Paulo R. S. Mendonca
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Paulo R. S. Mendonca has a total of 95 co-authors. Here are the top co-authors by number of shared publications:
Rahul Bhotika
GE Global Research (USA)
10 shared publications- Roberto Cipolla
University of Cambridge
8 shared publications James V. Miller
GE Global Research, One Research Circle, Niskayuna, NY 12309, USA
6 shared publicationsFei Zhao
GE Global Research, Niskayuna, USA
4 shared publicationsKwan-Yee K. Wong
4 shared publicationsPeter Lamb
Research Circle, GE Global Research, Niskayuna, NY, USA
4 shared publications
Latest Publications
- 24 February 2025
- Published by Elsevier in Statistics & Probability Letters
- Vol. 222
Explicit formulae for projectively transformed Cauchy distributions with applications
- 24 February 2025
- Published by Elsevier in Statistics & Probability Letters
- Vol. 222
Explicit formulae for projectively transformed Cauchy distributions with applications
- 1 January 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1014-1023
A Self-Supervised Bootstrap Method for Single-Image 3D Face Reconstruction
State-of-the-art methods for 3D reconstruction of faces from a single image require 2D-3D pairs of ground-truth data for supervision. Such data is costly to acquire, and most datasets available in the literature are restricted to pairs for which the input 2D images depict faces in a near fronto-parallel pose. Therefore, many data-driven methods for single-image 3D facial reconstruction perform poorly on profile and near-profile faces. We propose a method to improve the performance of single-image 3D facial reconstruction networks by utilizing the network to synthesize its own training data for fine-tuning, comprising: (i) single-image 3D reconstruction of faces in near-frontal images without ground-truth 3D shape; (ii) application of a rigid-body transformation to the reconstructed face model; (iii) rendering of the face model from new viewpoints; and (iv) use of the rendered image and corresponding 3D reconstruction as additional data for supervised fine-tuning. The new 2D-3D pairs thus produced have the same high-quality observed for near fronto-parallel reconstructions, thereby nudging the network towards more uniform performance as a function of the viewing angle of input faces. Application of the proposed technique to the fine-tuning of a state-of-the-art single-image 3D-reconstruction network for faces demonstrates the usefulness of the method, with particularly significant gains for profile or near-profile views.
- 1 January 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1014-1023
A Self-Supervised Bootstrap Method for Single-Image 3D Face Reconstruction
State-of-the-art methods for 3D reconstruction of faces from a single image require 2D-3D pairs of ground-truth data for supervision. Such data is costly to acquire, and most datasets available in the literature are restricted to pairs for which the input 2D images depict faces in a near fronto-parallel pose. Therefore, many data-driven methods for single-image 3D facial reconstruction perform poorly on profile and near-profile faces. We propose a method to improve the performance of single-image 3D facial reconstruction networks by utilizing the network to synthesize its own training data for fine-tuning, comprising: (i) single-image 3D reconstruction of faces in near-frontal images without ground-truth 3D shape; (ii) application of a rigid-body transformation to the reconstructed face model; (iii) rendering of the face model from new viewpoints; and (iv) use of the rendered image and corresponding 3D reconstruction as additional data for supervised fine-tuning. The new 2D-3D pairs thus produced have the same high-quality observed for near fronto-parallel reconstructions, thereby nudging the network towards more uniform performance as a function of the viewing angle of input faces. Application of the proposed technique to the fine-tuning of a state-of-the-art single-image 3D-reconstruction network for faces demonstrates the usefulness of the method, with particularly significant gains for profile or near-profile views.
- 28 August 2014
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Medical Imaging
- Vol. 34 (3) , 807-815
Stratification of Patients With Liver Fibrosis Using Dual-Energy CT
Assessing the severity of liver fibrosis has direct clinical implications for patient diagnosis and treatment. Liver biopsy, typically considered the gold standard, has limited clinical utility due to its invasiveness. Therefore, several imaging-based techniques for staging liver fibrosis have emerged, such as magnetic resonance elastography (MRE) and ultrasound elastography (USE), but they face challenges that include limited availability, high cost, poor patient compliance, low repeatability, and inaccuracy. Computed tomography (CT) can address many of these limitations, but is still hampered by inaccuracy in the presence of confounding factors, such as liver fat. Dual-energy CT (DECT), with its ability to discriminate between different tissue types, may offer a viable alternative to these methods. By combining the "multi-material decomposition" (MMD) algorithm with a biologically driven hypothesis we developed a method for assessing liver fibrosis from DECT images. On a twelve-patient cohort the method produced quantitative maps showing the spatial distribution of liver fibrosis, as well as a fibrosis score for each patient with statistically significant correlation with the severity of fibrosis across a wide range of disease severities. A preliminary comparison of the proposed algorithm against MRE showed good agreement between the two methods. Finally, the application of the algorithm to longitudinal DECT scans of the cohort produced highly repeatable results. We conclude that our algorithm can successfully stratify patients with liver fibrosis and can serve to supplement and augment current clinical practice and the role of DECT imaging in staging liver fibrosis.
