Analyzing the blood oxygenation level dependent (BOLD) effect in the functional magnetic resonanc... more Analyzing the blood oxygenation level dependent (BOLD) effect in the functional magnetic resonance imaging (fMRI) is typically based on recent ground-breaking time series analysis techniques. This work represents a significant improvement over existing approaches to system identification using nonlinear hemodynamic models. It is important for three reasons. First, instead of using linearized approximations of the dynamics, we present a nonlinear filtering based on the sequential Monte Carlo method to capture the inherent nonlinearities in the physiological system. Second, we simultaneously estimate the hidden physiological states and the system parameters through particle filtering with sequential parameter learning to fully take advantage of the dynamic information of the BOLD signals. Third, during the unknown static parameter learning, we employ the low-dimensional sufficient statistics for efficiency and avoiding potential degeneration of the parameters. The performance of the p...
Proceedings. 1998 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (Cat. No.98CB36231), 1998
We propose an approach for boundary finding where the correspondence of a subset of boundary poin... more We propose an approach for boundary finding where the correspondence of a subset of boundary points to a model is simultaneously determined. Global shape parameters derived from the statistical variation of object boundary points in a training set are used to model the object. A Bayesian formulation, based on this prior knowledge and the edge information of the input image, is employed to find the object boundary with its subset points in correspondence with boundaries in the training set or the mean boundary. We compared the use of a generic smoothness prior and a uniform independent prior with the training set prior in order to demonstrate the power of this statistical information. A number of experiments were performed on both synthetic and real medical images of the brain and heart to evaluate the approach, including the validation of the dependence of the method on image quality, different initialization and prior information.
This paper presents a new and general nonlinear framework for fMRI data analysis based on statist... more This paper presents a new and general nonlinear framework for fMRI data analysis based on statistical learning methodology: support vector machines. Unlike most current methods which assume a linear model for simplicity, the estimation and analysis of fMRI signal within the proposed framework is nonlinear, which matches recent findings on the dynamics underlying neural activity and hemodynamic physiology. The approach utilizes spatio-temporal support vector regression (SVR), within which the intrinsic spatio-temporal autocorrelations in fMRI data are reflected. The novel formulation of the problem allows merging model-driven with data-driven methods, and therefore unifies these two currently separate modes of fMRI analysis. In addition, multiresolution signal analysis is achieved and developed. Other advantages of the approach are: avoidance of interpolation after motion estimation, embedded removal of low-frequency noise components, and easy incorporation of multi-run, multi-subject, and multi-task studies into the framework.
This paper describes a new method of non-rigid registration using the combined power of elastic a... more This paper describes a new method of non-rigid registration using the combined power of elastic and statistical shape models. The transformations are constrained to be consistent with a physical model of elasticity to maintain smoothness and continuity. A Bayesian formulation, based on this model, on an intensity similarity measure, and on statistical shape information embedded in corresponding boundary points, is employed to find a more accurate and robust non-rigid registration. A dense set of forces arises from the intensity similarity measure to accommodate complex anatomical details. A sparse set of forces constrains consistency with statistical shape models derived from a training set. A number of experiments were performed on both synthetic and real medical images of the brain and heart to evaluate the approach. It is shown that statistical boundary shape information significantly augments and improves elastic model based non-rigid registration. 1
Recent advances in functional magnetic resonance imaging (fMRI) provide an unparalleled opportuni... more Recent advances in functional magnetic resonance imaging (fMRI) provide an unparalleled opportunity for measuring and characterizing brain function in humans. However, the typically small signal change is very noisy and susceptible to various artifacts, such as those caused by scanner drift, head motion, and cardio-respiratory effects. This paper presents an integrated and exploratory approach to characterize brain function from fMRI
Advances in neural information processing systems, 2009
In this paper, we develop an efficient moments-based permutation test approach to improve the tes... more In this paper, we develop an efficient moments-based permutation test approach to improve the test's computational efficiency by approximating the permutation distribution of the test statistic with Pearson distribution series. This approach involves the calculation of the first four moments of the permutation distribution. We propose a novel recursive method to derive these moments theoretically and analytically without any permutation. Experimental results using different test statistics are demonstrated using simulated data and real data. The proposed strategy takes advantage of nonparametric permutation tests and parametric Pearson distribution approximation to achieve both accuracy and efficiency.
