A histopatologia é considerada um dos recursos diagnósticos mais importantes na prática médica e caracteriza-se pelo estudo das alterações estruturais e morfológicas das células e dos tecidos causadas por doenças. Atualmente, o principal método utilizado no diagnóstico histopatológico de imagens microscópicas, obtidas por meio de amostras em exames convencionais, é a avaliação visual do patologista, a qual se baseia na experiência do mesmo. O uso de técnicas de processamento computacional de imagens possibilita a identificação de elementos estruturais e a determinação de características inerentes, subsidiando o estudo da organização estrutural das células e de suas variações patológicas. A utilização de métodos computacionais no auxílio ao diagnóstico visa diminuir a subjetividade do processo de avaliação e classificação realizado pelo médico. Diferentes características dos tecidos podem ser mapeadas por meio de métricas específicas que poderão ser utilizadas em sistemas de reconhecimento de padrões. Dentro desta perspectiva, o objetivo geral deste trabalho inclui a proposta, a implementação e a avaliação de um método para a identificação e a análise de estruturas histológicas, a ser utilizado para a análise de lesões neoplásicas do colo do útero (NICs) a partir de amostras histopatológicas. Este trabalho foi desenvolvido em colaboração com uma equipe de patologistas, especialistas do domínio. As imagens microscópicas digitalizadas foram adquiridas a partir de lâminas previamente fixadas, contendo amostras de biópsias. Para segmentação dos núcleos celulares, foi implementado um pipeline de operadores morfológicos. Métodos de segmentação baseados em cor também foram testados e comparados à abordagem morfológica. Foi proposta e implementada uma abordagem baseada em camadas para representação do tecido, adotando-se a Triangulação de Delaunay (TD) como modelo de grafo de vizinhança. A TD apresenta algumas propriedades particulares que permitem a extração de métricas específicas. Foram utilizados algoritmos de agrupamento e morfologia de grafos, adotando-se critérios de semelhança e relações de adjacência entre os triângulos da rede, a fim de se obter a fronteira entre as camadas histológicas do tecido epitelial de forma automática. As seguintes métricas foram extraídas dos agrupamentos resultantes: grau médio, entropia e taxa de ocupação dos triângulos da rede. Finalmente, foi projetado um classificador estatístico levando-se em consideração os diferentes agrupamentos que poderiam ser obtidos a partir das imagens de treinamento. Valores de acurácia, sensitividade e especificidade foram utilizadas para avaliação dos resultados obtidos. Foi implementada validação cruzada em todos os experimentos realizados e foi utilizado um total de 116 imagens. Primeiro, foi avaliado a acurácia da metodologia proposta na determinação correta da presença de anomalia no tecido, para isto, todas as imagens que apresentavam NICs foram agrupadas em uma mesma classe. A maior taxa de acurácia obtida neste experimento foi de 88%. Em uma segunda etapa, foram realizadas avaliações entre as seguintes classes: Normal e NIC-I; NIC-I e NIC-II, e, NIC-II e NIC-III, obtendo-se taxas de acurácia máximas de 73%, 77% e 86%, respectivamente. Além disso, foi verificada também, a acurácia na discriminação entre os três tipos de NICs e regiões normais, obtendo-se acurácia de 64%. As taxas de ocupação relativas aos agrupamentos representativos das camadas basais e superficiais, foram os atributos que levaram às maiores taxas de acurácia. Os resultados obtidos permitem verificar a adequação do método proposto na representação e análise do processo de evolução das NICs no tecido epitelial do colo uterino. / Histopathology is considered one of the most important diagnostic tools in medical routine and is characterized by the study of structural and morphological changes of the cells in biological tissues caused by diseases. Currently, the visual assessment of the pathologist is the main method used in the histopathological diagnosis of microscopic images obtained from biopsy samples. This diagnosis is usually based on the experience of the pathologist. The use of computational techniques in the processing of these images allows the identification of structural elements and the determination of inherent characteristics, supporting the study of the structural organization of tissues and their pathological changes. Also, the use of computational methods to improve diagnosis aims to reduce the subjectivity of the evaluation made by the physician. Besides, different tissue characteristics can be mapped through specific metrics that can be used in pattern recognition systems. Within this perspective, the overall objective of this work includes the proposal, the implementation and the evaluation of a methodology for the identification and analysis of histological structures. This methodology includes the specification of a method for the analysis of cervical intraepithelial neoplasias (CINs) from histopathological samples. This work was developed in collaboration with a team of pathologists. Microscopic images were acquired from blades previously stained, containing samples of biopsy examinations. For the segmentation of cell nuclei, a pipeline of morphological operators were implemented. Segmentation techniques based on color were also tested and compared to the morphological