Mammographic texture features have been shown to correlate with the risk of developing breast can... more Mammographic texture features have been shown to correlate with the risk of developing breast cancer. Digital breast tomosynthesis (DBT) is an emerging 3D x-ray breast imaging modality with superior tissue visualization compared to mammography, having the potential to provide more accurate estimation of parenchymal texture features. In this paper, we investigate the effect of DBT acquisition parameters on computer-extracted texture
Medical Imaging 2008: Visualization, Image-guided Procedures, and Modeling, 2008
Software breast phantoms offer greater flexibility in generating synthetic breast images compared... more Software breast phantoms offer greater flexibility in generating synthetic breast images compared to physical phantoms. The realism of such generated synthetic images depends on the method for simulating the three-dimensional breast anatomical structures. We present here a novel algorithm for computer simulation of breast anatomy. The algorithm simulates the skin, regions of predominantly adipose tissue and fibro-glandular tissue, and the matrix of adipose tissue compartments and Cooper's ligaments. The simulation approach is based upon a region growing procedure; adipose compartments are grown from a selected set of seed points with different orientation and growth rate. The simulated adipose compartments vary in shape and size similarly to the anatomical breast variation, resulting in much improved phantom realism compared to our previous simulation based on geometric primitives. The proposed simulation also has an improved control over the breast size and glandularity. Our software breast phantom has been used in a number of applications, including breast tomosynthesis and texture analysis optimization.
... of the University of Pennsylvania, Breast Imaging Section, 1 Silverstein Building HUP, 3400 S... more ... of the University of Pennsylvania, Breast Imaging Section, 1 Silverstein Building HUP, 3400 Spruce St., Philadelphia PA 19104-4206 {Despina.Kontos,Predrag ... and/or previously diagnosed breast cancer (age range 31-80 yrs, mean age 51.4 yrs, nor-mal distribution Lilliefors test ...
... In the study, bilateral DBT, digital mammography (DM), MRI, ultrasound, and PET of the same w... more ... In the study, bilateral DBT, digital mammography (DM), MRI, ultrasound, and PET of the same women ... Sixteen women were excluded due to the existence or suspicion of bilateral cancer, or due to ... CA (1998) 9. DeFreitas, KF, Ren, B., Ruth, C., Shaw, I., Smith, AP, Stein, JA: USA ...
Topological properties of the breast ductal network have shown the potential for classifying clin... more Topological properties of the breast ductal network have shown the potential for classifying clinical breast images with and without radiological findings. In this paper, we review three methods for the description and classification of breast ductal topology. The methods are based on ramification matrices and symbolic representation employing text mining techniques.. The performance of these methods has been compared using clinical x-ray and MR images of breast ductal networks. We observed the accuracy of the classification between the ductal trees segmented from the x-ray galactograms with radiological findings and normal cases in the range of 0.86-0.91%. The accuracy of the classification of the ductal trees segmented from the MR autogalactograms was observed in the range of 0.5-0.89%.
Medical Imaging 2013: Physics of Medical Imaging, 2013
ABSTRACT Software breast phantoms have been developed for pre-clinical validation of breast imagi... more ABSTRACT Software breast phantoms have been developed for pre-clinical validation of breast imaging systems. Realism is of great importance for the acceptance and the range of applications of breast phantoms. In this paper we have assessed the phantom realism based upon the analysis of mammographic texture properties. Texture analysis is of interest since it reflects the spatial tissue distribution, which is known to correlate with breast cancer risk. We compared texture properties of synthetic mammograms generated using software breast phantoms with clinical data. A total of 133 phantom images were synthesized using software phantoms developed at the University of Pennsylvania. The phantoms were designed using two different anatomy simulation methods: an octree-based recursive partitioning method and a region growing method. The synthetic images were generated assuming a clinically used acquisition geometry and mono-energetic x-ray beam with no scatter. The clinical data included 60 anonymized mammograms selected retrospectively from screening cases at the University of Pennsylvania. The same postprocessing was applied to clinical and phantom images. The texture analysis was performed using fully automated software which extracts a battery of features from analyzed images. The histograms of texture properties extracted from phantom images were compared with those from clinical mammograms, separately for the two anatomy simulation methods. The histogram agreement was quantified using symmetrized Kulback-Leibler divergence. We observed good agreement for most of the analyzed 25 features. In more than a half of the features, the octree-based simulation method yielded better agreement with clinical data as compared with the region growing method.
