Johan Chaniot, M. Moreaud, L. Sorbier, D. Jeulin, J. Becker, T. Fournel
Morphological characterization of porous media is of paramount interest, mainly due to the connections between their physicochemical properties and their porous microstructure geometry. Heterogeneity can be seen as a geometric characteristic of porous microstructures. In this paper, two novel topological descriptors are proposed, based on the M-tortuosity formalism. Using the concept of geometric tortuosity or morphological tortuosity, a first operator is defined, the H-tortuosity . It estimates the average variations of the morphological tortuosity as a function of the scale, based on Monte Carlo method and assessing the heterogeneity of porous networks. The second descriptor is an extension, named the H-tortuosity-by-iterativeerosions , taking into account different percolating particle sizes. These two topological operators are applied on Cox multi-scale Boolean models, to validate their behaviors and to highlight their discriminative power.
{"title":"HETEROGENEITY ASSESSMENT BASED ON AVERAGE VARIATIONS OF MORPHOLOGICAL TORTUOSITY FOR COMPLEX POROUS STRUCTURES CHARACTERIZATION","authors":"Johan Chaniot, M. Moreaud, L. Sorbier, D. Jeulin, J. Becker, T. Fournel","doi":"10.5566/ias.2370","DOIUrl":"https://doi.org/10.5566/ias.2370","url":null,"abstract":"Morphological characterization of porous media is of paramount interest, mainly due to the connections between their physicochemical properties and their porous microstructure geometry. Heterogeneity can be seen as a geometric characteristic of porous microstructures. In this paper, two novel topological descriptors are proposed, based on the M-tortuosity formalism. Using the concept of geometric tortuosity or morphological tortuosity, a first operator is defined, the H-tortuosity . It estimates the average variations of the morphological tortuosity as a function of the scale, based on Monte Carlo method and assessing the heterogeneity of porous networks. The second descriptor is an extension, named the H-tortuosity-by-iterativeerosions , taking into account different percolating particle sizes. These two topological operators are applied on Cox multi-scale Boolean models, to validate their behaviors and to highlight their discriminative power.","PeriodicalId":49062,"journal":{"name":"Image Analysis & Stereology","volume":"61 1","pages":"111-128"},"PeriodicalIF":0.9,"publicationDate":"2020-06-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"80580270","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Most of the methods of classification of breast lesions in ultrasound (US) images have been tested on B-mode images from the commercial equipment. The new possibility of further analysis of this problem showed up with the availability of a public database containing original raw radio frequency (RF) signals. In particular, it appeared that the original texture might contain diagnostic information which could be modified in the typical image processing and which is more difficult to perceive than the details of lesion shape/contour. In this paper a detailed analysis of the lesion texture is conducted by means of the decision trees and logistic regression. The decision trees turned out useful mainly for selecting texture features to be employed in the stepwise logistic regression. The RF signals database of 200 breast lesions was used for testing the performance of the benign vs malignant lesion classifier. The Gray Level Cooccurrence Matrix (GLCM) was calculated with the vertical/horizontal offset of up to five pixels. For each of these matrices six features were calculated resulting in a total of 210 features. Using these features a sufficient number of decision trees were generated to calculate pseudo-Receiver Operating Characteristics (ROCs). The outcome of this process is a collection of generated trees for which the employed features are known. These features were then used for generating generalized linear model by means of stepwise logistic regression. The analyzed regression models included the coefficients of up-to-the second degree terms. The texture features were further completed by a single shape feature, that is tumor circularity. The automatic procedure for finding the exact mask of a lesion is also provided for the conditions when the acoustic shadowing makes it impossible to obtain the entire contour reliably and a half-contour has to be used. The selected logistic regression models gave ROCs with the Area Under Curve (AUC) of up to 0.83 and the 95 % confidence region (0.63 0.96). Analyzing classification results one comes to the conclusion that the tumor circularity, which is the most informative among shape/contour features, is not essential against the background of textural features. The reported study shows that a relatively straightforward procedure can be employed to obtain benign vs malignant classifier comparable with that originally used for the database of the raw RF signals and based on the more complicated segmentation of the parameter maps of homodyned K distribution.