approach. For the representation of the tissue architecture an approach based on the tissue layers was proposed and implemented adopting the Delaunay Triangulation (DT) as neighborhood graph. The DT has some special properties that allow the extraction of specific metrics. Clustering algorithms and graph morphology were used in order to automatically obtain the boundary between the histological layers of the epithelial tissue. For this purpose, similarity criteria and adjacency relations between the triangles of the network were explored. The following metrics were extracted from the resulting clusters: mean degree, entropy and the occupation rate of the clusters. Finally, a statistical classifier was designed taking into account the different combinations of clusters that could be obtained from the training process. Values of accuracy, sensitivity and specificity were used to evaluate the results. All the experiments were taken in a cross-validation process (5-fold) and a total of 116 images were used. First, it was evaluated the accuracy in determining the correct presence of abnormalities in the tissue. For this, all images presenting CINs were grouped in the same class. The highest accuracy rate obtained for this evaluation was 88%. In a second step, the discrimination between the following classes were analyzed: Normal/CIN 1; CIN 1/CIN 2, and, CIN 2/CIN 3, which represents the histological grading of the CINs. In a similar way, the highest accuracy rates obtained were 73%, 77% and 86%, respectively. In addition, it was also calculated the accuracy rate in discriminating between the four classes analyzed in this work: the three types of CINs and the normal region. In this last case, it was obtained a rate of 64%.The occupation rate for the basal and superficial layers were the attributes that led to the highest accuracy rates. The results obtained shows the adequacy of the proposed method in the representation and classification of the CINs evolution in the cervical epithelial tissue.

A evolução dos dispositivos de armazenamento e das redes de computadores permitiram que os vídeos digitais assumissem um importante papel no desenvolvimento de sistemas de informação multimídia. Com a finalidade de aproveitar todo o potencial dos vídeos digitais no desenvolvimento desses sistemas, técnicas automatizadas eficientes para análise, interpretação e recuperação são necessárias. A recuperação de vídeos baseada em conteúdo (CBVR, do inglês content-based video retrieval) permite o processamento e a análise do conteúdo de vídeos digitais visando à extração de informações relevantes que viabilizem indexação e recuperação. Trabalhos científicos têm proposto a aplicação de CBVR em bases de vídeos médicos a fim de proporcionar diferentes contribuições como diagnóstico auxiliado por computador, suporte à tomada de decisão e disponibilização de bases de vídeos para utilização em treinamento e educação médica. Em geral, características visuais são as principais informações utilizadas no contexto de CBVR aplicada em vídeos médicos. No entanto, muitos diagnósticos são realizados por meio da análise dos sons produzidos em diferentes estruturas e órgãos do corpo humano. Um exemplo é o diagnóstico cardíaco que, além de exames de imagem como ecocardiografia e ressonância magnética, também pode empregar a análise dos sons provenientes do coração por meio da auscultação. O objetivo deste trabalho consistiu em aplicar e avaliar extratores de características de som em conjunto com extratores de características visuais para viabilizar CBVR e, então, inferir se a abordagem resultou em ganhos com relação ao desempenho de recuperação quando comparada à utilização apenas das características visuais. Vídeos médicos constituíram nosso principal interesse, porém o trabalho considerou também vídeos não relacionados à área médica para a validação da abordagem. Justifica-se o objetivo, pois a análise do som, visando a obter descritores relevantes para melhorar os resultados de recuperação, ainda é pouco explorada na literatura científica. Essa afirmação foi evidenciada com a condução de uma revisão sistemática sobre o tema. Dois conjuntos de experimentos foram conduzidos visando a validar a abordagem de CBVR mencionada. O primeiro conjunto de experimentos foi aplicado sobre uma base de vídeos sintetizados para validação da abordagem. Já o segundo, foi aplicado em uma base de vídeos construídos utilizando-se imagens provenientes de exames de ressonância magnética em conjunto com sons provenientes de auscultação do coração. Os resultados foram analisados utilizando-se as métricas de revocação e precisão, bem como o gráfico que as relaciona. Demonstrou-se que a abordagem é promissora por meio da melhora significativa dos resultados de recuperação nos diferentes cenários de combinação entre características visuais e sonoras experimentados / Advance of storage devices and computer networks has contributed to digital videos assume an important role in the development of multimedia information systems. In order to take advantage of the full potential of digital videos in the development of these systems, it is necessary the development of efficient techniques for automated data analysis, interpretation and retrieval. Content-based video retrieval (CBVR) allows processing and analysis of content in digital videos to extract relevant information and enable