Le diagnostic assisté par ordinateur du cancer du sein suscite de plus en plus un réel engouement en raison de la quantité sans cesse croissante d'images mammographiques issues des campagnes de dépistage. L'accent est mis sur les opacités mammaires en raison du risque élevé de cancer qui leur est associé. En effet, la variabilité des formes rencontrées et la difficulté à discerner les masses surtout quand ces dernières sont embarquées dans des densités importantes exigent une nouvelle stratégie plutôt adaptée aux cas les plus complexes à savoir les masses appartenant aux classes BI-RADS IV et V, c-à-d. respectivement les masses malignes spiculées et les distorsions architecturales. Dans ce travail, un système de diagnostic assisté par ordinateur entièrement automatique et conçu pour la segmentation et la classification des opacités dans les catégories bénigne/maligne ou graisseuse/dense, spécifiquement pour celles de type BI-RADS IV et V est abordé. Dans un premier temps, nous avons développé une approche de pré-traitement des images fondée sur l'apprentissage d'un dictionnaire parcimonieux sur les bases d'images, combiné à une réduction de dimension afin de supprimer de façon efficace et rapide le bruit de numérisation des images mammographiques présentes dans les bases utilisées pour concevoir notre système de diagnostic en comparaison des approches déjà existantes. Une fois les images pré-traitées, nous avons mis en place une procédure de segmentation non-supervisée des masses basée sur les champs de Markov et qui a l'avantage d'être à la fois plus rapide, plus efficace et plus robuste que les meilleures techniques de segmentation disponibles dans l'état-de-l'art. De plus, la méthode proposée s'affranchit de la variabilité des masses et ce quelque soit la densité de l'image. Dans l'idée de décrire convenablement les lésions malignes spiculées, nous avons conçu une méthode de segmentation des spicules qui présente la particularité de ne pas recourir à l'utilisation de descripteurs extraits manuellement dont les performances peuvent varier en fonction de leur qualité. L'approche proposée repose sur des hypothèses que nous avons formulées concernant l'aspect des spicules. Celles-ci nous ont conduits à développer un modèle Markovien combiné à une transformée de Radon locale pour extraire les structures curvilignes de l'image. Ensuite, nous servant d'un modèle a contrario, nous avons pu extraire les spicules de l'ensemble des structures détectées. Cette phase, vient clore la première partie de la conception de notre système, qui est en mesure d'extraire soit des masses spiculées, soit des distorsions architecturales. Afin de finaliser sa conception, nous avons procédé à la création d'un modèle d'aide à la décision qui, à l'inverse de ce qui s'est toujours fait dans l'état-de-l'art pour la discrimination des masses, procède à une extraction non-supervisée des descripteurs à l'aide d'une méthode issue du Deep learning, à savoir les réseaux de neurones à convolution. Les descripteurs extraits, sont ensuite utilisés dans un classifieur SVM pour apprendre un modèle. Ce modèle servira par la suite à la reconnaissance du cancer du sein. Les résultats obtenus pour chacune des étapes du système de diagnostic sont très intéressants et viennent combler un vide important dans la classification des masses en général et dans la distinction des masses malignes entre elles en particulier en se fondant sur trois niveaux de décision que sont la forme, la densité et les spicules. / Computer-aided diagnosis of breast cancer is raising increasingly a genuine enthusiasm because of the ever-increasing quantity of mammographic images from breast cancer screening campaigns. The focus is on breast masses due to the high risk of cancer associated with them. Indeed, the variability of shape encountered and the difficulty to discern the masses especially when theyare embedded in a high density require a new approach especially suited for the most complex cases namely the masses which belong to classes BI-RADS IV and V, i.e. spiculated breast mass and architectural distortion. In this work, a fully automatic computer-aided diagnosis system is designed for the segmentation and classification of breast mass especially for malignant masses of classes BI-RADS IV and BI-RADS V. Initially, we developped a pre-processing method combined with the reduction of the dictionary size in order to remove effectively and quickly the digitization noise of the mammographic images that make up the database used to design our