Search

Background Diagnosis of pulmonary thromboembolism (PTE) in dogs relies on computed tomography pulmonary angiography (CTPA), but detailed interpretation of CTPA images is demanding for the radiologist and only large vessels may be evaluated. New approaches for better detection of smaller thrombi include dual energy computed tomography (DECT) as well as computer assisted diagnosis (CAD) techniques. The purpose of this study was to investigate the performance of quantitative texture analysis for detecting dogs with PTE using grey-level co-occurrence matrices (GLCM) and multivariate statistical classification analyses. CT images from healthy (n = 6) and diseased (n = 29) dogs with and without PTE confirmed on CTPA were segmented so that only tissue with CT numbers between −1024 and −250 Houndsfield Units (HU) was preserved. GLCM analysis and subsequent multivariate classification analyses were performed on texture parameters extracted from these images. Results Leave-one-dog-out cross validation and receiver operator characteristic (ROC) showed that the models generated from the texture analysis were able to predict healthy dogs with optimal levels of performance. Partial Least Square Discriminant Analysis (PLS-DA) obtained a sensitivity of 94% and a specificity of 96%, while Support Vector Machines (SVM) yielded a sensitivity of 99% and a specificity of 100%. The models, however, performed worse in classifying the type of disease in the diseased dog group: In diseased dogs with PTE sensitivities were 30% (PLS-DA) and 38% (SVM), and specificities were 80% (PLS-DA) and 89% (SVM). In diseased dogs without PTE the sensitivities of the models were 59% (PLS-DA) and 79% (SVM) and specificities were 79% (PLS-DA) and 82% (SVM). Conclusion The results indicate that texture analysis of CTPA images using GLCM is an effective tool for distinguishing healthy from abnormal lung. Furthermore the texture of pulmonary parenchyma in dogs with PTE is

Background Diagnosis of pulmonary thromboembolism (PTE) in dogs relies on computed tomography pulmonary angiography (CTPA), but detailed interpretation of CTPA images is demanding for the radiologist and only large vessels may be evaluated. New approaches for better detection of smaller thrombi include dual energy computed tomography (DECT) as well as computer assisted diagnosis (CAD) techniques. The purpose of this study was to investigate the performance of quantitative texture analysis for detecting dogs with PTE using grey-level co-occurrence matrices (GLCM) and multivariate statistical classification analyses. CT images from healthy (n = 6) and diseased (n = 29) dogs with and without PTE confirmed on CTPA were segmented so that only tissue with CT numbers between −1024 and −250 Houndsfield Units (HU) was preserved. GLCM analysis and subsequent multivariate classification analyses were performed on texture parameters extracted from these images. Results Leave-one-dog-out cross validation and receiver operator characteristic (ROC) showed that the models generated from the texture analysis were able to predict healthy dogs with optimal levels of performance. Partial Least Square Discriminant Analysis (PLS-DA) obtained a sensitivity of 94% and a specificity of 96%, while Support Vector Machines (SVM) yielded a sensitivity of 99% and a specificity of 100%. The models, however, performed worse in classifying the type of disease in the diseased dog group: In diseased dogs with PTE sensitivities were 30% (PLS-DA) and 38% (SVM), and specificities were 80% (PLS-DA) and 89% (SVM). In diseased dogs without PTE the sensitivities of the models were 59% (PLS-DA) and 79% (SVM) and specificities were 79% (PLS-DA) and 82% (SVM). Conclusion The results indicate that texture analysis of CTPA images using GLCM is an effective tool for distinguishing healthy from abnormal lung. Furthermore the texture of pulmonary parenchyma in dogs with PTE is