indexing and retrieval. Scientific studies have proposed the application of CBVR in medical video databases in order to provide different contributions like computer-aided diagnosis, decision-making support or availability of video databases for use in medical training and education. In general, visual characteristics are the main information used in the context of CBVR applied in medical videos. However, many diagnoses are performed by analysing the sounds produced in different structures and organs of the human body. An example is the cardiac diagnosis which, in addition to images generated by echocardiography and magnetic resonance imaging, for example, may also employ the analysis of sounds from the heart by means of auscultation. The objective of this work was evaluating combination between audio signal and visual features to enable CBVR and investigating how much this approach can improve retrieval results comparing to using only visual features. Medical videos are the main data of interest in this work, but video segments not related to the medical field were also used to validate the approach. The objectives of this work are justifiable because audio signal analysis, in order to get relevant descriptors to improve retrieval results, is still little explored in the scientific literature. This statement was evidenced by results of a systematic review. Two experiment sets were conducted to validate the CBVR approach described. The first experiment set was applied to a synthetic images database specially built to validate the approach, while the second experiment was applied to a database composed of digital videos created from magnetic resonance imaging and heart sounds from auscultation. Results were analyzed using the recall and precision metrics, as well as the graph which relates these metrics. Results showed that this approach is promising due the significantly improvement obtained in retrieval results to different scenarios of combination between visual and audio signal features

<p>Mammographic screening is an effective way to detect breast cancer. Early detection of breast cancer depends to a high degree on the adequacy of the mammogram from which the diagnosis is made. Today, most of the analysis of the mammogram is performed by radiologists. Computer-aided diagnosis (CAD) systems have been proposed as an aid to increase the efficiency and effectiveness of the screening procedure by automatically indicating abnormalities in the mammograms. However, in order for a CAD system to be stable and efficient, the input images need to be adequate. Criteria for adequacy include: high resolution, low image noise and high image contrast. Additionally, the breast needs to be adequately positioned and compressed to properly visualise the entire breast and especially the glandular tissue.</p><p>This thesis addresses questions regarding the automatic determination of mammogram adequacy with the focus on breast positioning and segmentation evaluation. The goal and, thus, the major technical challenge is to develop algorithms that support fully automatic quality checks. The relevant quality criteria are discussed in Chapter 2. The aim of this discussion is to compile a comprehensive list of necessary criteria that a system for automatic assessment of mammographic adequacy must satisfy. Chapter 3 gives an overview of research performed in computer-aided analysis of mammograms. It also provides basic knowledge about image analysis involved in the research area of computerized mammography in general, and in the papers of this thesis, in particular. In contrast, Chapter 4 describes basic knowledge about segmentation evaluation, which is a highly important component in image analysis. Papers I–IV propose algorithms for measuring the quality of a mammogram according to certain criteria and addresses problems related to them. A method for automatic analysis of the shape of the pectoralis muscle is presented in Paper I. Paper II proposes a fully automatic method for extracting the breast border. A geometric assumption used by radiologists is that the nipple is located at the point on the breast border being furthest away from the pectoralis muscle. This assumption is investigated in Paper III, and a method for automatically restricting the search area is proposed. There has been an increasing need to develop an automated segmentation algorithm for extracting the glandular tissue, where the majority of breast cancer occur. In Paper IV, a novel approach for solving this problem in a robust and accurate way is proposed. Paper V discusses the challenges involved in evaluating the quality of segmentation algorithms based on ground truths provided by an expert panel. A method to relate ground truths provided by several experts to each other in order to establish levels of agreement is proposed. Furthermore, this work is used to develop an algorithm that combines an ensemble of markings into one surrogate ground truth.</p>