computer-aided diagnosis system in comparison with the existing approaches. After the image pre-processing, we haveproposed an unsupervised segmentation method based on a Markov random field which has the advantage of being faster, more efficient and more robust than the state-of-art segmentation methods. Furthermore, the proposed method overcomes the variability of the breast masses whatever the image density. In purpose to describe correctly the spiculated malignant lesions, we proposed anapproach which avoid the computation and extraction of local features, and to rely on general-purpose classification procedures whose performance and computational efficiency can greatly vary depending on design and image characteristics. The proposed method is based on several assumptions on the structure of spicules as they appear in mammograms which have been reported in the literature. In order to make use of the above assumptions, the proposed method proceeds the following steps: first the mammogram is separated into patches onto which the curvilinear structures are discretized into segments due to Radon transform. Then, Markov modeling and contextual information are used to refine the segment positions and associate segments into curvilinear structures. Finally, spicules are detected based on a contrario model. This stage conclude the first part of the design of our computer-aided diagnosis system, that is able to extract both spiculated masses and architectural distortion. In order to complete the design of the diagnosis system, we carried out the creation of a decision support model which, contrary to what has always been done in the state-of-art for discrimination of the masses, conducts an unsupervised extraction of features through Deep learning approach - namely convolutional artificial neural networks -, combined with an SVM-type classifier. The obtained model is then stored and used as a classifier for breast cancer recognition tasks during the generalization phase. The results obtained for each step of the design of our system are very interesting and come to fill an important gap in the distinction of different type of malignant masses.

Made available in DSpace on 2014-06-12T23:04:14Z (GMT). No. of bitstreams: 2 arquivo8851_1.pdf: 4717828 bytes, checksum: 592acd5b2b91d1cead758a1666f091db (MD5) license.txt: 1748 bytes, checksum: 8a4605be74aa9ea9d79846c1fba20a33 (MD5) Previous issue date: 2007 / As indicações clínicas da core biopsy obedece aos mesmos critérios utilizados para biópsia cirúrgica, mas não há consenso na literatura quanto a sua indicação para avaliação das lesões na categoria mamográfica BI-RADS® 5 (ACR/Breast Imaging Reporting and Data System). Objetivo: Determinar a associação entre as alterações mamográficas e o diagnóstico histopatológico de material obtido por core biopsy estereotáxica de lesões mamárias impalpáveis classificadas na categoria mamográfica BI-RADS® 5, estabelecendo o valor preditivo positivo da mamografia nas lesões altamente suspeitas de malignidade. Materiais e métodos: Por meio de estudo retrospectivo, transversal, analítico, de comparação entre métodos diagnósticos, foram analisadas 70 core biopsies de lesões mamárias impalpáveis, classificadas radiologicamente como altamente suspeitas de malignidade (BI-RADS® 5), de 70 pacientes, atendidas em serviços privados de Anatomia Patológica e Radiologia da cidade do Recife, Pernambuco, no período de 2001 a 2006. Resultados: Eram do sexo feminino 68 (97,1%) pacientes e 2 (2,9%), do masculino. A idade variou de 17 a 87 anos, com média de 58 ± 15 anos. A mama esquerda foi acometida em 42 (60%) casos e a direita, em 28 (40%). Predominaram localização das lesões no QSE (44 casos; 62,9%) e nódulos irregulares espiculados (49 casos; 70%), 11 (15,7%) dos quais associados a microcalcificações. As microcalcificações estavam presentes em 31(44,3%) casos; sendo 16 (22,9%) casos não associados a nódulos, distorção arquitetural ou densidade assimétrica. As core biopsies foram constituídas por 3 a 16 fragmentos (média: 6±2). Não houve diferença na distribuição de freqüência de número de fragmentos em função dos diagnósticos histopatológicos (p>0,05) ou radiológicos (p=0,63). Houve diagnóstico de: 59 (84,3%) casos de carcinoma, 7 (10%) casos de lesões benignas e 4 (5,7%) com lesão borderline. O carcinoma invasivo foi o mais freqüente (49 casos; 70%) e em 15 (21,4%) casos associou-se a componente in situ. O carcinoma in situ puro correspondeu a 10 (14,3%) casos. Houve associação significante entre nódulos