{"title":"Study of classification of breast lesions using texture GLCM features obtained from the raw ultrasound signal","authors":"Mariusz Nieniewski, L. Chmielewski","doi":"10.5566/ias.2113","DOIUrl":"https://doi.org/10.5566/ias.2113","url":null,"abstract":"Most of the methods of classification of breast lesions in ultrasound (US) images have been tested on B-mode images from the commercial equipment. The new possibility of further analysis of this problem showed up with the availability of a public database containing original raw radio frequency (RF) signals. In particular, it appeared that the original texture might contain diagnostic information which could be modified in the typical image processing and which is more difficult to perceive than the details of lesion shape/contour. In this paper a detailed analysis of the lesion texture is conducted by means of the decision trees and logistic regression. The decision trees turned out useful mainly for selecting texture features to be employed in the stepwise logistic regression. The RF signals database of 200 breast lesions was used for testing the performance of the benign vs malignant lesion classifier. The Gray Level Cooccurrence Matrix (GLCM) was calculated with the vertical/horizontal offset of up to five pixels. For each of these matrices six features were calculated resulting in a total of 210 features. Using these features a sufficient number of decision trees were generated to calculate pseudo-Receiver Operating Characteristics (ROCs). The outcome of this process is a collection of generated trees for which the employed features are known. These features were then used for generating generalized linear model by means of stepwise logistic regression. The analyzed regression models included the coefficients of up-to-the second degree terms. The texture features were further completed by a single shape feature, that is tumor circularity. The automatic procedure for finding the exact mask of a lesion is also provided for the conditions when the acoustic shadowing makes it impossible to obtain the entire contour reliably and a half-contour has to be used. The selected logistic regression models gave ROCs with the Area Under Curve (AUC) of up to 0.83 and the 95 % confidence region (0.63 0.96). Analyzing classification results one comes to the conclusion that the tumor circularity, which is the most informative among shape/contour features, is not essential against the background of textural features. The reported study shows that a relatively straightforward procedure can be employed to obtain benign vs malignant classifier comparable with that originally used for the database of the raw RF signals and based on the more complicated segmentation of the parameter maps of homodyned K distribution.","PeriodicalId":49062,"journal":{"name":"Image Analysis & Stereology","volume":"42 1","pages":"129-145"},"PeriodicalIF":0.9,"publicationDate":"2020-06-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"86357473","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The segmentation of rock grains on images depicting bulk rock materials is considered. The rocks’ material images are transformed by selected texture operators, to obtain a set of features describing them. The first order features, second-order features, run-length matrix, grey tone difference matrix, and Laws’ energies are used for this purpose. The features are classified using k-nearest neighbours, support vector machines, and artificial neural networks classifiers. The results show that the border of rocks grains can be determined with above 75% accuracy. The multi-texture approach was also investigated, leading to an increase in accuracy to over 79% for the early-fusion of features. Attempts were made to reduce feature space dimensionality by manually picking features as well as by the use of principal component analysis. The outcomes showed a significant decrease in accuracy. The obtained results have been visually compared with the ground truth. The compliance observed can be considered to be satisfactory.