- 28 August 2014
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Medical Imaging
- Vol. 34 (3) , 807-815
Stratification of Patients With Liver Fibrosis Using Dual-Energy CT
Assessing the severity of liver fibrosis has direct clinical implications for patient diagnosis and treatment. Liver biopsy, typically considered the gold standard, has limited clinical utility due to its invasiveness. Therefore, several imaging-based techniques for staging liver fibrosis have emerged, such as magnetic resonance elastography (MRE) and ultrasound elastography (USE), but they face challenges that include limited availability, high cost, poor patient compliance, low repeatability, and inaccuracy. Computed tomography (CT) can address many of these limitations, but is still hampered by inaccuracy in the presence of confounding factors, such as liver fat. Dual-energy CT (DECT), with its ability to discriminate between different tissue types, may offer a viable alternative to these methods. By combining the "multi-material decomposition" (MMD) algorithm with a biologically driven hypothesis we developed a method for assessing liver fibrosis from DECT images. On a twelve-patient cohort the method produced quantitative maps showing the spatial distribution of liver fibrosis, as well as a fibrosis score for each patient with statistically significant correlation with the severity of fibrosis across a wide range of disease severities. A preliminary comparison of the proposed algorithm against MRE showed good agreement between the two methods. Finally, the application of the algorithm to longitudinal DECT scans of the cohort produced highly repeatable results. We conclude that our algorithm can successfully stratify patients with liver fibrosis and can serve to supplement and augment current clinical practice and the role of DECT imaging in staging liver fibrosis.
- 1 April 2014
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 19457928,p. 685-689
Topological texture-based method for mass detection in breast ultrasound image
Texture analysis plays an important role in many image processing tasks. In this work, we present a texture descriptor based on the topology of excursion sets, derived from the concept of Minkowski functionals, and evaluate their usefulness in the detection of breast masses in 2D breast ultrasound images. The application includes three major stages: preprocessing, including candidate generation through computation of gradient concentration under a Fisher-Tippet noise model (in itself another contribution of the paper); texture feature extraction; and region classification using a Random Forests classifier. Performance of the proposed method is evaluated on 135 2D BUS images with 139 masses. Our method reaches 91% sensitivity with an averaged 1.19 false detections, and the proposed texture feature compares favorably against the often-used grey level co-occurrence matrices on the exact the same task.
- 1 April 2014
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 19457928,p. 685-689
Topological texture-based method for mass detection in breast ultrasound image
Texture analysis plays an important role in many image processing tasks. In this work, we present a texture descriptor based on the topology of excursion sets, derived from the concept of Minkowski functionals, and evaluate their usefulness in the detection of breast masses in 2D breast ultrasound images. The application includes three major stages: preprocessing, including candidate generation through computation of gradient concentration under a Fisher-Tippet noise model (in itself another contribution of the paper); texture feature extraction; and region classification using a Random Forests classifier. Performance of the proposed method is evaluated on 135 2D BUS images with 139 masses. Our method reaches 91% sensitivity with an averaged 1.19 false detections, and the proposed texture feature compares favorably against the often-used grey level co-occurrence matrices on the exact the same task.
