Search

U-Net, as an encoder-decoder architecture with forward skip connections, has achieved promising results in various medical image analysis tasks. Many recent approaches have also extended U-Net with more complex building blocks, which typically increase the number of network parameters considerably. Such complexity makes the inference stage highly inefficient for clinical applications. Towards an effective yet economic segmentation network design, in this work, we propose backward skip connections that bring decoded features back to the encoder. Our design can be jointly adopted with forward skip connections in any encoder-decoder architecture forming a recurrence structure without introducing extra parameters. With the backward skip connections, we propose a U-Net based network family, namely Bi-directional O-shape networks, which set new benchmarks on multiple public medical imaging segmentation datasets. On the other hand, with the most plain architecture (BiO-Net), network computations inevitably increase along with the pre-set recurrence time. We have thus studied the deficiency bottleneck of such recurrent design and propose a novel two-phase Neural Architecture Search (NAS) algorithm, namely BiX-NAS, to search for the best multi-scale bi-directional skip connections. The ineffective skip connections are then discarded to reduce computational costs and speed up network inference. The finally searched BiX-Net yields the least network complexity and outperforms other state-of-the-art counterparts by large margins. We evaluate our methods on both 2D and 3D segmentation tasks in a total of six datasets. Extensive ablation studies have also been conducted to provide a comprehensive analysis for our proposed methods., Comment: Medical Image Analysis 2022

Simultaneous contrast enhancement and speckle suppression in optical coherence tomography (OCT) are of great significance to medical diagnosis. In this paper, we propose a selective weighted variational enhancement (SWVE) model to enhance the structural parts of OCT images, and then present a shape-preserving fourth-order-oriented partial differential equations (SP-FOOPDE) algorithm to suppress speckle noise. To be specific, in the SWVE model, we first introduce the fast and robust fuzzy c-means clustering (FRFCM) algorithm to generate masks based on the gray-level histograms of the reconstructed OCT images and utilize the masks to distinguish the structural parts from the background. Then the retinex-based weighted variational model, combined with gamma correction, is adopted to enhance the structural parts by multiplying the estimated reflectance with the adjusted illumination. In the despeckling process, we present an SP-FOOPDE algorithm with the fidelity term modified by the shearlet transform to strike a splendid balance between noise suppression and structural preservation. Experimental results show that the proposed method performs well in contrast enhancement and speckle suppression, with better quality metrics of the MSE, PSNR, CNR, ENL, EKI, and ν and better noise immunity than the related method. Moreover, the application to the segmentation preprocessing exhibits that the retinal structure of the OCT images processed by the proposed method can be completely segmented.

Gene expressions profiling empowers many biological studies in various fields by comprehensive characterization of cellular status under different experimental conditions. Despite the recent advances in high-throughput technologies, profiling the whole-genome set is still challenging and expensive. Based on the fact that there is high correlation among the expression patterns of different genes, the above issue can be addressed by a cost-effective approach that collects only a small subset of genes, called landmark genes, as the representative of the entire genome set and estimates the remaining ones, called target genes, via the computational model. Several shallow and deep regression models have been presented in the literature for inferring the expressions of target genes. However, the shallow models suffer from underfitting due to their insufficient capacity in capturing the complex nature of gene expression data, and the existing deep models are prone to overfitting due to the lack of using the interrelations of target genes in the learning framework. To address these challenges, we formulate the gene expression inference as a multi-task learning problem and propose a novel deep multi-task learning algorithm with automatically learning the biological interrelations among target genes and utilizing such information to enhance the prediction. In particular, we employ a multi-layer sub-network with low dimensional latent variables for learning the interrelations among target genes (i.e. distinct predictive tasks), and impose a seamless and easy to implement regularization on deep models. Unlike the conventional complicated multi-task learning methods, which can only deal with tens or hundreds of tasks, our proposed algorithm can effectively learn the interrelations from the large-scale (\(\sim \)10,000) tasks on the gene expression inference problem, and does not suffer from cost-prohibitive operations. Experimental results indicate the superiority of our method compared to the existing gene expression inference models and alternative multi-task learning algorithms on two large-scale datasets.