Mammographic screening is an effective way to detect breast cancer. Early detection of breast cancer depends to a high degree on the adequacy of the mammogram from which the diagnosis is made. Today, most of the analysis of the mammogram is performed by radiologists. Computer-aided diagnosis (CAD) systems have been proposed as an aid to increase the efficiency and effectiveness of the screening procedure by automatically indicating abnormalities in the mammograms. However, in order for a CAD system to be stable and efficient, the input images need to be adequate. Criteria for adequacy include: high resolution, low image noise and high image contrast. Additionally, the breast needs to be adequately positioned and compressed to properly visualise the entire breast and especially the glandular tissue. This thesis addresses questions regarding the automatic determination of mammogram adequacy with the focus on breast positioning and segmentation evaluation. The goal and, thus, the major technical challenge is to develop algorithms that support fully automatic quality checks. The relevant quality criteria are discussed in Chapter 2. The aim of this discussion is to compile a comprehensive list of necessary criteria that a system for automatic assessment of mammographic adequacy must satisfy. Chapter 3 gives an overview of research performed in computer-aided analysis of mammograms. It also provides basic knowledge about image analysis involved in the research area of computerized mammography in general, and in the papers of this thesis, in particular. In contrast, Chapter 4 describes basic knowledge about segmentation evaluation, which is a highly important component in image analysis. Papers I–IV propose algorithms for measuring the quality of a mammogram according to certain criteria and addresses problems related to them. A method for automatic analysis of the shape of the pectoralis muscle is presented in Paper I. Paper II proposes a fully automatic method for extracting the breast border. A geometric assumption used by radiologists is that the nipple is located at the point on the breast border being furthest away from the pectoralis muscle. This assumption is investigated in Paper III, and a method for automatically restricting the search area is proposed. There has been an increasing need to develop an automated segmentation algorithm for extracting the glandular tissue, where the majority of breast cancer occur. In Paper IV, a novel approach for solving this problem in a robust and accurate way is proposed. Paper V discusses the challenges involved in evaluating the quality of segmentation algorithms based on ground truths provided by an expert panel. A method to relate ground truths provided by several experts to each other in order to establish levels of agreement is proposed. Furthermore, this work is used to develop an algorithm that combines an ensemble of markings into one surrogate ground truth.

Vascular diseases are among the leading causes of death in Canada and around the globe. A major underlying cause of most such medical conditions is atherosclerosis, a gradual accumulation of plaque on the walls of blood vessels. Particularly vulnerable to atherosclerosis is the carotid artery, which carries blood to the brain. Dangerous narrowing of the carotid artery can lead to embolism, a dislodgement of plaque fragments which travel to the brain and are the cause of most strokes. If this pathology can be detected early, such a deadly scenario can be potentially prevented through treatment or surgery. This not only improves the patient's prognosis, but also dramatically lowers the overall cost of their treatment. Medical imaging is an indispensable tool for early detection of atherosclerosis, in particular since the exact location and shape of the plaque need to be known for accurate diagnosis. This can be achieved by locating the plaque inside the artery and measuring its volume or texture, a process which is greatly aided by image segmentation. In particular, the use of ultrasound imaging is desirable because it is a cost-effective and safe modality. However, ultrasonic images depict sound-reflecting properties of tissue, and thus suffer from a number of unique artifacts not present in other medical images, such as acoustic shadowing, speckle noise and discontinuous tissue boundaries. A robust ultrasound image segmentation technique must take these properties into account. Prior to segmentation, an important pre-processing step is the extraction of a series of features from the image via application of various transforms and non-linear filters. A number of such features are explored and evaluated, many of them resulting in piecewise smooth images. It is also proposed to decompose the ultrasound image into several statistically distinct components. These components can be then used as features directly, or other features can be obtained from them instead of the original image. The decomposition scheme is derived using Maximum-a-Posteriori estimation framework and is efficiently computable. Furthermore, this work presents and evaluates an algorithm for segmenting the carotid artery in 3D ultrasound images from other tissues. The algorithm incorporates information from different sources using an energy minimization framework. Using the ultrasound image itself, statistical differences between the region of interest and its background are exploited, and maximal overlap with strong image edges encouraged. In order to aid the convergence to anatomically accurate shapes, as well as to deal with the above-mentioned artifacts, prior knowledge is incorporated into the algorithm by using weak geometric priors. The performance of the algorithm is tested on a number of available 3D images, and encouraging results are obtained and discussed.