irregulares espiculados e carcinoma invasivo (41 casos; 58,6%; p=0,005). O tipo histológico mais encontrado foi o carcinoma ductal invasivo (34 casos; 69,4%). Dentre os carcinomas invasivos, 36 (73,5%) casos tiveram grau histológico 2, com predomínio da soma dos escores igual a 6 (34 casos; 69,4%). O CDIS padrão comedônico puro associou-se mais freqüentemente às microcalcificações. O valor preditivo positivo da avaliação mamográfica na categoria BI-RADS® 5 foi de 84,3%. O maior valor preditivo positivo foi verificado em nódulo irregular espiculado com microcalcificações, com ou sem, sem microcalcificações e microcalcificações sem nódulo (100%, 87,8%, 84,2% e 75%, respectivamente). Conclusões: A avaliação mamográfica das lesões impalpáveis enquadradas como altamente suspeitas de malignidade foi de alto valor preditivo para o diagnóstico de câncer, a maioria correspondendo a carcinoma invasivo. Os nódulos irregulares espiculados tiveram um alto valor preditivo para o diagnóstico de carcinoma, particularmente, quando associados às microcalcificações

Submitted by Nadir Basilio (nadirsb@uninove.br) on 2015-07-27T13:39:57Z No. of bitstreams: 1 Renata Mara Bueno Aguiar.pdf: 3052806 bytes, checksum: 7fc933a7df31cf76bd53ddfab6dc00d3 (MD5) / Made available in DSpace on 2015-07-27T13:39:57Z (GMT). No. of bitstreams: 1 Renata Mara Bueno Aguiar.pdf: 3052806 bytes, checksum: 7fc933a7df31cf76bd53ddfab6dc00d3 (MD5) Previous issue date: 2013-12-11 / Objective: To analyze the mammographic radiologic findings during screening campaigns of a Brazilian Oncology Center. The mammographic radiologic findings, the complementary diagnostic tests eventually requested and the patient’s adherence to the screening recommendations were studied. Methods: A retrospective study was conducted including all 771 patients that participated in the stimulated and gratuitous screening program of the A. C. Camargo Cancer Center, São Paulo, Brazil, during 2008. The patient’s records were analyzed in order to obtain result from mammography tests, breast ultrasonography, biopsies and surgeries. Results: The women`s age varied from 24 to 107 years old. (average: 55 years old), 259 (34%) under 50 years old, 423 (55%) between 50 and 69 years old and 89 (11%) older than 69 years old. The mammography results were classified as follows: BI-RADS 0 =186 (25,1%), BI-RADS 1 = 114 (14,8%), BI-RADS 2 =395 (51,2%), BI-RADS 3 =50 (6,5%), BI-RADS 4 = 16 (2,1%) e BI-RADS 5 =2 (0,3%). Complementary ultrasonography exam was performed in 184 (24%) of all patients. Thirty three (4%) lesions were submitted to histopathological analyses revealing 6 (8%) cases of cancer. They were 3 invasive ductal carcinomas, 2 in situ ductal carcinomas and 1 lobular carcinoma in situ. Five of those cancers were surgically classified as Stage 1 disease and One as Stage 2 disease. After 4 years 87 (15%) patients returned for new screening exams. Conclusion: There was a meaning ratio of patients out of Brazilian National Cancer Institute (INCA) recommendation for breast cancer screening. The rates of complementary ultrasonography were high. The number of patients diagnosed with cancer and the rates of diagnoses in initial grades are in consonance with the literature. The opportunistic screening model revealed low return rate for new screening tests. / Objetivo: Analisar durante uma campanha de rastreamento para câncer de mama realizada em um Centro Brasileiro de Oncologia. Foram estudados os achados radiológicos das mamografias e as eventuais recomendações de testes diagnósticos complementares e a adesão das pacientes a essas recomendações. Metodologia: Esta foi uma análise retrospectiva incluindo todas as 771 pacientes que participaram das campanhas estimulada e gratuita de rastreamento para câncer de mama do A. C. Camargo Cancer Center, em São Paulo, Brasil, no ano de 2008. Foram avaliados os resultados de mamografias, ultrassonografias, biópsias e cirurgias de mama realizadas. Para os diagnósticos de câncer registramos a histologia e o estadiamento clínico da doença. Resultados: A idade das mulheres variou de 24 a 107 anos (média: 55 anos), 259 (34%) com idade inferior a 50 anos, 423 (55%) com idade entre 50 e 69 anos e 89 (11%) com idades maiores que 69 anos. A classificação segundo o BI-RADS do resultado das mamografias foi: BI-RADS 0 186 (25,1%), BI-RADS 1 114 (14,8%), BI-RADS 2 = 395 (51,2%), BI-RADS 3 = 50 (6,5%), BI-RADS 4 =16 (2,1%) e BI-RADS 5 = 2 (0,3%).O exame ultrassonográfico complementar à mamografia foi realizado por 184 (24%) da amostra. Foram