Alzheimer’s disease (AD) is a serious neurodegenerative condition that affects millions of individuals across the world. As the average age of individuals in the United States and the world increases, the prevalence of AD will continue to grow. To address this public health problem, the research community has developed computational approaches to sift through various aspects of clinical data and uncover their insights, among which one of the most challenging problem is to determine the biological mechanisms that cause AD to develop. To study this problem, in this paper we present a novel Joint Multi-Modal Longitudinal Regression and Classification method and show how it can be used to identify the cognitive status of the participants in the Alzheimer’s Disease Neuroimaging Initiative (ADNI) cohort and the underlying biological mechanisms. By intelligently combining clinical data of various modalities ( i.e. , genetic information and brain scans) using a variety of regularizations that can identify AD-relevant biomarkers, we perform the regression and classification tasks simultaneously. Because the proposed objective is a non-smooth optimization problem that is difficult to solve in general, we derive an efficient iterative algorithm and rigorously prove its convergence. To validate our new method in predicting the cognitive scores of patients and their clinical diagnosis, we conduct comprehensive experiments on the ADNI cohort. Our promising results demonstrate the benefits and flexibility of the proposed method. We anticipate that our new method is of interest to clinical communities beyond AD research and have open-sourced the code of our method online. 1 1 The code package for the proposed Joint Multi-Modal Longitudinal Regression and Classification model have been made publicly available online at https://github.com/minds-mines/jmmlrc .

With the rapid development of high-throughput genotyping and neuroimaging techniques, imaging genetics has drawn significant attention in the study of complex brain diseases such as Alzheimer's disease (AD). Research on the associations between genotype and phenotype improves the understanding of the genetic basis and biological mechanisms of brain structure and function. AD is a progressive neurodegenerative disease; therefore, the study on the relationship between single nucleotide polymorphism (SNP) and longitudinal variations of neuroimaging phenotype is crucial. Although some machine learning models have recently been proposed to capture longitudinal patterns in genotype-phenotype association studies, most machine-learning models base the learning on fixed structure among longitudinal prediction tasks rather than automatically learning the interrelationships. In response to this challenge, we propose a new automated time structure learning model to automatically reveal the longitudinal genotype-phenotype interactions and exploits such learned structure to enhance the phenotypic predictions. We proposed an efficient optimization algorithm for our model and provided rigorous theoretical convergence proof. We performed experiments on the Alzheimer's Disease Neuroimaging Initiative (ADNI) cohort for longitudinal phenotype prediction, including 3123 SNPs and 2 biomarkers (Voxel-Based Morphometry and FreeSurfer). The empirical results validate that our proposed model is superior to the counterparts. In addition, the best SNPs identified by our model have been replicated in the literature, which justifies our prediction.

Microscopy image classification is important in various biomedical applications, such as cancer subtype identification, and protein localization for high content screening. To achieve automated and effective microscopy image classification, the representative and discriminative capability of image feature descriptors is essential. To this end, in this paper, we propose a new feature representation algorithm to facilitate automated microscopy image classification. In particular, we incorporate Fisher vector (FV) encoding with multiple types of local features that are handcrafted or learned, and we design a separation-guided dimension reduction method to reduce the descriptor dimension while increasing its discriminative capability. Our method is evaluated on four publicly available microscopy image data sets of different imaging types and applications, including the UCSB breast cancer data set, MICCAI 2015 CBTC challenge data set, and IICBU malignant lymphoma, and RNAi data sets. Our experimental results demonstrate the advantage of the proposed low-dimensional FV representation, showing consistent performance improvement over the existing state of the art and the commonly used dimension reduction techniques.