This Master Thesis presents a novel classification approach for microcalcifications (MCs) extracted from core biopsy tissue samples digitized using micro-CT, a high-resolution 3D imaging modality. MCs are tiny spots of calcium that may occur in the female breast. Although they are common in healthy woman, they are often an early sign of breast cancer. The shape of the MCs is an important factor used to discriminate between benign and malignant abnormalities. However, the current standard imaging modalities (i.e. mammography) are not efficient for a clear shape based analysis. In case of suspiciousness, a biopsy is conducted and the extracted tissue is anatomopathologically investigated for the presence of cancer cells. Nevertheless, only 20-35% of biopsies turn out to be positive. As such, the question whether some unnecessary biopsies can be avoided if the shape of the MCs could be analysed in more detail has been raised. In addition, the MCs themselves are not analysed, but they are characterised as benign (or malignant) according to whether they were found into a benign (or malignant) tissue. As a result, there is a ground truth for the tissue samples but not for the individual MCs. So, when a classifier of a Computer Aided Diagnosis System will be asked to classify a MC according to its shape, there will be a degree of ambiguity and uncertainty. This master thesis investigates whether the use of a clustering method as a preprocessing step before training the classifier could avoid the ground truth issues and could improve the obtained classification results. / Η παρούσα μεταπτυχιακή εργασία παρουσιάζει μια νέα μέθοδο για την ταξινόμηση μικροαποτιτανώσεων μαστού που έχουν εξαχθεί από βιοψίες και έχουν ψηφιοποιηθεί με χρήση micro-CT, μια υψηλής ανάλυσης, τρισδιάστατη τεχνική απεικόνισης. Οι μικροαποτιτανώσεις (ή αλλιώς μικροασβεστώσεις) αποτελούν μικρά αποθέματα ασβεστίου στον μαστικό αδένα. Παρόλο που μπορεί να εμφανιστούν και σε υγιείς γυναίκες, μπορούν να αποτελέσουν ένα πρώιμο σημάδι καρκίνου του στήθους. Το σχήμα είναι ένας από τους σημαντικότερους παράγοντες ο οποίος βοηθάει στη διάκριση ανάμεσα σε καλοήθεις και κακοήθεις μικροασβεστώσεις, ωστόσο δεν μπορεί να απεικονιστεί επαρκώς μέσω των στανταρ απεικονιστικών τεχνικών (μαστογραφία). Σε περίπτωση υποψίας κακοήθειας, διεξάγεται βιοψία με σκοπό την απομάκρυνση ιστού από την ύποπτη περιοχή και την ανατομοπαθολογική του εξέταση για την παρουσία καρκινικών κυττάρων. Ωστόσο, μονο το 20%-35% των βιοψιών αποδεικνύονται κακοήθεις. Ως εκ τούτου, έχει τεθεί το ερώτημα κατά πόσο μπορούν να αποφευχθούν οι μη απαραίτητες βιοψίες εάν το σχήμα των μικροασβεστώσεων μπορούσε να μελετηθεί πιο λεπτομερώς. Επιπροσθέτως, οι μικροασβεστώσεις αυτές καθ’ εαυτές δεν αναλύονται αλλά χαρακτηρίζονται ως καλοήθεις (ή κακοήθεις) με βάση το αν βρέθηκαν μεσα σε καλοήθες (ή κακοήθες) ιστό. Ως αποτέλεσμα, υπάρχει βάση αναφοράς για τα δείγματα ιστού αλλά όχι για τις μικροασβεστώσεις. Έτσι, όταν ζητηθεί από έναν ταξινομητή ενός συστήματος υποβοηθούμενης διάγνωσης με υπολογιστή να ταξινομήσει μικροασβεστώσεις με βάση το σχήμα τους υπάρχει ένα μεγάλο ποσοστό ασάφειας και αβεβαιότητας. Αυτή η μεταπτυχιακή εργασία έχει σκοπό να ερευνήσει εάν η εισαγωγή ενός βήματος συσταδοποίησης πριν αυτού της ταξινόμησης μπορεί να αποφύγει το πρόβλημα έλλειψης βάσης αναφοράς και να βελτιώσει τα αποτελέσματα της ταξινόμησης.