- 21 March 2014
- Published by SPIE-Intl Soc Optical Eng
- Vol. 9034, 90342N
False positive reduction of microcalcification cluster detection in digital breast tomosynthesis
Digital breast tomosynthesis (DBT) is a new modality that has strong potential in improving the sensitivity and specificity of breast mass detection. However, the detection of microcalcifications (MCs) in DBT is challenging because radiologists have to search for the often subtle signals in many slices. We are developing a computer-aided detection (CAD) system to assist radiologists in reading DBT. The system consists of four major steps, namely: image enhancement; pre-screening of MC candidates; false-positive (FP) reduction, and detection of MC cluster candidates of clinical interest. We propose an algorithm for reducing FPs by using 3D characteristics of MC clusters in DBT. The proposed method takes the MC candidates from the pre-screening step described in [14] as input, which are then iteratively clustered to provide training samples to a random-forest classifier and a rule-based classifier. The random forest classifier is used to learn a discriminative model of MC clusters using 3D texture features, whereas the rule-based classifier revisits the initial training samples and enhances them by combining median filtering and graph-cut-based segmentation followed by thresholding on the final number of MCs belonging to the candidate cluster. The outputs of these two classifiers are combined according to the prediction confidence of the random-forest classifier. We evaluate the proposed FP-reduction algorithm on a data set of two-view DBT from 40 breasts with biopsy-proven MC clusters. The experimental results demonstrate a significant reduction in FP detections, with a final sensitivity of 92.2% for an FP rate of 50%.
- 21 March 2014
- Published by SPIE-Intl Soc Optical Eng
- Vol. 9034, 90342N
False positive reduction of microcalcification cluster detection in digital breast tomosynthesis
Digital breast tomosynthesis (DBT) is a new modality that has strong potential in improving the sensitivity and specificity of breast mass detection. However, the detection of microcalcifications (MCs) in DBT is challenging because radiologists have to search for the often subtle signals in many slices. We are developing a computer-aided detection (CAD) system to assist radiologists in reading DBT. The system consists of four major steps, namely: image enhancement; pre-screening of MC candidates; false-positive (FP) reduction, and detection of MC cluster candidates of clinical interest. We propose an algorithm for reducing FPs by using 3D characteristics of MC clusters in DBT. The proposed method takes the MC candidates from the pre-screening step described in [14] as input, which are then iteratively clustered to provide training samples to a random-forest classifier and a rule-based classifier. The random forest classifier is used to learn a discriminative model of MC clusters using 3D texture features, whereas the rule-based classifier revisits the initial training samples and enhances them by combining median filtering and graph-cut-based segmentation followed by thresholding on the final number of MCs belonging to the candidate cluster. The outputs of these two classifiers are combined according to the prediction confidence of the random-forest classifier. We evaluate the proposed FP-reduction algorithm on a data set of two-view DBT from 40 breasts with biopsy-proven MC clusters. The experimental results demonstrate a significant reduction in FP detections, with a final sensitivity of 92.2% for an FP rate of 50%.
- 16 September 2013
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Medical Imaging
- Vol. 33 (1) , 99-116
A Flexible Method for Multi-Material Decomposition of Dual-Energy CT Images
The ability of dual-energy computed-tomographic (CT) systems to determine the concentration of constituent materials in a mixture, known as material decomposition, is the basis for many of dual-energy CT's clinical applications. However, the complex composition of tissues and organs in the human body poses a challenge for many material decomposition methods, which assume the presence of only two, or at most three, materials in the mixture. We developed a flexible, model-based method that extends dual-energy CT's core material decomposition capability to handle more complex situations, in which it is necessary to disambiguate among and quantify the concentration of a larger number of materials. The proposed method, named multi-material decomposition (MMD), was used to develop two image analysis algorithms. The first was virtual unenhancement (VUE), which digitally removes the effect of contrast agents from contrast-enhanced dual-energy CT exams. VUE has the ability to reduce patient dose and improve clinical workflow, and can be used in a number of clinical applications such as CT urography and CT angiography. The second algorithm developed was liver-fat quantification (LFQ), which accurately quantifies the fat concentration in the liver from dual-energy CT exams. LFQ can form the basis of a clinical application targeting the diagnosis and treatment of fatty liver disease. Using image data collected from a cohort consisting of 50 patients and from phantoms, the application of MMD to VUE and LFQ yielded quantitatively accurate results when compared against gold standards. Furthermore, consistent results were obtained across all phases of imaging (contrast-free and contrast-enhanced). This is of particular importance since most clinical protocols for abdominal imaging with CT call for multi-phase imaging. We conclude that MMD can successfully form the basis of a number of dual-energy CT image analysis algorithms, and has the potential to improve the clinical utility of dual-energy CT in disease management.
