Objetivo The problem of efficient screening of the population for early diagnosis of cancer is of great importance. The main objective of the project is to study intrinsic links between ultrasonic and histological images sampled from thyroid gland masses (tumors). Ultrasonic examination is a non-invasive technique well suitable for screening purposes as well as for mass localization and navigation. However, its diagnostic abilities are rather poor, particularly for lesions less than 10 mm. Conversely, histopathological studies are invasive techniques that remain the gold standard for cancer diagnosis. It is expected that the results of the project will contribute to two different domains.Firstly, a deeper understanding will be gained of the mutual relationship between pathological tissue structure at a micro level and ultrasound image characteristics at a macro level.Secondly, results of ultrasound examinations will be obtained that are adaptable for clinical use at the early steps of cancer diagnosis. Since histological and ultrasonic images are of very different scale, imaging principles and visual content, there is no chance to compare them directly.The major subject of the research proposal is a joint analysis of ultrasonic and histological images. These images are obtained for patients according to the protocols agreed by medical teams. Image features include volume, shape and texture features of ultrasonic images as well as calorimetric features and features computed by trace-transformation, principal and independent component analysis methods for histological images. All features are analyzed together with clinical data in order to discover correlations between them, to rank the relative importance of image features, to assess the statistical significance of the links between the data, and to evaluate the diagnostic abilities of the features. The approaches employed are principal/independent component analysis methods, statistical significance tests, multivariate statistical models, multiple linear regression, multidimensional scaling, hierarchical clustering, support vector machines (SVMs), Random Forest, logical-combinatorial methods and algebraic approaches for classification. The medical aspect of the study is to continue 2D and 3D ultrasound examination of the thyroid, and to accomplish thyroid surgery and pathology examination of tissue removed (if required).All data obtained will be classified by pathological analysis into several groups as follows:normal thyroid;non-neoplastic lesions (adenomatous hyperplasia, colloid nodule, cyst, etc.);benign (adenomas) and malignant (carcinomas) neoplastic lesions.The project involves a unique international consortium combining image analysis, medical, and data processing teams. Each team has extensive experience in its research field and is recognized by high quality publications in international peer reviewed journals. Programa(s) IC-INTAS - International Association for the promotion of cooperation with scientists from the independent states of the former Soviet Union (INTAS), 1993- Tema(s) IT 2004 - Thematic Call - Information Technology Convocatoria de propuestas Data not available Régimen de financiación Data not available Coordinador University Medical Center Utrecht University, Medical Center Utrecht Aportación de la UE Sin datos Dirección Heidelberglaan 100 3584 CX Utrecht Países Bajos Ver en el mapa Coste total Sin datos Participantes (7) Ordenar alfabéticamente Ordenar por aportación de la UE Ampliar todo Contraer todo BELARUS STATE MEDICAL UNIVERSITY Bielorrusia Aportación de la UE Sin datos Dirección DZERZHINSKY AV., 83 MINSK Ver en el mapa Coste total Sin datos Belarussian Medical Academy for Postgraduate Bielorrusia Aportación de la UE Sin datos Dirección Belskogo 16 220092 Minsk Ver en el mapa Coste total Sin datos National Academy of Sciences of Belarus United, Institute of Informatics Problems, Bielorrusia Aportación de la UE Sin datos Dirección Surganova 6 220012 Minsk Ver en el mapa Coste total Sin datos PENZA STATE UNIVERSITY Rusia Aportación de la UE Sin datos Dirección KRASNAYA, 40 PENZA Ver en el mapa Coste total Sin datos Russian Academy of Sciences, Dorodnicyn Computing Centre of the RAS Rusia Aportación de la UE Sin datos Dirección Vavilova 40 119991 Moscow Ver en el mapa Coste total Sin datos University of Surrey Reino Unido Aportación de la UE Sin datos Dirección GU2 7XH Guildford Ver en el mapa Coste total Sin datos University of Wurzburg Alemania Aportación de la UE Sin datos Dirección Josef-Schneider-Strasse 2 D-97080 Wurzburg Ver en el mapa Coste total Sin datos