This paper presents novel statistical methods for estimating brain networks from fMRI data. Funct... more This paper presents novel statistical methods for estimating brain networks from fMRI data. Functional interactions are detected by simultaneously examining multi-seed correlations via multiple correlation coefficients. Spatially structured noise in fMRI is also taken into account during the identification of functional interconnection networks through non-central F hypothesis tests. Furthermore, partial multiple correlations are introduced and formulated to measure any additional task-induced but not stimulus-locked relation over brain regions so that we can take the analysis of functional connectivity closer to the characterization of direct functional interactions of the brain. Evaluation for accuracy and advantages of the new approaches and comparison with the existing single-seed method were performed extensively using both simulated data and real fMRI data.
2009 IEEE International Symposium on Biomedical Imaging: From Nano to Macro, 2009
ABSTRACT This paper presents a novel spatial Bayesian method for simultaneous activation detectio... more ABSTRACT This paper presents a novel spatial Bayesian method for simultaneous activation detection and hemodynamic response function (HRF) estimation of functional magnetic resonance imaging (fMRI) data. A Bayesian variable selection approach is used to induce shrinkage and sparsity, with a spatial prior on latent variables representing activated hemodynamic response components. Then, the activation map is generated from the full spectrum of posterior inference constructed through a Markov chain Monte Carlo scheme, and HRFs at different voxels are estimated non-parametrically with information pooling from neighboring voxels. By integrating functional activation detection and HRFs estimation in a unified framework, our method is more robust to noise and less sensitive to model mis-specification.
International journal of statistics in medical research, Jan 30, 2014
In this paper, we present a new blockwise permutation test approach based on the moments of the t... more In this paper, we present a new blockwise permutation test approach based on the moments of the test statistic. The method is of importance to neuroimaging studies. In order to preserve the exchangeability condition required in permutation tests, we divide the entire set of data into certain exchangeability blocks. In addition, computationally efficient moments-based permutation tests are performed by approximating the permutation distribution of the test statistic with the Pearson distribution series. This involves the calculation of the first four moments of the permutation distribution within each block and then over the entire set of data. The accuracy and efficiency of the proposed method are demonstrated through simulated experiment on the magnetic resonance imaging (MRI) brain data, specifically the multi-site voxel-based morphometry analysis from structural MRI (sMRI).
Outlier detection is a primary step in many data mining and analysis applications, including heal... more Outlier detection is a primary step in many data mining and analysis applications, including healthcare and medical research. This paper presents a general method to identify outliers in multivariate time series based on a Voronoi diagram, which we call Multivariate Voronoi Outlier Detection (MVOD). The approach copes with outliers in a multivariate framework, via designing and extracting effective attributes or features from the data that can take parametric or nonparametric forms. Voronoi diagrams allow for automatic configuration of the neighborhood relationship of the data points, which facilitates the differentiation of outliers and non-outliers. Experimental evaluation demonstrates that our MVOD is an accurate, sensitive, and robust method for detecting outliers in multivariate time series data.
Page 1. 17 Methods for Nonrigid Image Registration Lawrence H. Staib1 and YongmeiMichelle Wang2 .... more Page 1. 17 Methods for Nonrigid Image Registration Lawrence H. Staib1 and YongmeiMichelle Wang2 ... Page 2. 572 Lawrence H. Staib and Yongmei Michelle Wang (MR) and computed tomography (CT) brain images to an atlas. ...