- 16 September 2013
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Medical Imaging
- Vol. 33 (1) , 99-116
A Flexible Method for Multi-Material Decomposition of Dual-Energy CT Images
The ability of dual-energy computed-tomographic (CT) systems to determine the concentration of constituent materials in a mixture, known as material decomposition, is the basis for many of dual-energy CT's clinical applications. However, the complex composition of tissues and organs in the human body poses a challenge for many material decomposition methods, which assume the presence of only two, or at most three, materials in the mixture. We developed a flexible, model-based method that extends dual-energy CT's core material decomposition capability to handle more complex situations, in which it is necessary to disambiguate among and quantify the concentration of a larger number of materials. The proposed method, named multi-material decomposition (MMD), was used to develop two image analysis algorithms. The first was virtual unenhancement (VUE), which digitally removes the effect of contrast agents from contrast-enhanced dual-energy CT exams. VUE has the ability to reduce patient dose and improve clinical workflow, and can be used in a number of clinical applications such as CT urography and CT angiography. The second algorithm developed was liver-fat quantification (LFQ), which accurately quantifies the fat concentration in the liver from dual-energy CT exams. LFQ can form the basis of a clinical application targeting the diagnosis and treatment of fatty liver disease. Using image data collected from a cohort consisting of 50 patients and from phantoms, the application of MMD to VUE and LFQ yielded quantitatively accurate results when compared against gold standards. Furthermore, consistent results were obtained across all phases of imaging (contrast-free and contrast-enhanced). This is of particular importance since most clinical protocols for abdominal imaging with CT call for multi-phase imaging. We conclude that MMD can successfully form the basis of a number of dual-energy CT image analysis algorithms, and has the potential to improve the clinical utility of dual-energy CT in disease management.
- 1 September 2013
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 15224880,p. 2762-2766
Learning-based automatic defect recognition with computed tomographic imaging
The use of image-based automatic defect recognition (ADR) systems in a production line often requires strict processing-time specifications. On the other hand, the typical high-performance requirement of such system calls for the use of sophisticated, computationally-complex algorithms. Addressing the conflicting requirements of fast throughput and high detection performance is a significant challenge. In this paper we present a 3D learning-based ADR approach for industrial parts. The proposed method first extracts defect candidate regions using morphological closing and template matching. Then a local registration-based approach is utilized to produce accurate defect segmentation mask. Finally, 29 features including geometric features and texture features derived from grey level co-occurrence matrix are calculated for each candidate region, and a fast random forests classifier is used to classify the candidate regions as defect or defect-free. This approach was developed into a fully automated system for detecting casting defects in aluminum industrial parts depicted in 3D Computed Tomographic (CT) images. The system was tested on 31 images with 49 cavities and porosities defects, achieving a sensitivity of 94% with an average 3.5 false detections per part.
- 1 September 2013
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 15224880,p. 2762-2766
Learning-based automatic defect recognition with computed tomographic imaging
The use of image-based automatic defect recognition (ADR) systems in a production line often requires strict processing-time specifications. On the other hand, the typical high-performance requirement of such system calls for the use of sophisticated, computationally-complex algorithms. Addressing the conflicting requirements of fast throughput and high detection performance is a significant challenge. In this paper we present a 3D learning-based ADR approach for industrial parts. The proposed method first extracts defect candidate regions using morphological closing and template matching. Then a local registration-based approach is utilized to produce accurate defect segmentation mask. Finally, 29 features including geometric features and texture features derived from grey level co-occurrence matrix are calculated for each candidate region, and a fast random forests classifier is used to classify the candidate regions as defect or defect-free. This approach was developed into a fully automated system for detecting casting defects in aluminum industrial parts depicted in 3D Computed Tomographic (CT) images. The system was tested on 31 images with 49 cavities and porosities defects, achieving a sensitivity of 94% with an average 3.5 false detections per part.
- 1 January 2013
- Published by Springer Nature in Lecture Notes in Computer Science
- Vol. 16, 526-533
Tracking of Carotid Arteries in Ultrasound Images
We introduce an automated method for the 3D tracking of carotids acquired as a sequence of 2D ultrasound images. The method includes an image stabilization step that compensates for the cardiac and respiratory motion of the carotid, and tracks the carotid wall via a shape and appearance model trained from representative images. Envisaging an application in automatic detection of plaques, the algorithm was tested on ultrasound volumes from 4,000 patients and its accuracy was evaluated by measuring the distance between the location of more than 4,000 carotid plaques and the location of the carotid wall as estimated by the proposed algorithm. Results show that the centroids of over 95% of the carotid plaques in the dataset were located within 3 mm of the estimated carotid wall, indicating the accuracy of the tracking algorithm.