{"title":"Application of texture features and machine learning methods to grains segmentation in rock material images","authors":"K. Nurzynska, S. Iwaszenko","doi":"10.5566/ias.2186","DOIUrl":"https://doi.org/10.5566/ias.2186","url":null,"abstract":"The segmentation of rock grains on images depicting bulk rock materials is considered. The rocks’ material images are transformed by selected texture operators, to obtain a set of features describing them. The first order features, second-order features, run-length matrix, grey tone difference matrix, and Laws’ energies are used for this purpose. The features are classified using k-nearest neighbours, support vector machines, and artificial neural networks classifiers. The results show that the border of rocks grains can be determined with above 75% accuracy. The multi-texture approach was also investigated, leading to an increase in accuracy to over 79% for the early-fusion of features. Attempts were made to reduce feature space dimensionality by manually picking features as well as by the use of principal component analysis. The outcomes showed a significant decrease in accuracy. The obtained results have been visually compared with the ground truth. The compliance observed can be considered to be satisfactory.","PeriodicalId":49062,"journal":{"name":"Image Analysis & Stereology","volume":"4 1","pages":"73-90"},"PeriodicalIF":0.9,"publicationDate":"2020-06-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"87094112","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Knowledge of pedicle morphometry is valuable for a safe and reliable pedicle screw placement. In this study, we performed and evaluated computerized pedicle morphometry measurements from preoperative computed tomography (CT) images of the thoracic spine from 26 subjects. Manual measurements of the pedicle width, height and chord length were obtained for 540 thoracic pedicles in selected cross sections of orthogonal and oblique multiplanar reconstructions (MPRs). Computerized measurements of the pedicle width, height, length, chord length, transverse angulation, sagittal angulation and cross-sectional area were obtained for the same pedicles by an automated method that is based on parametric modeling of vertebral structures in three dimensions (3D). Statistical analysis revealed that manual measurements from orthogonal MPRs were significantly different (p ≤ 0.0011) when compared to those from oblique MPRs and computerized measurement in 3D, with the respective mean absolute difference (MAD) ± standard deviation (SD) of 0.77 ± 0.56 mm and 0.74 ± 0.57 mm for the pedicle width, and 1.31 ± 1.08 mm and 1.45 ± 1.10 mm for the pedicle height. No statistically significant differences (p ≥ 0.12) were observed between manual measurements from oblique MPRs and computerized measurements in 3D, with MAD ± SD of 0.44 ± 0.35 mm, 0.56 ± 0.52 mm and 1.72 ± 1.29 mm for the pedicle width, height and chord length, respectively. The advantage of computerized measurements is that they allow the extraction of additional pedicle morphometric parameters, which are important for preoperative planning of pedicle screw placement, or can be used for population and demographic studies using larger pedicle databases.
{"title":"Computerized three-dimensional pedicle morphometry from computed tomography images of the thoracic spine","authors":"Dejan Knez, T. Vrtovec","doi":"10.5566/ias.2349","DOIUrl":"https://doi.org/10.5566/ias.2349","url":null,"abstract":"Knowledge of pedicle morphometry is valuable for a safe and reliable pedicle screw placement. In this study, we performed and evaluated computerized pedicle morphometry measurements from preoperative computed tomography (CT) images of the thoracic spine from 26 subjects. Manual measurements of the pedicle width, height and chord length were obtained for 540 thoracic pedicles in selected cross sections of orthogonal and oblique multiplanar reconstructions (MPRs). Computerized measurements of the pedicle width, height, length, chord length, transverse angulation, sagittal angulation and cross-sectional area were obtained for the same pedicles by an automated method that is based on parametric modeling of vertebral structures in three dimensions (3D). Statistical analysis revealed that manual measurements from orthogonal MPRs were significantly different (p ≤ 0.0011) when compared to those from oblique MPRs and computerized measurement in 3D, with the respective mean absolute difference (MAD) ± standard deviation (SD) of 0.77 ± 0.56 mm and 0.74 ± 0.57 mm for the pedicle width, and 1.31 ± 1.08 mm and 1.45 ± 1.10 mm for the pedicle height. No statistically significant differences (p ≥ 0.12) were observed between manual