realizadas 33 (4%) investigações anátomo-patológicas identificando 6 (8%) casos de câncer, 3 Carcinomas ductais invasivos, 2 carcinomas ductais in situ e 1 carcinoma lobular in situ. O estadiamento cirúrgico dessas neoplasias foi Estadio 1 para 5 pacientes e Estadio 2 em uma paciente. A revisão dos registros realizada após 4 anos identificou 87 (15%) pacientes que haviam retornado ao serviço para novos exames de rastreamento ao menos uma vez. Conclusão: Uma proporção significativa das pacientes de campanha está fora da faixa etária de recomendação do Instituto Nacional do Câncer do Ministério da Saúde (INCA) para rastreamento do câncer de mama. As taxas de complementação com ultrassonografia foram elevadas. O número de cânceres e a taxa de diagnósticos em estádio inicial estão em consonância com os índices preconizados. O modelo de rastreamento oportunístico mostrou baixa taxa de retorno periódico ao mesmo serviço.

Made available in DSpace on 2014-06-11T19:26:18Z (GMT). No. of bitstreams: 0 Previous issue date: 2006-10-30Bitstream added on 2014-06-13T19:13:27Z : No. of bitstreams: 1 giandon_cas_me_botfm_prot.pdf: 2640324 bytes, checksum: 4b109b85ba7836d8177933cba98a32e4 (MD5) / Fundação para o Desenvolvimento Médico e Hospitalar (Famesp) / O câncer de mama é um problema de saúde pública, com 49.470 casos em 2005 no Brasil. A mamografia é o método de escolha para rastreamento e diagnóstico precoce do câncer de mama. O sistema BI RADS veio tentar organizar e padronizar os laudos mamográficos. O VPP do Bl-RADS IV varia de 2 a 90% e da categoria V > 90%. Objetivos: avaliar as principais características das lesões mamográficas suspeitas e calcular o VPP da categoria IV e V. Sujeitos e métodos: foram estudados 309 laudos mamográficos de Bl-RADS IV e V e correlacionados com seus resultados de histopatológicos. A análise estatística foi o teste diagnóstico do VPP. Resultados: idade média 54 anos, mama esquerda acometida em 165 (53,4%) pacientes e mama direita em 144 (46,6%) pacientes, BI RADS IV 265 (85,8%), Bl-RADS V 44 (14,2%), lesões benignas 163 (52,8%), lesões proliferativas de risco 48 (15,5%), lesões malignas 98 (31,7%). Os principais achados mamográficos na categoria IV foram as microcalcificações 163 (61,5%) e na categoria V foram os nódulos em 22 (50,0%). A modalidade de biopsia mais usada na categoria IV foi o agulhamento mamário em 188 (70,9%) e na categoria V foi à biopsia per cutânea em 16 (36,4%). O carcinoma ductal infiltrante foi o tipo histológico mais freqüente na categoria IV em 45 (78,9%) e na categoria V em 35 (85,4%) pacientes. O VPP para a categoria IV foi de 39,2% e da categoria V de 95,5%. Conclusão: Concluímos que os principais achados mamográficos na categoria IV foram as microcalcificações, com VPP desta categoria de 39,2%; na categoria V foram os nódulos os principais achados mamográficos, com VPP de 95,5%, valor este que confirma esta categoria como alta probabilidade de malignidade. / Breast cancer is a serious world health problem. The mammography is the better method for the screening of earlier breast cancer. The Bl-RADS system was introduced to standardize the report results. Bl-RADS IV has a Positive Predictive Value (PPV) ranged from 2 to 90% and over 90% to category V. Objectives: to evaluate the principal characteristics of the suspicious mammographic lesions and calculate PPV of the Bl-RADS category IV and V. Subjects and methods: it was evaluated 309 reports of mammographic abnormalities classified as Bl-RADS IV and V; and their correlations with biopsies results. Resulted: medium age was 54 years, the lesions occurred in 165 left-sided breast (53.4%) and in 144 right-sided breast (46.6%). Bl-RADS IV pattern was identified in 265 (85.8%) reports and BI-RADS V in 44 (14.2%). Benign lesions were found in 163 reports (52.8%), proliferative risk lesions were found in 48 (15.5%), malign lesions were found in 98 (31.7%). The principal mammographic lesion was the microcalcification in category IV and nodules were in category V. The most used procedure for diagnosis was core biopsy in 204 patients (66%). The invasive ductal carcinoma was the predominant histological type in malignant breast tumors. PPV of 39.2% was observed for BI-RADS IV and for BI-RÃDS V was 95.5%. Conclusion: the principal mammographic pattern lesion in the Bl-RADS IV was the microcalcification, with PPV of 39.2% and the nodules were the principal mammographic pattern lesions in the BI-RADS V with PPV of 95.5%. It confirms the high probability of malignancy in BI-RADS V category.