Vascular diseases are among the leading causes of death in Canada and around the globe. A major underlying cause of most such medical conditions is atherosclerosis, a gradual accumulation of plaque on the walls of blood vessels. Particularly vulnerable to atherosclerosis is the carotid artery, which carries blood to the brain. Dangerous narrowing of the carotid artery can lead to embolism, a dislodgement of plaque fragments which travel to the brain and are the cause of most strokes. If this pathology can be detected early, such a deadly scenario can be potentially prevented through treatment or surgery. This not only improves the patient's prognosis, but also dramatically lowers the overall cost of their treatment. Medical imaging is an indispensable tool for early detection of atherosclerosis, in particular since the exact location and shape of the plaque need to be known for accurate diagnosis. This can be achieved by locating the plaque inside the artery and measuring its volume or texture, a process which is greatly aided by image segmentation. In particular, the use of ultrasound imaging is desirable because it is a cost-effective and safe modality. However, ultrasonic images depict sound-reflecting properties of tissue, and thus suffer from a number of unique artifacts not present in other medical images, such as acoustic shadowing, speckle noise and discontinuous tissue boundaries. A robust ultrasound image segmentation technique must take these properties into account. Prior to segmentation, an important pre-processing step is the extraction of a series of features from the image via application of various transforms and non-linear filters. A number of such features are explored and evaluated, many of them resulting in piecewise smooth images. It is also proposed to decompose the ultrasound image into several statistically distinct components. These components can be then used as features directly, or other features can be obtained from them instead of the original image. The decomposition scheme is derived using Maximum-a-Posteriori estimation framework and is efficiently computable. Furthermore, this work presents and evaluates an algorithm for segmenting the carotid artery in 3D ultrasound images from other tissues. The algorithm incorporates information from different sources using an energy minimization framework. Using the ultrasound image itself, statistical differences between the region of interest and its background are exploited, and maximal overlap with strong image edges encouraged. In order to aid the convergence to anatomically accurate shapes, as well as to deal with the above-mentioned artifacts, prior knowledge is incorporated into the algorithm by using weak geometric priors. The performance of the algorithm is tested on a number of available 3D images, and encouraging results are obtained and discussed.

Nowadays,hospitals are equipped with high resolution medical imaging systems such as MRI, CT that help the radiologists to make more accurate diagnosis. However these systems cannot give any information of the explicit content that is on the image pixels. The vast amount of images that are produced in hospitals is processed mainly by the medical domain users. Even systems such as PACS cannot retrieve images with anatomical or disease-­‐related criteria. The integrating of semantic web technologies in health care can provide a solution. The benefits for the semantic web technologies are owed to the core element of the semantic web, which is the ontology. The ontology sets strict relationships between its entities. The main goal of this thesis is to design and develop an online approach for Semantic Annotation and Retrieval of Medical Images. The architecture of the proposed system is based on a service oriented approach that enables the expandability of the system by integrating new features such as image processing algorithms to perform Computer Aided Diagnosis (CAD) tasks and to make queries with low -­‐ level image characteristics. Also the adopting of such an approach for the architecture allows to add new reference ontologies to the system without redesigning the core architecture. The ontology framework of the system includes (a) three reference ontologies, namely the Foundational Model of Anatomy (FMA) for the anatomy annotation, the International Classification of Disease (ICD-­‐10) for the disease annotation and the RadLex for the radiological findings and (b) an application ontology that connects the medical document with the concepts of the medical ontologies (FMA, ICD-­‐10, Radlex) and it also contains information about patient, hospital and image modality. Part of application ontology information is extracted from the DICOM header. In the context of the current thesis, the system was used to annotate and retrieve several medical images. The proposed online approach for annotation and retrieval of medical images system can enable the interoperability between different Health Information Systems (HIS) and can constitute a tool for discovering the hidden knowledge in medical image data. / -