measurements from oblique MPRs and computerized measurements in 3D, with MAD ± SD of 0.44 ± 0.35 mm, 0.56 ± 0.52 mm and 1.72 ± 1.29 mm for the pedicle width, height and chord length, respectively. The advantage of computerized measurements is that they allow the extraction of additional pedicle morphometric parameters, which are important for preoperative planning of pedicle screw placement, or can be used for population and demographic studies using larger pedicle databases.","PeriodicalId":49062,"journal":{"name":"Image Analysis & Stereology","volume":"36 1","pages":""},"PeriodicalIF":0.9,"publicationDate":"2020-06-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"74416778","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
To perform a detailed evaluation of reciprocating motion using a computer-aided phase identification and frame-to-frame analysis, continuous rotation at 300 rpm, RECIPROC ALL mode and WAVEONE ALL-mode were recorded with a high-speed camera. Movie files were automatically analyzed with digital video analysis and modeling tool. RECIPROC ALL mode parameters were 186.34°±1.02 at 428.32 rpm ±7.61 and 65.07°±0.93 at 261.06 rpm ± 7.72; WAVEONE ALL-mode parameters were 191.39°±1.32 at 523.83 rpm ±14.36 and 70.13°±1.26 at 316.06 rpm ± 8.75. The variability of rotational speed during the cycle and distinct acceleration –deceleration patterns, was similar for both reciprocating modes. The computer-aided frame-to-frame analysis revealed that asymmetrical reciprocating motion has more complex kinematics demonstrating high peak rotational speed values and different patterns of acceleration and deceleration. While there was a difference in reciprocating cycle duration and rotational speed, both cycles demonstrated a similar dynamic of rotational speed during the cycle.
{"title":"COMPUTER-AIDED PHASE IDENTIFICATION AND FRAME-TO-FRAME ANALYSIS OF ENDODONTIC ASYMMETRIC RECIPROCATION ROTATION: A PRELIMINARY STUDY","authors":"A. Fidler, E. O. Orhan, Özgür Irmak","doi":"10.5566/ias.2335","DOIUrl":"https://doi.org/10.5566/ias.2335","url":null,"abstract":"To perform a detailed evaluation of reciprocating motion using a computer-aided phase identification and frame-to-frame analysis, continuous rotation at 300 rpm, RECIPROC ALL mode and WAVEONE ALL-mode were recorded with a high-speed camera. Movie files were automatically analyzed with digital video analysis and modeling tool. RECIPROC ALL mode parameters were 186.34°±1.02 at 428.32 rpm ±7.61 and 65.07°±0.93 at 261.06 rpm ± 7.72; WAVEONE ALL-mode parameters were 191.39°±1.32 at 523.83 rpm ±14.36 and 70.13°±1.26 at 316.06 rpm ± 8.75. The variability of rotational speed during the cycle and distinct acceleration –deceleration patterns, was similar for both reciprocating modes. The computer-aided frame-to-frame analysis revealed that asymmetrical reciprocating motion has more complex kinematics demonstrating high peak rotational speed values and different patterns of acceleration and deceleration. While there was a difference in reciprocating cycle duration and rotational speed, both cycles demonstrated a similar dynamic of rotational speed during the cycle.","PeriodicalId":49062,"journal":{"name":"Image Analysis & Stereology","volume":"94 1","pages":""},"PeriodicalIF":0.9,"publicationDate":"2020-06-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"91065891","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
J. Kopeček, J. Stanek, S. Habr, F. Seitl, L. Petrich, V. Schmidt, V. Beneš
The aim of this paper is to evaluate an ambitious imaging experiment and to contribute to the methodology of statistical inference of the three-dimensional microstructure of polycrystalline materials. The microstructure of the considered Al-3Mg-0.2Sc alloy was investigated by three-dimensional electron backscattered diffraction (3D-EBSD), i.e., tomographic imaging with xenon plasma focused ion beam (Xe-FIB) alongside EBSD. The samples were subjected to severe plastic deformations by equal channel angular pressing (ECAP) and annealed subsequently prior to the mapping. First we compared the misorientation level needed for a reliable segmentation of grains distinguishing between conventional evaluation of two-dimensional cuts and the 3D data set. Then, using methods of descriptive spatial statistics, various morphological characteristics of a large number of grains were analyzed, as well as the crystallographic texture and the spatial distribution of grain boundaries. According to the results stated so far in the literature, an even microstructure was expected, nevertheless local inhomogeneities in grains and grain boundaries with regard to their size, texture and spatial distribution were observed and justified.