No / Radiographer abnormality detection schemes (RADS) were first introduced in the United Kingdom (UK) in the mid 1980s with the development of the ‘red dot scheme’. This article establishes the current position of UK RADS practice and provides insight into specific areas for development. Method: A postal questionnaire was distributed to 456 sites, including 270 emergency departments and 186 minor injuries units (MIU). Information was sought relating to: the type of emergency department and radiography service provided; details of RADS operated including any education and audit to support radiographer participation; and the mandatory/voluntary nature of the system adopted. Results: A total of 306 (n = 306/456; 74%) responses were received. The large majority of respondents (n = 284/306; 92.8%) indicated that a RADS was in operation. Of these, 221 sites operated a red dot scheme, 7 sites operated a radiographer comment system, and a further 54 sites operated both a red dot and comment scheme. Two sites indicated that a RADS other than red dot or radiographer commenting was operated. Twenty-one different methods of highlighting abnormal images were identified and eight different commenting methods. The RADS was considered mandatory at 25% of sites. Conclusion: This study confirms the continued widespread contribution of radiographers to the trauma diagnostic process through the use of RADS. The informal nature of the systems, inconsistent approaches to audit and education, and variations in the methods employed are issues which require national guidance.

Este trabalho apresenta um método computacional que classifica as mamografias no padrão de densidade BI-RADS, visando auxiliar a detecção precoce do câncer de mama, seja essa realizada por análise visual ou por auxílio computadorizado. A classificação das mamografias em bancos padronizados objetiva eliminar conflitos entre laudos mamográficos de diferentes profissionais, bem como quanto à conduta médica a ser seguida. Entretanto, o estabelecimento de bancos feito visualmente e principalmente em períodos diferentes dificulta sua uniformização, proporcionando uma classificação muito subjetiva e relativamente grosseira em conseqüência a grande variação entre e inter observadores. O método desenvolvido permitiu classificar as imagens independentemente da subjetividade própria à observação visual de quem organizou o banco ou da técnica de exposição aos raios X utilizada. Os resultados foram superiores a 92% mesmo para bancos de imagens totalmente diferentes. Esses resultados foram obtidos respeitando-se as possíveis diferenças de interpretações de diversas equipes médicas. Além do estabelecimento de banco de mamografias com limiares entre as composições bem quantificadas, com esta ferramenta, tanto os estagiários poderão ser treinados para classificar as imagens no padrão de densidades do BI-RADS, respeitando as particularidades locais, quanto os resultados dos CAD poderão ser comparados. / This thesis presents a computational method that classifies the mammography into the composition of the breast tissue density patterns described in the BI-RADS protocol, intended to help in the early detection of breast cancer, either if this detection happens to be realized by visual analysis or by computerized support. The classification of the mammography in standardized database intends to eliminate issues between mammography awards of distinct professionals and the correct medical conduct to be followed. However, the determination of database only visually, especially in different periods, difficult it\'s to standardize, causing an extremely subjective classification and relatively superficial in consequence of the large inter-and intraobserver variability. The method allows classifying the images independently of the subjective quality of the visual analysis from who organized the database or from the technique of the exposition to X-ray employed. The results were superior of 92% even to database totally distinct. These results were obtained respecting eventual differences of interpretation from several medical groups. Beside the establishment of mammography database with thresholding between the well quantified categories, this methodology will consent to probationers to be trained for classify the images according to the composition of the breast tissue density patterns described in the BI-RADS, respecting its local particularity. Likewise, with this methodology, the results from CAD would be compared.