Este trabalho apresenta um estudo de classificação de nódulos em mamograma digitalizados através de um classificador de rede neural artificial (RNA). O algoritmo de treinamento de \"backpropagation\" foi utilizado para ajustar os pesos da RNA. O objetivo principal deste trabalho foi determinar um método para analisar e selecionar a melhor configuração de atributos e topologia da RNA para classificar lesões mamárias do tipo nódulo. Foram escolhidas 118 imagens de regiões de interesse (ROI), sendo 68 benignas e 50 malignas de duas bases de imagens: uma do Hospital das Clínicas de Ribeirão Preto, da Universidade de São Paulo, e outra do MIAS-UK (Mammographic Image Analysis Society). O processo completo envolveu quatro etapas: detecção, extração e seleção de atributos, e classificação. Na etapa de detecção, as imagens foram submetidas ao processo combinado das técnicas segmentação de thresholding, morfologia matemática e crescimento de região. Foram extraídos 14 atributos de textura e 14 atributos de forma. Para selecionar os atributos mais discriminantes, foi utilizado o método de Jeffries-Matusita. Foram selecionados três grupos de atributos de forma, três de atributos de textura e três de atributos combinados. Análise pela curva ROC foram dirigidas para avaliar o desempenho do classificador de rede neural artificial (RNA). Os melhores resultados obtidos foram: para o grupo de atributos de forma com 5 unidades escondidas, a área dentro da curva ROC foi de 0.99, taxa de acerto de 98,21%, taxa de especificidade de 98,37% e taxa de sensibilidade de 98.00%; para o grupo de atributos de textura com 4 unidades escondidas, a área dentro da curva foi de 0.98, taxa de acerto de 97,08%, taxa de especificidade de 98,53% e taxa de sensibilidade de 95.11%; para o grupo de atributos combinados de textura e forma com 3 unidades escondidas, a área dentro da curva foi de 0.99, taxa de acerto de 98,21%, taxa de especificidade de 100.00% e taxa de sensibilidade de 95.78%. / This work presents a study of masses classification in digitized mammograms by means of artificial neural network (ANN). The backpropagation training algorithm was used to adjust the weights of ANN. The aim of this work was to determine a methodology to analyze and selection of the best feature subset and ANN topology to classify masses lesions. A total of 118 regions of interest images were chosen (68 benign and 50 malignant lesions) from two image databases: one from \"Hospital das Clínicas de Ribeirão Preto\", at the University of São Paulo, and other from Mammographic lmage Analysis Society (MIAS-UK). The whole process involved four steps: segmentation, feature extraction, selection, and classification. In the first step, the images were submitted to a combined process of thresholding, mathematical morphology, and region growing techniques. In the second step, fourteen texture features and fourteen shape features were extracted. The Jeffries-Matusita method was used to select the best features. The results of this stage were the selection of three shape feature sets, three texture feature sets, and three combined feature sets. The Receiver Operating Characteristic (ROC) analysis were conducted to evaluated the ANN classifier performance. The best result obtained for shape feature set was obtained using a ANN with 5 hidden units, the area under ROC curve was of 0.99, classification rate of 98.21%, specificity rate of 98.37% and sensitivity rate of 98.00%. For texture feature set, the best result was using a ANN with 4 hidden units, the area under ROC curve was of 0.98, classification rate of 97.08%, specificity rate of 98.53% and sensitivity rate of 95.11%. Finally, for the combined feature set (texture and shape) the best result obtained was using a ANN with 3 hidden units, the area under ROC curve was of 0.99, classification rate of 98.21%, specificity rate of 100.00% and sensitivity rate of 95.78%.

Neste trabalho foi desenvolvida uma metodologia de avaliação e monitoramento de pacientes com úlceras de perna baseada nas características dos tecidos internos dessas feridas. Os tecidos internos podem ser classificados como granulado, fibrina e necrosado, e a avaliação da área de cada um desses tecidos fornece para o clínico dados referentes ao estado da úlcera.A metodologia extrai essas informações a partir de imagens digitalizadas das lesões. Para tanto, a área referente à úlcera é segmentada e em seguida a área interna processada por uma rede neural, que tem o propósito de classificar cada ponto para um dos tecidos analisados. Os algoritmos desenvolvidos operam sobre imagens coloridas, já que cada tecido em uma imagem só pode ser identificado por sua cor. Este trabalho propõe ainda uma metodologia de extração de características das lesões através de uma forma não invasiva utilizando, para tanto, algoritmos de visão computacional. / The aim of this work was the development of a monitoring and evaluation methodology of leg ulcers patients based on the features of the inner tissues of these wounds. The internal tissues can be classified as granulation, slough and necrotic, and the evaluation of the area of each one of these tissues can be used by the specialist to help with the patient''s diagnosis. The methodology extracts these information from the wound digitized images. For this, the wound area is segmented and the inner region or the segmented area is processed by a neural network that classifies each point of the analyzed tissues. The developed algorithms operate on color images since each tissue in an image can only be analyzed by its colors. In this work has also proposed a feature extraction methodology of the wounds through a non-invasive way using computer vision algorithms.