{"title":"Analysis of polycrystalline microstructure of AlMgSc alloy observed by 3D EBSD","authors":"J. Kopeček, J. Stanek, S. Habr, F. Seitl, L. Petrich, V. Schmidt, V. Beneš","doi":"10.5566/ias.2224","DOIUrl":"https://doi.org/10.5566/ias.2224","url":null,"abstract":"The aim of this paper is to evaluate an ambitious imaging experiment and to contribute to the methodology of statistical inference of the three-dimensional microstructure of polycrystalline materials. The microstructure of the considered Al-3Mg-0.2Sc alloy was investigated by three-dimensional electron backscattered diffraction (3D-EBSD), i.e., tomographic imaging with xenon plasma focused ion beam (Xe-FIB) alongside EBSD. The samples were subjected to severe plastic deformations by equal channel angular pressing (ECAP) and annealed subsequently prior to the mapping. First we compared the misorientation level needed for a reliable segmentation of grains distinguishing between conventional evaluation of two-dimensional cuts and the 3D data set. Then, using methods of descriptive spatial statistics, various morphological characteristics of a large number of grains were analyzed, as well as the crystallographic texture and the spatial distribution of grain boundaries. According to the results stated so far in the literature, an even microstructure was expected, nevertheless local inhomogeneities in grains and grain boundaries with regard to their size, texture and spatial distribution were observed and justified.","PeriodicalId":49062,"journal":{"name":"Image Analysis & Stereology","volume":"26 1","pages":"1-11"},"PeriodicalIF":0.9,"publicationDate":"2020-04-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"86163637","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Katja Fink, M. Prebil, N. Vardjan, Jørgen Jensen, R. Zorec, M. Kreft
Glycogen synthase kinase 3 (GSK-3) plays an important role in metabolic regulation in skeletal muscles, and both insulin and adrenaline stimulate GSK-3 phosphorylation. The aim of the present study was to study the effect of insulin and adrenaline on GSK-3 localisation in skeletal muscles. We characterized subcellular localization of (GSK-3) signal protein in fully differentiated muscle fibre by immunofluorescence and confocal microscopy. We stimulated muscle fibres with insulin and/or adrenaline. Images were analysed by segmentation of single central optical section of the muscle. We found GSK-3 to be localised in clusters. The number of GSK-3 clusters and their average size were increased after stimulation with insulin and/or adrenaline. Average GSK-3 particle size is linearly related to their quantity. We conclude that subcellular GSK-3 in isolated skeletal muscle fibres is localized in clusters and clustering increased after stimulation with insulin and/or adrenaline.
