdiques, anomenada Starviewer, que desenvolupen conjuntament el laboratori de Gràfics i Imatge (GiLab) de la Universitat de Girona i l’ Institut de Diagnòstic per la Imatge (IDI) de l’Hospital Josep Trueta de Girona. En particular, en aquest projecte es centra en el diagnòstic del càncer colorectal i el desenvolupament de mètodes i tècniques de suport al seu diagnòstic. Els dos punts claus en el tractament d’aqueta patologia són: la detecció de les lesions I l’estudi de l’evolució d’aquestes lesions, una vegada s’ha iniciat el tractament tumoral. L’objectiu principal d’aquest projecte és implementar i integrar en la plataforma Starviewer les tècniques de visualització i processament de dades necessàries per donar suport als especialistes en el diagnòstic de les lesions del colon. Donada la dificultat en el processament de les dades reals del budell ens proposem: dissenyar i implementar un sistema per crear models sintètics del budell; estudiar, implementar i avaluar les tècniques de processament d’imatge que calen per segmentar lesions de budell; dissenyar i implementar un sistema d’exploració del budell iintegrar de tots els mòduls implementats en la plataforma starviewer

Visualització de models fusionats

L’objectiu principal d’aquest projecte era implementar la visualització 3D demodels fusionats i aplicar totes les tècniques possibles per realitzar aquesta fusió. Aquestes tècniques s’integraran en la plataforma de visualització i processament de dades mèdiques STARVIEWER. Per assolir l’ objectiu principal s’ han definit els següents objectius específics:1- estudiar els algoritmes de visualització de models simples i analitzar els diferents paràmetres a tenir en compte. 2- ampliació de la tècnica de visualització bàsica seleccionada per tal de suportar els models fusionats. 3- avaluar i compar tots els mètodes implementats per poder determinar quin ofereix les millors visualitzacions

Integració d’un visualitzador de volums en una plataforma de reconstrucció de mapes de fibres del cervell

Estudi, disseny i implementació d’un algorisme de visualització de volums i integrar-lo en la plataforma DTIWeb de visualització i processament de dades de DTI. La plataforma DTIWeb és una plataforma desenvolupada conjuntament entre el Laboratori de Gràfics i Imatge de la Universitat de Girona i d’Institut de Diagnòstic per la imatge de l’Hospital Josep Trueta de Girona. Aquesta plataforma integra els mètodes bàsics de reconstrucció de fibres del cervell. La principal limitació de la plataforma és que no suporta la visualització de models 3D. Aquest fet limita el seu us en la pràctica clínica habitual ja que es fa difícil la interpretació dels mapes de connectivitat que genera

Caracterització i diagnosi de masses en imatges de mamografia

Els objectius del projecte es divideixen en tres blocs: Primerament, realitzar unasegmentació automàtica del contorn d'una imatge on hi ha una massa central. Tot seguit, a partir del contorn trobat, caracteritzar la massa. I finalment, utilitzant les característiques anteriors classificar la massa en benigne o maligne. En el projecte s'utilitza el Matlab com a eina de programació. Concretament les funcions enfocades al processat de imatges del toolbox de Image processing (propi de Matlab) i els classificadors de la PRTools de la Delft University of Technology

Entorn de suport a la inserció de pròtesis

L’objectiu d’aquest projecte és integrar a la plataforma Starviewer ( plataforma informàtica de processament i visualització d’imatges mèdiques creada fruit de la col•laboració del Laboratori de Gràfics i Imatge (GILab) de la Universitat de Girona i l’Institut de Diagnòstic per la Imatge (IDI) de l’hospital Dr. Josep Trueta de Girona) per donar suport al diagnòstic un entorn de suport a la inserció de pròtesis, que permeti automatitzar al màxim les operacions que actualment es realitzen de forma manual. Hem de tenir en compte que, tot i que, la imatge més usada pel radiòleg es la radiografia (Rx) també treballa amb tomografia computada (TAC). El TAC dona una visió 3D de l’organisme, mentre que la Rx és 2D