2008 International Conference on BioMedical Engineering and Informatics, 2008
A new statistical permutation analysis method is presented in this paper to efficiently and accur... more A new statistical permutation analysis method is presented in this paper to efficiently and accurately localize regionally specific shape differences between groups of 3D biomedical images. It can improve the system's efficiency by approximating the permutation distribution of the test statistic with Pearson distribution series. This procedure involves the calculation of the first four moments of the permutation distribution, which are derived theoretically and analytically without any permutation. Furthermore, bioequivalence testing aims for practical significances between the two groups that are statistically significant with the shape differences larger than a desired threshold. Experimental results based on both classical and bioequivalence hypothesis tests using simulated data and real biomedical images are presented to demonstrate the advantages of the proposed approach.
Proceedings IEEE Workshop on Mathematical Methods in Biomedical Image Analysis (MMBIA 2001), 2001
Image interpolation is of great importance in biomedical visualization and analysis. In this pape... more Image interpolation is of great importance in biomedical visualization and analysis. In this paper, we present a novel gray-level interpolation method called Directional Coherence Interpolation (DCI). The principal advantage of the proposed approach is that it leads to significantly higher visual quality in 3D rendering when compared with traditional image interpolation methods. The basis of DCI is a form of directional image-space coherence. DCI interpolates the missing image data along the maximum coherence directions (MCD), which are estimated from the local image intensity yet constrained by a generic smoothness term. Since the edges of the image and the contents of the objects are well preserved along the MCDs, DCI can incorporate image shape and structure information without the prior requirement of explicit representation of object boundary / surface. A number of experiments were performed on both synthetic and real medical images to evaluate the proposed approach. The experimental results show that in addition to the substantial improvement of visual effects (qualitative evaluation), the quantitative error measures of DCI are also better than the conventional gray level linear interpolation. Comparing with the shape-based interpolation scheme applied on gray-level images, DCI has much lower computation cost.
Proceedings IEEE International Symposium on Biomedical Imaging
A multiresolution technique for biomedical image interpolation is presented in this paper. It is ... more A multiresolution technique for biomedical image interpolation is presented in this paper. It is an extension of the work on Directional Coherence Interpolation (DCI) [15]-a novel gray-level image interpolation method that interpolates the missing image data along the smoothed Maximum Coherence Directions (MCDs). We propose to apply a pyramidal search strategy for MCD estimation. This coarse-to-fine scheme requires less computation time by starting with the reduced amount of data and propagating searching results to finer resolutions. In addition, it also improves robustness compared with our previous single resolution DCI.
ABSTRACT Deformable models for medical image segmentation are often enhanced by their use of prio... more ABSTRACT Deformable models for medical image segmentation are often enhanced by their use of prior shape information. Some problems are well suited to the constraints that global shape infor-mation provides, where the shapes of the organs or structures are very consistent and are well characterized by a specific shape model. Other problems involve structures whose shapes are highly variable or have no consistent shape at all and thus require more generic shape infor-mation. We describe approaches to these two types of segmentation problems illustrating the varying uses of shape information. We describe approaches in a maximum a posteriori formu-lation using parametric models with associated probability densities as a way to incorporate specific shape information. We show different forms of prior information and how to combine specific and generic information in this framework. We describe level set methods which can incorporate powerful generic shape constraints, in particular, a thickness constraint. We discuss the development of these ideas, and illustrate these approaches with examples from images of the heart and brain.
2008 IEEE International Conference on Acoustics, Speech and Signal Processing, 2008
It is a fascinating yet challenging problem to accurately and efficiently localize regionally dis... more It is a fascinating yet challenging problem to accurately and efficiently localize regionally distinct features between face groups in multi-dimensional signal processing and analysis. Given a data with unknown distribution and small sample size, we propose a new statistical analysis framework using hybrid randomization (i.e., permutation) tests to improve the system's efficiency in identifying distinct features. The proposed method fits the nonparametric distribution of the test statistic with Pearson distribution series. We bypass the tedious online randomization via calculating the first four moments of the permutation distribution. This can reduce the computational complexity from O(n!) to O(n 2) over traditional methods for the modified Hotelling's T 2 test statistics. Experiments on simulated data and 3D face analysis demonstrate the efficiency, accuracy and robustness of the proposed approach.