{"title":"Increase in Subcellular GSK-3 Clusters in Insulin- and Adrenaline-treated Differentiated Rat Skeletal Muscle Fibres","authors":"Katja Fink, M. Prebil, N. Vardjan, Jørgen Jensen, R. Zorec, M. Kreft","doi":"10.5566/ias.2356","DOIUrl":"https://doi.org/10.5566/ias.2356","url":null,"abstract":"Glycogen synthase kinase 3 (GSK-3) plays an important role in metabolic regulation in skeletal muscles, and both insulin and adrenaline stimulate GSK-3 phosphorylation. The aim of the present study was to study the effect of insulin and adrenaline on GSK-3 localisation in skeletal muscles. We characterized subcellular localization of (GSK-3) signal protein in fully differentiated muscle fibre by immunofluorescence and confocal microscopy. We stimulated muscle fibres with insulin and/or adrenaline. Images were analysed by segmentation of single central optical section of the muscle. We found GSK-3 to be localised in clusters. The number of GSK-3 clusters and their average size were increased after stimulation with insulin and/or adrenaline. Average GSK-3 particle size is linearly related to their quantity. We conclude that subcellular GSK-3 in isolated skeletal muscle fibres is localized in clusters and clustering increased after stimulation with insulin and/or adrenaline.","PeriodicalId":49062,"journal":{"name":"Image Analysis & Stereology","volume":"1 1","pages":"25-32"},"PeriodicalIF":0.9,"publicationDate":"2020-04-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83074192","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Shape analysis is of great importance in many fields, such as computer vision, medical imaging, and computational biology. This analysis can be performed considering shapes as closed planar curves in the shape space. This approach has been used for the first time to obtain the morphological classification of erythrocytes in digital images of sickle cell disease considering the shape space S1, which has the property of being isometric to an infinite-dimensional Grassmann manifold of two-dimensional subspaces (Younes et al. , 2008), without taking advantage of all the features offered by the elastic metric related to the possibility of stretching and bending of the curves. In this paper, we study this deformation in the shape space, S2, which is based on the representation of closed planar curves by means of the square-root velocity function (SRVF) (Srivastava et al. , 2011), using the elastic metric of this space to obtain more efficient geodesics and geodesic lengths between planar curves. Supervised classification with this approach achieved an accuracy of 94.3%, classification using templates achieved 94.2% and unsupervised clustering in three groups achieved 94.7%, considering three classes of erythrocytes: normal, sickle, and with other deformations. These results are better than those previously achieved in the morphological analysis of erythrocytes and the method can be used in different applications related to the treatment of sickle cell disease, even in cases where it is necessary to study the process of evolution of the deformation, something that can not be done in a natural way in the feature space.
形状分析在许多领域都很重要，如计算机视觉、医学成像和计算生物学。这种分析可以将形状视为形状空间中的封闭平面曲线。该方法首次用于考虑形状空间S1的镰状细胞病数字图像中红细胞的形态分类，该形状空间S1具有与二维子空间的无限维Grassmann流形等长的特性(Younes et al.， 2008)，而没有利用与曲线拉伸和弯曲可能性相关的弹性度量所提供的所有特征。本文在形状空间S2中研究这种变形，该空间基于用平方根速度函数(SRVF)表示封闭平面曲线(Srivastava et al.， 2011)，利用该空间的弹性度量来获得更有效的平面曲线之间的测地线和测地线长度。该方法的监督分类准确率为94.3%，模板分类准确率为94.2%，三组无监督聚类准确率为94.7%，考虑了正常红细胞、镰状红细胞和其他变形红细胞三种类型。这些结果比以前在红细胞形态分析中取得的结果要好，并且该方法可以用于与镰状细胞病治疗相关的不同应用，甚至在需要研究变形演变过程的情况下，这是在特征空间中无法以自然方式完成的。
{"title":"MORPHOLOGICAL ANALYSIS OF CELLS BY MEANS OF AN ELASTIC METRIC IN THE SHAPE SPACE","authors":"I. Epifanio, X. Gual-Arnau, S. Herold-García","doi":"10.5566/ias.2183","DOIUrl":"https://doi.org/10.5566/ias.2183","url":null,"abstract":"Shape analysis is of great importance in many fields, such as computer vision, medical imaging, and computational biology. This analysis can be performed considering shapes as closed planar curves in the shape space. This approach has been used for the first time to obtain the morphological classification of erythrocytes in digital images of sickle cell disease considering the shape space S1, which has the property of being isometric to an infinite-dimensional Grassmann manifold of two-dimensional subspaces (Younes et al. , 2008), without taking advantage of all the features offered by the elastic metric related to the possibility of stretching and bending of the curves. In this paper, we study this deformation in the shape space, S2, which is based on the representation of closed planar curves by means of the square-root velocity function (SRVF) (Srivastava et al. , 2011), using the elastic metric of this space to obtain more efficient geodesics and geodesic lengths between planar curves. Supervised