2010 IEEE International Symposium on Biomedical Imaging: From Nano to Macro, 2010
In this paper we propose using histogram intersection for mammographic image classification. Firs... more In this paper we propose using histogram intersection for mammographic image classification. First, we use the bagof-words model for image representation, which captures the texture information by collecting local patch statistics. Then, we propose using normalized histogram intersection (HI) as a similarity measure with the K-nearest neighbor (KNN) classifier. Furthermore, by taking advantage of the fact that HI forms a Mercer kernel, we combine HI with support vector machines (SVM), which further improves the classification performance. The proposed methods are evaluated on a galactographic dataset and are compared with several previously used methods. In a thorough evaluation containing about 288 different experimental configurations, the proposed methods demonstrate promising results.
IEEE transactions on medical imaging, Jan 20, 2015
A modification to our previous simulation of breast anatomy is proposed to improve the quality of... more A modification to our previous simulation of breast anatomy is proposed to improve the quality of simulated x-ray projections images. The image quality is affected by the voxel size of the simulation. Large voxels can cause notable spatial quantization artifacts; small voxels extend the generation time and increase the memory requirements. An improvement in image quality is achievable without reducing voxel size by the simulation of partial volume averaging in which voxels containing more than one simulated tissue type are allowed. The linear x-ray attenuation coefficient of voxels is, thus, the sum of the linear attenuation coefficients weighted by the voxel subvolume occupied by each tissue type. A local planar approximation of the boundary surface is employed. In the twomaterial case, the partial volume in each voxel is computed by decomposition into up to four simple geometric shapes. In the three-material case, by application of the Gauss-Ostrogradsky theorem, the 3D partial vo...
... 1 University of Pennsylvania, Dept. of Radiology, Philadelphia, PA 19104 2 Real-Time Tomograp... more ... 1 University of Pennsylvania, Dept. of Radiology, Philadelphia, PA 19104 2 Real-Time Tomography, LLC, Villanova, PA19085 {Predrag.Bakic,Andrew.Maidment}@uphs.upenn.edu, {Peter.Ringer,Johnny.Kuo,Susan.Ng}@realtimetomography.com Abstract. ... Giles, UK). ...
1996 IEEE International Symposium on Circuits and Systems. Circuits and Systems Connecting the World. ISCAS 96, 1996
... Thns. Information Theory, vol. IT-20, pp. 525-5S3, July 1974. [6] Werblin, F., Roska, T., and... more ... Thns. Information Theory, vol. IT-20, pp. 525-5S3, July 1974. [6] Werblin, F., Roska, T., and Chua, LO, "The Ana-logic Cellular Neural Network as a Bionic Eye," Int. J. Cir. Theor. Appl., vol. 23, pp. 541-569, Sept./Oct. 1995. [7 ...
1 Hospital of the University of Pennsylvania, Breast Imaging Division, 1 Silverstein Building HUP... more 1 Hospital of the University of Pennsylvania, Breast Imaging Division, 1 Silverstein Building HUP, 3400 Spruce St., Philadelphia PA 19104-4206 2 University of Pennsylvania, Department of Radiology, Physics Section, 1 Silverstein Building HUP, 3400 Spruce St., Philadelphia ...
The poor specificity of galactography, the imaging modality generally indicated in cases of nippl... more The poor specificity of galactography, the imaging modality generally indicated in cases of nipple discharge, has led to a large number of biopsies with negative results. A quantitative scheme for classifying galactographic findings might help reduce the number of such biopsies in the future. As a first step toward that goal, the authors have studied one quantitative method for describing the branching of ducts by using ramification matrices (R matrices), and the correlation of the values of the matrix elements with clinical findings. The ductal trees were manually segmented for 25 galactographic views from 15 patients, and corresponding R matrices were calculated. Patients were divided into two groups: those with no reported galactographic findings (NF) and those with reported findings (RF) of ductal ectasia, cysts, or papilloma. In a leave-one-out fashion, the authors evaluated a classification scheme that was based on R-matrix coefficients and used a Bayesian decision rule. The e...