Analyzing the blood oxygenation level dependent (BOLD) effect in the functional magnetic resonanc... more Analyzing the blood oxygenation level dependent (BOLD) effect in the functional magnetic resonance imaging (fMRI) is typically based on recent ground-breaking time series analysis techniques. This work represents a significant improvement over existing approaches to system identification using nonlinear hemodynamic models. It is important for three reasons. First, instead of using linearized approximations of the dynamics, we present a nonlinear filtering based on the sequential Monte Carlo method to capture the inherent nonlinearities in the physiological system. Second, we simultaneously estimate the hidden physiological states and the system parameters through particle filtering with sequential parameter learning to fully take advantage of the dynamic information of the BOLD signals. Third, during the unknown static parameter learning, we employ the low-dimensional sufficient statistics for efficiency and avoiding potential degeneration of the parameters. The performance of the p...
Proceedings. 1998 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (Cat. No.98CB36231), 1998
We propose an approach for boundary finding where the correspondence of a subset of boundary poin... more We propose an approach for boundary finding where the correspondence of a subset of boundary points to a model is simultaneously determined. Global shape parameters derived from the statistical variation of object boundary points in a training set are used to model the object. A Bayesian formulation, based on this prior knowledge and the edge information of the input image, is employed to find the object boundary with its subset points in correspondence with boundaries in the training set or the mean boundary. We compared the use of a generic smoothness prior and a uniform independent prior with the training set prior in order to demonstrate the power of this statistical information. A number of experiments were performed on both synthetic and real medical images of the brain and heart to evaluate the approach, including the validation of the dependence of the method on image quality, different initialization and prior information.
This paper presents a new and general nonlinear framework for fMRI data analysis based on statist... more This paper presents a new and general nonlinear framework for fMRI data analysis based on statistical learning methodology: support vector machines. Unlike most current methods which assume a linear model for simplicity, the estimation and analysis of fMRI signal within the proposed framework is nonlinear, which matches recent findings on the dynamics underlying neural activity and hemodynamic physiology. The approach utilizes spatio-temporal support vector regression (SVR), within which the intrinsic spatio-temporal autocorrelations in fMRI data are reflected. The novel formulation of the problem allows merging model-driven with data-driven methods, and therefore unifies these two currently separate modes of fMRI analysis. In addition, multiresolution signal analysis is achieved and developed. Other advantages of the approach are: avoidance of interpolation after motion estimation, embedded removal of low-frequency noise components, and easy incorporation of multi-run, multi-subject, and multi-task studies into the framework.
This paper describes a new method of non-rigid registration using the combined power of elastic a... more This paper describes a new method of non-rigid registration using the combined power of elastic and statistical shape models. The transformations are constrained to be consistent with a physical model of elasticity to maintain smoothness and continuity. A Bayesian formulation, based on this model, on an intensity similarity measure, and on statistical shape information embedded in corresponding boundary points, is employed to find a more accurate and robust non-rigid registration. A dense set of forces arises from the intensity similarity measure to accommodate complex anatomical details. A sparse set of forces constrains consistency with statistical shape models derived from a training set. A number of experiments were performed on both synthetic and real medical images of the brain and heart to evaluate the approach. It is shown that statistical boundary shape information significantly augments and improves elastic model based non-rigid registration. 1
Recent advances in functional magnetic resonance imaging (fMRI) provide an unparalleled opportuni... more Recent advances in functional magnetic resonance imaging (fMRI) provide an unparalleled opportunity for measuring and characterizing brain function in humans. However, the typically small signal change is very noisy and susceptible to various artifacts, such as those caused by scanner drift, head motion, and cardio-respiratory effects. This paper presents an integrated and exploratory approach to characterize brain function from fMRI
Advances in neural information processing systems, 2009
In this paper, we develop an efficient moments-based permutation test approach to improve the tes... more In this paper, we develop an efficient moments-based permutation test approach to improve the test's computational efficiency by approximating the permutation distribution of the test statistic with Pearson distribution series. This approach involves the calculation of the first four moments of the permutation distribution. We propose a novel recursive method to derive these moments theoretically and analytically without any permutation. Experimental results using different test statistics are demonstrated using simulated data and real data. The proposed strategy takes advantage of nonparametric permutation tests and parametric Pearson distribution approximation to achieve both accuracy and efficiency.