classification with this approach achieved an accuracy of 94.3%, classification using templates achieved 94.2% and unsupervised clustering in three groups achieved 94.7%, considering three classes of erythrocytes: normal, sickle, and with other deformations. These results are better than those previously achieved in the morphological analysis of erythrocytes and the method can be used in different applications related to the treatment of sickle cell disease, even in cases where it is necessary to study the process of evolution of the deformation, something that can not be done in a natural way in the feature space.","PeriodicalId":49062,"journal":{"name":"Image Analysis & Stereology","volume":"14 5","pages":"13-23"},"PeriodicalIF":0.9,"publicationDate":"2020-04-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"72613031","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Random planar tessellations are presented which are generated by subsequent division of their polygonal cells. The purpose is to develop parametric models for crack patterns appearing at length scales which can change by orders of magnitude in areas such as nanotechnology, materials science, soft matter, and geology. Using the STIT tessellation as a reference model and comparing with phenomena in real crack patterns, three modifications of STIT are suggested. For all these models a simulation tool, which also yields several statistics for the tessellation cells, is provided on the web. The software is freely available via a link given in the bibliography of this article. The present paper contains results of a simulation study indicating some essential features of the models. Finally, an example of a real fracture pattern is considered which is obtained using the deposition of a thin metallic film onto an elastomer material – the results of this are compared to the predictions of the model.
{"title":"Modeling Crack Patterns by Modified STIT Tessellations","authors":"R. Leon, W. Nagel, J. Ohser, S. Arscott","doi":"10.5566/ias.2245","DOIUrl":"https://doi.org/10.5566/ias.2245","url":null,"abstract":"Random planar tessellations are presented which are generated by subsequent division of their polygonal cells. The purpose is to develop parametric models for crack patterns appearing at length scales which can change by orders of magnitude in areas such as nanotechnology, materials science, soft matter, and geology. Using the STIT tessellation as a reference model and comparing with phenomena in real crack patterns, three modifications of STIT are suggested. For all these models a simulation tool, which also yields several statistics for the tessellation cells, is provided on the web. The software is freely available via a link given in the bibliography of this article. The present paper contains results of a simulation study indicating some essential features of the models. Finally, an example of a real fracture pattern is considered which is obtained using the deposition of a thin metallic film onto an elastomer material – the results of this are compared to the predictions of the model.","PeriodicalId":49062,"journal":{"name":"Image Analysis & Stereology","volume":"23 1","pages":"33-46"},"PeriodicalIF":0.9,"publicationDate":"2020-04-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"76811515","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The direct superimposition of a standard test grid of congruent quadrats onto an image bearing a population of particles exhibiting perspective artifacts, tends to increase the variance of the population size estimator, because the quadrat contents become unbalanced. If the quadrats are transformed according to the same projection mechanism affecting the particles, however, then the variance is restored into moderate values. Our purpose was to provide exact, easily programmable equations for the relevant transform.
{"title":"Transformation of a Grid of Quadrats to Cope With Perspective Artifacts","authors":"L. Cruz-Orive, M. Cruz","doi":"10.5566/ias.2296","DOIUrl":"https://doi.org/10.5566/ias.2296","url":null,"abstract":"The direct superimposition of a standard test grid of congruent quadrats onto an image bearing a population of particles exhibiting perspective artifacts, tends to increase the variance of the population size estimator, because the quadrat contents become unbalanced. If the quadrats are transformed according to the same projection mechanism affecting the particles, however, then the variance is restored into moderate values. Our purpose was to provide exact, easily programmable equations for the relevant transform.","PeriodicalId":49062,"journal":{"name":"Image Analysis & Stereology","volume":"22 1","pages":"47-52"},"PeriodicalIF":0.9,"publicationDate":"2020-04-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"80971314","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：