Mammographic texture features have been shown to correlate with the risk of developing breast can... more Mammographic texture features have been shown to correlate with the risk of developing breast cancer. Digital breast tomosynthesis (DBT) is an emerging 3D x-ray breast imaging modality with superior tissue visualization compared to mammography, having the potential to provide more accurate estimation of parenchymal texture features. In this paper, we investigate the effect of DBT acquisition parameters on computer-extracted texture
Medical Imaging 2008: Visualization, Image-guided Procedures, and Modeling, 2008
Software breast phantoms offer greater flexibility in generating synthetic breast images compared... more Software breast phantoms offer greater flexibility in generating synthetic breast images compared to physical phantoms. The realism of such generated synthetic images depends on the method for simulating the three-dimensional breast anatomical structures. We present here a novel algorithm for computer simulation of breast anatomy. The algorithm simulates the skin, regions of predominantly adipose tissue and fibro-glandular tissue, and the matrix of adipose tissue compartments and Cooper's ligaments. The simulation approach is based upon a region growing procedure; adipose compartments are grown from a selected set of seed points with different orientation and growth rate. The simulated adipose compartments vary in shape and size similarly to the anatomical breast variation, resulting in much improved phantom realism compared to our previous simulation based on geometric primitives. The proposed simulation also has an improved control over the breast size and glandularity. Our software breast phantom has been used in a number of applications, including breast tomosynthesis and texture analysis optimization.
... of the University of Pennsylvania, Breast Imaging Section, 1 Silverstein Building HUP, 3400 S... more ... of the University of Pennsylvania, Breast Imaging Section, 1 Silverstein Building HUP, 3400 Spruce St., Philadelphia PA 19104-4206 {Despina.Kontos,Predrag ... and/or previously diagnosed breast cancer (age range 31-80 yrs, mean age 51.4 yrs, nor-mal distribution Lilliefors test ...
... In the study, bilateral DBT, digital mammography (DM), MRI, ultrasound, and PET of the same w... more ... In the study, bilateral DBT, digital mammography (DM), MRI, ultrasound, and PET of the same women ... Sixteen women were excluded due to the existence or suspicion of bilateral cancer, or due to ... CA (1998) 9. DeFreitas, KF, Ren, B., Ruth, C., Shaw, I., Smith, AP, Stein, JA: USA ...
Topological properties of the breast ductal network have shown the potential for classifying clin... more Topological properties of the breast ductal network have shown the potential for classifying clinical breast images with and without radiological findings. In this paper, we review three methods for the description and classification of breast ductal topology. The methods are based on ramification matrices and symbolic representation employing text mining techniques.. The performance of these methods has been compared using clinical x-ray and MR images of breast ductal networks. We observed the accuracy of the classification between the ductal trees segmented from the x-ray galactograms with radiological findings and normal cases in the range of 0.86-0.91%. The accuracy of the classification of the ductal trees segmented from the MR autogalactograms was observed in the range of 0.5-0.89%.
Medical Imaging 2013: Physics of Medical Imaging, 2013
ABSTRACT Software breast phantoms have been developed for pre-clinical validation of breast imagi... more ABSTRACT Software breast phantoms have been developed for pre-clinical validation of breast imaging systems. Realism is of great importance for the acceptance and the range of applications of breast phantoms. In this paper we have assessed the phantom realism based upon the analysis of mammographic texture properties. Texture analysis is of interest since it reflects the spatial tissue distribution, which is known to correlate with breast cancer risk. We compared texture properties of synthetic mammograms generated using software breast phantoms with clinical data. A total of 133 phantom images were synthesized using software phantoms developed at the University of Pennsylvania. The phantoms were designed using two different anatomy simulation methods: an octree-based recursive partitioning method and a region growing method. The synthetic images were generated assuming a clinically used acquisition geometry and mono-energetic x-ray beam with no scatter. The clinical data included 60 anonymized mammograms selected retrospectively from screening cases at the University of Pennsylvania. The same postprocessing was applied to clinical and phantom images. The texture analysis was performed using fully automated software which extracts a battery of features from analyzed images. The histograms of texture properties extracted from phantom images were compared with those from clinical mammograms, separately for the two anatomy simulation methods. The histogram agreement was quantified using symmetrized Kulback-Leibler divergence. We observed good agreement for most of the analyzed 25 features. In more than a half of the features, the octree-based simulation method yielded better agreement with clinical data as compared with the region growing method.