Background Diagnosis of pulmonary thromboembolism (PTE) in dogs relies on computed tomography pulmonary angiography (CTPA), but detailed interpretation of CTPA images is demanding for the radiologist and only large vessels may be evaluated. New approaches for better detection of smaller thrombi include dual energy computed tomography (DECT) as well as computer assisted diagnosis (CAD) techniques. The purpose of this study was to investigate the performance of quantitative texture analysis for detecting dogs with PTE using grey-level co-occurrence matrices (GLCM) and multivariate statistical classification analyses. CT images from healthy (n = 6) and diseased (n = 29) dogs with and without PTE confirmed on CTPA were segmented so that only tissue with CT numbers between −1024 and −250 Houndsfield Units (HU) was preserved. GLCM analysis and subsequent multivariate classification analyses were performed on texture parameters extracted from these images. Results Leave-one-dog-out cross validation and receiver operator characteristic (ROC) showed that the models generated from the texture analysis were able to predict healthy dogs with optimal levels of performance. Partial Least Square Discriminant Analysis (PLS-DA) obtained a sensitivity of 94% and a specificity of 96%, while Support Vector Machines (SVM) yielded a sensitivity of 99% and a specificity of 100%. The models, however, performed worse in classifying the type of disease in the diseased dog group: In diseased dogs with PTE sensitivities were 30% (PLS-DA) and 38% (SVM), and specificities were 80% (PLS-DA) and 89% (SVM). In diseased dogs without PTE the sensitivities of the models were 59% (PLS-DA) and 79% (SVM) and specificities were 79% (PLS-DA) and 82% (SVM). Conclusion The results indicate that texture analysis of CTPA images using GLCM is an effective tool for distinguishing healthy from abnormal lung. Furthermore the texture of pulmonary parenchyma in dogs with PTE is

Introduction: Co-occurrence grey level matrix (GLCM) is a textural analysis method that have been useful to discriminate patterns, but no used on tendon ultrasound image. Objective: Textural analysis of patellar tendon ultrasonograph. Method: Longitudinal analytic study with 16 subjects (8 women and 8 men) young, healthy and sedentary people with training by means wholebody vibration platform (Fitvibe Medical) for 2 days x 14 weeks. Cross-sectional of patellar tendon ultrasonographics were taken with a Sonosite-180 ultrasonograph (L 5-10 MHz). By means GLCM algorithm of Image J v1.38 it were calculated five textural parameters: Uniformity (ASM), Contrast, Correlation, Homogeneity (IDM) and Entropy in four orientations (0º, 90º, 180º and 270º) and three distances between pixels (d=1, 5 and 10 pixels). Wilcoxon test (C.I. 95%) for related samples was applied (SPSS 15.0). Results: Entropy (d=5) was the most sensible to detect textural changes; perhaps ASM and Contrast can be also useful. It seems that distances between pairs of pixels that the algorithm uses affects the results. Conclusions: The use of GLCM in the textural analysis of tendon ultrasonography is innovating and it appears like a useful tool as much to evaluate the histological evolution of tendon tissue, like coming up and detecting future pathologies precociously. But more analyses will be necessary to study how different parameters affect texture and how they are related to each other and which can be the best adjustments of the algorithm to detect changes in textural patterns., Introducción: Las matrices de co-ocurrencia del nivel de gris (GLCM) son útiles para el análisis textural de imágenes ya la discriminación de patrones pero hasta ahora no se han aplicado sobre imágenes ecográficas del tendón. Objetivo: Análisis textural ecográfico del tendón rotuliano. Método: Estudio longitudinal analítico con 16 sujetos (8 mujeres y 8 hombres) jóvenes, sanos y sedentarios entrenados con una plataforma de vibración vertical (Fitvibe Medical) 2 días x 14 semanas. Se tomaron cortes ecográficos transversales del tendón rotuliano antes y después del entrenamiento con un ecógrafo Sonosite-180 (Lineal 5-10 MHz). Mediante el algoritmo GLCM de Image J v1.38 se calcularon las variables texturales Uniformidad (ASM), Contraste, Correlación, Homogeneidad (IDM) y la Entropía para cuatro orientaciones (0º, 90º, 180º y 270º) y tres distancias (d=1, 5 y 10 px). Se aplicó la prueba de Wilcoxon (i.c.95%) para muestras relacionadas (SPSS 15.0). Resultados: la Entropía (d=5) fue la más sensible a los cambios texturales; quizá la variable ASM, pueda resultar también de interés junto con el Contraste. Conclusiones: Ante la falta de referencias con el uso de la GLCM en el análisis textural de ecografía de tendón son necesarios más análisis que estudien cómo afectan los distintos parámetros a las variables texturales, cómo se relacionan entre sí y cuáles pueden ser los mejores ajustes del algoritmo para detectar cambios en el patrón textural.