- 1 January 2013
- Published by Springer Nature in Lecture Notes in Computer Science
- Vol. 16, 526-533
Tracking of Carotid Arteries in Ultrasound Images
We introduce an automated method for the 3D tracking of carotids acquired as a sequence of 2D ultrasound images. The method includes an image stabilization step that compensates for the cardiac and respiratory motion of the carotid, and tracks the carotid wall via a shape and appearance model trained from representative images. Envisaging an application in automatic detection of plaques, the algorithm was tested on ultrasound volumes from 4,000 patients and its accuracy was evaluated by measuring the distance between the location of more than 4,000 carotid plaques and the location of the carotid wall as estimated by the proposed algorithm. Results show that the centroids of over 95% of the carotid plaques in the dataset were located within 3 mm of the estimated carotid wall, indicating the accuracy of the tracking algorithm.
- 1 January 2013
- Published by Springer Nature in Lecture Notes in Computer Science
- Vol. 16, 743-750
Deformable Atlas for Multi-structure Segmentation
We develop a novel deformable atlas method for multi-structure segmentation that seamlessly combines the advantages of image-based and atlas-based methods. The method formulates a probabilistic framework that combines prior anatomical knowledge with image-based cues that are specific to the subject’s anatomy, and solves it using expectation-maximization method. It improves the segmentation over conventional label fusion methods especially around the structure boundaries, and is robust to large anatomical variation. The proposed method was applied to segment multiple structures in both normal and diseased brains and was shown to significantly improve results especially in diseased brains.
- 1 January 2013
- Published by Springer Nature in Lecture Notes in Computer Science
- Vol. 16, 743-750
Deformable Atlas for Multi-structure Segmentation
We develop a novel deformable atlas method for multi-structure segmentation that seamlessly combines the advantages of image-based and atlas-based methods. The method formulates a probabilistic framework that combines prior anatomical knowledge with image-based cues that are specific to the subject’s anatomy, and solves it using expectation-maximization method. It improves the segmentation over conventional label fusion methods especially around the structure boundaries, and is robust to large anatomical variation. The proposed method was applied to segment multiple structures in both normal and diseased brains and was shown to significantly improve results especially in diseased brains.
- 1 January 2013
- Published by Springer Nature
- Vol. 16, 324-331
Contrast-Independent Liver-Fat Quantification from Spectral CT Exams
The diagnosis and treatment of fatty liver disease requires accurate quantification of the amount of fat in the liver. Image-based methods for quantification of liver fat are of increasing interest due to the high sampling error and invasiveness associated with liver biopsy, which despite these difficulties remains the gold standard. Current computed tomography (CT) methods for liverfat quantification are only semi-quantitative and infer the concentration of liver fat heuristically. Furthermore, these techniques are only applicable to images acquired without the use of contrast agent, even though contrast-enhanced CT imaging is more prevalent in clinical practice. In this paper, we introduce a method that allows for direct quantification of liver fat for both contrast-free and contrastenhanced CT images. Phantom and patient data are used for validation, and we conclude that our algorithm allows for highly accurate and repeatable quantification of liver fat for spectral CT.
- 1 January 2013
- Published by Springer Nature
- Vol. 16, 324-331
Contrast-Independent Liver-Fat Quantification from Spectral CT Exams
The diagnosis and treatment of fatty liver disease requires accurate quantification of the amount of fat in the liver. Image-based methods for quantification of liver fat are of increasing interest due to the high sampling error and invasiveness associated with liver biopsy, which despite these difficulties remains the gold standard. Current computed tomography (CT) methods for liverfat quantification are only semi-quantitative and infer the concentration of liver fat heuristically. Furthermore, these techniques are only applicable to images acquired without the use of contrast agent, even though contrast-enhanced CT imaging is more prevalent in clinical practice. In this paper, we introduce a method that allows for direct quantification of liver fat for both contrast-free and contrastenhanced CT images. Phantom and patient data are used for validation, and we conclude that our algorithm allows for highly accurate and repeatable quantification of liver fat for spectral CT.