No description available.

Este trabalho apresenta um método computacional que classifica as mamografias no padrão de densidade BI-RADS, visando auxiliar a detecção precoce do câncer de mama, seja essa realizada por análise visual ou por auxílio computadorizado. A classificação das mamografias em bancos padronizados objetiva eliminar conflitos entre laudos mamográficos de diferentes profissionais, bem como quanto à conduta médica a ser seguida. Entretanto, o estabelecimento de bancos feito visualmente e principalmente em períodos diferentes dificulta sua uniformização, proporcionando uma classificação muito subjetiva e relativamente grosseira em conseqüência a grande variação entre e inter observadores. O método desenvolvido permitiu classificar as imagens independentemente da subjetividade própria à observação visual de quem organizou o banco ou da técnica de exposição aos raios X utilizada. Os resultados foram superiores a 92% mesmo para bancos de imagens totalmente diferentes. Esses resultados foram obtidos respeitando-se as possíveis diferenças de interpretações de diversas equipes médicas. Além do estabelecimento de banco de mamografias com limiares entre as composições bem quantificadas, com esta ferramenta, tanto os estagiários poderão ser treinados para classificar as imagens no padrão de densidades do BI-RADS, respeitando as particularidades locais, quanto os resultados dos CAD poderão ser comparados. / This thesis presents a computational method that classifies the mammography into the composition of the breast tissue density patterns described in the BI-RADS protocol, intended to help in the early detection of breast cancer, either if this detection happens to be realized by visual analysis or by computerized support. The classification of the mammography in standardized database intends to eliminate issues between mammography awards of distinct professionals and the correct medical conduct to be followed. However, the determination of database only visually, especially in different periods, difficult it\'s to standardize, causing an extremely subjective classification and relatively superficial in consequence of the large inter-and intraobserver variability. The method allows classifying the images independently of the subjective quality of the visual analysis from who organized the database or from the technique of the exposition to X-ray employed. The results were superior of 92% even to database totally distinct. These results were obtained respecting eventual differences of interpretation from several medical groups. Beside the establishment of mammography database with thresholding between the well quantified categories, this methodology will consent to probationers to be trained for classify the images according to the composition of the breast tissue density patterns described in the BI-RADS, respecting its local particularity. Likewise, with this methodology, the results from CAD would be compared.

京都大学 / 0048 / 新制・課程博士 / 博士(医学) / 甲第22741号 / 医博第4659号 / 新制||医||1046(附属図書館) / 京都大学大学院医学研究科医学専攻 / (主査)教授 辻川 明孝, 教授 伊達 洋至, 教授 羽賀 博典 / 学位規則第4条第1項該当 / Doctor of Medical Science / Kyoto University / DFAM

(AJ) The three aims of the work were as follows: 1. Comparison of prostate magnetic resonance (MR) examination results from 1.5 T and 3 T scanners in patients with prostate carcinoma (PCa). MR findings of 103 patients (ages 44-72 years) were compared with histopathological results after radical prostatectomy. The work was focused on the accuracy of predicting local cancer staging and determining prostate tumour location. Patients were divided into three groups (A, B and C) based on the type of MR scanner and protocol used. Patient groups A and B were examined in 1.5T and 3T MR scanners equipped with surface coils in the identical multiparametric MR imaging protocol included dynamic contrast examination (DCE). Patient group C was examined in a 3T MR scanner without DCE. The highest accuracy of predicting the stage of PCa was seen in patients examined in 3 T MR scanner with DCE included in the protocol, however, no significant differences were seen between results from 1.5 T and 3.T MR scanners. No significant difference was also found in the accuracy of determining the location of prostate tumour between 1.5 T and 3T MR examinations, however, there were significant differences between sequences used, with the highest accuracy attained by using a combination of T2 weighted sequences and diffusion...