2010 IEEE International Symposium on Biomedical Imaging: From Nano to Macro, 2010
In this paper we propose using histogram intersection for mammographic image classification. Firs... more In this paper we propose using histogram intersection for mammographic image classification. First, we use the bagof-words model for image representation, which captures the texture information by collecting local patch statistics. Then, we propose using normalized histogram intersection (HI) as a similarity measure with the K-nearest neighbor (KNN) classifier. Furthermore, by taking advantage of the fact that HI forms a Mercer kernel, we combine HI with support vector machines (SVM), which further improves the classification performance. The proposed methods are evaluated on a galactographic dataset and are compared with several previously used methods. In a thorough evaluation containing about 288 different experimental configurations, the proposed methods demonstrate promising results.
IEEE transactions on medical imaging, Jan 20, 2015
A modification to our previous simulation of breast anatomy is proposed to improve the quality of... more A modification to our previous simulation of breast anatomy is proposed to improve the quality of simulated x-ray projections images. The image quality is affected by the voxel size of the simulation. Large voxels can cause notable spatial quantization artifacts; small voxels extend the generation time and increase the memory requirements. An improvement in image quality is achievable without reducing voxel size by the simulation of partial volume averaging in which voxels containing more than one simulated tissue type are allowed. The linear x-ray attenuation coefficient of voxels is, thus, the sum of the linear attenuation coefficients weighted by the voxel subvolume occupied by each tissue type. A local planar approximation of the boundary surface is employed. In the twomaterial case, the partial volume in each voxel is computed by decomposition into up to four simple geometric shapes. In the three-material case, by application of the Gauss-Ostrogradsky theorem, the 3D partial vo...
... 1 University of Pennsylvania, Dept. of Radiology, Philadelphia, PA 19104 2 Real-Time Tomograp... more ... 1 University of Pennsylvania, Dept. of Radiology, Philadelphia, PA 19104 2 Real-Time Tomography, LLC, Villanova, PA19085 {Predrag.Bakic,Andrew.Maidment}@uphs.upenn.edu, {Peter.Ringer,Johnny.Kuo,Susan.Ng}@realtimetomography.com Abstract. ... Giles, UK). ...
1996 IEEE International Symposium on Circuits and Systems. Circuits and Systems Connecting the World. ISCAS 96, 1996
... Thns. Information Theory, vol. IT-20, pp. 525-5S3, July 1974. [6] Werblin, F., Roska, T., and... more ... Thns. Information Theory, vol. IT-20, pp. 525-5S3, July 1974. [6] Werblin, F., Roska, T., and Chua, LO, "The Ana-logic Cellular Neural Network as a Bionic Eye," Int. J. Cir. Theor. Appl., vol. 23, pp. 541-569, Sept./Oct. 1995. [7 ...
1 Hospital of the University of Pennsylvania, Breast Imaging Division, 1 Silverstein Building HUP... more 1 Hospital of the University of Pennsylvania, Breast Imaging Division, 1 Silverstein Building HUP, 3400 Spruce St., Philadelphia PA 19104-4206 2 University of Pennsylvania, Department of Radiology, Physics Section, 1 Silverstein Building HUP, 3400 Spruce St., Philadelphia ...
The poor specificity of galactography, the imaging modality generally indicated in cases of nippl... more The poor specificity of galactography, the imaging modality generally indicated in cases of nipple discharge, has led to a large number of biopsies with negative results. A quantitative scheme for classifying galactographic findings might help reduce the number of such biopsies in the future. As a first step toward that goal, the authors have studied one quantitative method for describing the branching of ducts by using ramification matrices (R matrices), and the correlation of the values of the matrix elements with clinical findings. The ductal trees were manually segmented for 25 galactographic views from 15 patients, and corresponding R matrices were calculated. Patients were divided into two groups: those with no reported galactographic findings (NF) and those with reported findings (RF) of ductal ectasia, cysts, or papilloma. In a leave-one-out fashion, the authors evaluated a classification scheme that was based on R-matrix coefficients and used a Bayesian decision rule. The e...
Uploads
Papers by Predrag Bakic