This paper presents novel statistical methods for estimating brain networks from fMRI data. Funct... more This paper presents novel statistical methods for estimating brain networks from fMRI data. Functional interactions are detected by simultaneously examining multi-seed correlations via multiple correlation coefficients. Spatially structured noise in fMRI is also taken into account during the identification of functional interconnection networks through non-central F hypothesis tests. Furthermore, partial multiple correlations are introduced and formulated to measure any additional task-induced but not stimulus-locked relation over brain regions so that we can take the analysis of functional connectivity closer to the characterization of direct functional interactions of the brain. Evaluation for accuracy and advantages of the new approaches and comparison with the existing single-seed method were performed extensively using both simulated data and real fMRI data.
2009 IEEE International Symposium on Biomedical Imaging: From Nano to Macro, 2009
ABSTRACT This paper presents a novel spatial Bayesian method for simultaneous activation detectio... more ABSTRACT This paper presents a novel spatial Bayesian method for simultaneous activation detection and hemodynamic response function (HRF) estimation of functional magnetic resonance imaging (fMRI) data. A Bayesian variable selection approach is used to induce shrinkage and sparsity, with a spatial prior on latent variables representing activated hemodynamic response components. Then, the activation map is generated from the full spectrum of posterior inference constructed through a Markov chain Monte Carlo scheme, and HRFs at different voxels are estimated non-parametrically with information pooling from neighboring voxels. By integrating functional activation detection and HRFs estimation in a unified framework, our method is more robust to noise and less sensitive to model mis-specification.
International journal of statistics in medical research, Jan 30, 2014
In this paper, we present a new blockwise permutation test approach based on the moments of the t... more In this paper, we present a new blockwise permutation test approach based on the moments of the test statistic. The method is of importance to neuroimaging studies. In order to preserve the exchangeability condition required in permutation tests, we divide the entire set of data into certain exchangeability blocks. In addition, computationally efficient moments-based permutation tests are performed by approximating the permutation distribution of the test statistic with the Pearson distribution series. This involves the calculation of the first four moments of the permutation distribution within each block and then over the entire set of data. The accuracy and efficiency of the proposed method are demonstrated through simulated experiment on the magnetic resonance imaging (MRI) brain data, specifically the multi-site voxel-based morphometry analysis from structural MRI (sMRI).
Outlier detection is a primary step in many data mining and analysis applications, including heal... more Outlier detection is a primary step in many data mining and analysis applications, including healthcare and medical research. This paper presents a general method to identify outliers in multivariate time series based on a Voronoi diagram, which we call Multivariate Voronoi Outlier Detection (MVOD). The approach copes with outliers in a multivariate framework, via designing and extracting effective attributes or features from the data that can take parametric or nonparametric forms. Voronoi diagrams allow for automatic configuration of the neighborhood relationship of the data points, which facilitates the differentiation of outliers and non-outliers. Experimental evaluation demonstrates that our MVOD is an accurate, sensitive, and robust method for detecting outliers in multivariate time series data.
Page 1. 17 Methods for Nonrigid Image Registration Lawrence H. Staib1 and YongmeiMichelle Wang2 .... more Page 1. 17 Methods for Nonrigid Image Registration Lawrence H. Staib1 and YongmeiMichelle Wang2 ... Page 2. 572 Lawrence H. Staib and Yongmei Michelle Wang (MR) and computed tomography (CT) brain images to an atlas. ...
2008 International Conference on BioMedical Engineering and Informatics, 2008
A new statistical permutation analysis method is presented in this paper to efficiently and accur... more A new statistical permutation analysis method is presented in this paper to efficiently and accurately localize regionally specific shape differences between groups of 3D biomedical images. It can improve the system's efficiency by approximating the permutation distribution of the test statistic with Pearson distribution series. This procedure involves the calculation of the first four moments of the permutation distribution, which are derived theoretically and analytically without any permutation. Furthermore, bioequivalence testing aims for practical significances between the two groups that are statistically significant with the shape differences larger than a desired threshold. Experimental results based on both classical and bioequivalence hypothesis tests using simulated data and real biomedical images are presented to demonstrate the advantages of the proposed approach.
Proceedings IEEE Workshop on Mathematical Methods in Biomedical Image Analysis (MMBIA 2001), 2001
Image interpolation is of great importance in biomedical visualization and analysis. In this pape... more Image interpolation is of great importance in biomedical visualization and analysis. In this paper, we present a novel gray-level interpolation method called Directional Coherence Interpolation (DCI). The principal advantage of the proposed approach is that it leads to significantly higher visual quality in 3D rendering when compared with traditional image interpolation methods. The basis of DCI is a form of directional image-space coherence. DCI interpolates the missing image data along the maximum coherence directions (MCD), which are estimated from the local image intensity yet constrained by a generic smoothness term. Since the edges of the image and the contents of the objects are well preserved along the MCDs, DCI can incorporate image shape and structure information without the prior requirement of explicit representation of object boundary / surface. A number of experiments were performed on both synthetic and real medical images to evaluate the proposed approach. The experimental results show that in addition to the substantial improvement of visual effects (qualitative evaluation), the quantitative error measures of DCI are also better than the conventional gray level linear interpolation. Comparing with the shape-based interpolation scheme applied on gray-level images, DCI has much lower computation cost.
Proceedings IEEE International Symposium on Biomedical Imaging
A multiresolution technique for biomedical image interpolation is presented in this paper. It is ... more A multiresolution technique for biomedical image interpolation is presented in this paper. It is an extension of the work on Directional Coherence Interpolation (DCI) [15]-a novel gray-level image interpolation method that interpolates the missing image data along the smoothed Maximum Coherence Directions (MCDs). We propose to apply a pyramidal search strategy for MCD estimation. This coarse-to-fine scheme requires less computation time by starting with the reduced amount of data and propagating searching results to finer resolutions. In addition, it also improves robustness compared with our previous single resolution DCI.
ABSTRACT Deformable models for medical image segmentation are often enhanced by their use of prio... more ABSTRACT Deformable models for medical image segmentation are often enhanced by their use of prior shape information. Some problems are well suited to the constraints that global shape infor-mation provides, where the shapes of the organs or structures are very consistent and are well characterized by a specific shape model. Other problems involve structures whose shapes are highly variable or have no consistent shape at all and thus require more generic shape infor-mation. We describe approaches to these two types of segmentation problems illustrating the varying uses of shape information. We describe approaches in a maximum a posteriori formu-lation using parametric models with associated probability densities as a way to incorporate specific shape information. We show different forms of prior information and how to combine specific and generic information in this framework. We describe level set methods which can incorporate powerful generic shape constraints, in particular, a thickness constraint. We discuss the development of these ideas, and illustrate these approaches with examples from images of the heart and brain.
2008 IEEE International Conference on Acoustics, Speech and Signal Processing, 2008
It is a fascinating yet challenging problem to accurately and efficiently localize regionally dis... more It is a fascinating yet challenging problem to accurately and efficiently localize regionally distinct features between face groups in multi-dimensional signal processing and analysis. Given a data with unknown distribution and small sample size, we propose a new statistical analysis framework using hybrid randomization (i.e., permutation) tests to improve the system's efficiency in identifying distinct features. The proposed method fits the nonparametric distribution of the test statistic with Pearson distribution series. We bypass the tedious online randomization via calculating the first four moments of the permutation distribution. This can reduce the computational complexity from O(n!) to O(n 2) over traditional methods for the modified Hotelling's T 2 test statistics. Experiments on simulated data and 3D face analysis demonstrate the efficiency, accuracy and robustness of the proposed approach.
Uploads
Papers by Test Account