Sequence describes the primary structure of a protein, which contains important structural, characteristic, and genetic information and thereby motivates many sequence-based computational approaches to infer protein function. Among them, feature-base approaches attract increased attention because they make prediction from a set of transformed and more biologically meaningful sequence features. However, original features extracted from sequence are usually of high dimensionality and often compromised by irrelevant patterns, therefore dimension reduction is necessary prior to classification for efficient and effective protein function prediction. A protein usually performs several different functions within an organism, which makes protein function prediction a multi-label classification problem. In machine learning, multi-label classification deals with problems where each object may belong to more than one class. As a well-known feature reduction method, linear discriminant analysis (LDA) has been successfully applied in many practical applications. It, however, by nature is designed for single-label classification , in which each object can belong to exactly one class. Because directly applying LDA in multi-label classification causes ambiguity when computing scatters matrices, we apply a new Multi-label Linear Discriminant Analysis (MLDA) approach to address this problem and meanwhile preserve powerful classification capability inherited from classical LDA. We further extend MLDA by $\ell _1$ -normalization to overcome the problem of over-counting data points with multiple labels. In addition, we incorporate biological network data using Laplacian embedding into our method, and assess the reliability of predicted putative functions. Extensive empirical evaluations demonstrate promising results of our methods.

Accurate and reliable segmentation of the prostate gland using magnetic resonance (MR) imaging has critical importance for the diagnosis and treatment of prostate diseases, especially prostate cancer. Although many automated segmentation approaches, including those based on deep learning have been proposed, the segmentation performance still has room for improvement due to the large variability in image appearance, imaging interference, and anisotropic spatial resolution. In this paper, we propose the 3D adversarial pyramid anisotropic convolutional deep neural network (3D APA-Net) for prostate segmentation in MR images. This model is composed of a generator (i.e., 3D PA-Net) that performs image segmentation and a discriminator (i.e., a six-layer convolutional neural network) that differentiates between a segmentation result and its corresponding ground truth. The 3D PA-Net has an encoder-decoder architecture, which consists of a 3D ResNet encoder, an anisotropic convolutional decoder, and multi-level pyramid convolutional skip connections. The anisotropic convolutional blocks can exploit the 3D context information of the MR images with anisotropic resolution, the pyramid convolutional blocks address both voxel classification and gland localization issues, and the adversarial training regularizes 3D PA-Net and thus enables it to generate spatially consistent and continuous segmentation results. We evaluated the proposed 3D APA-Net against several state-of-the-art deep learning-based segmentation approaches on two public databases and the hybrid of the two. Our results suggest that the proposed model outperforms the compared approaches on three databases and could be used in a routine clinical workflow.

Studying a common architecture reflecting both brain's structural and functional organizations across individuals and populations in a hierarchical way has been of significant interest in the brain mapping field. Recently, deep learning models exhibited ability in extracting meaningful hierarchical structures from brain imaging data, e.g., fMRI and DTI. However, deep learning models have been rarely used to explore the relation between brain structure and function yet. In this paper, we proposed a novel multimodal deep believe network (DBN) model to discover and quantitatively represent the hierarchical organizations of common and consistent brain networks from both fMRI and DTI data. A prominent characteristic of DBN is that it is capable of extracting meaningful features from complex neuroimaging data with a hierarchical manner. With our proposed DBN model, three hierarchical layers with hundreds of common and consistent brain networks across individual brains are successfully constructed through learning a large dimension of representative features from fMRI/DTI data.

It is largely unknown whether there is functional role difference between cortical gyral and sulcal regions. Recent advancements in neuroimaging studies demonstrate clear difference of structural connection profiles in gyral and sulcal areas, suggesting possible functional role difference in these convex and concave cortical regions. To explore and confirm such possible functional difference, we design and apply a powerful deep learning model of convolutional neural networks (CNN) that has been proven to be superior in learning discriminative and meaningful patterns on fMRI. By using the CNN model, gyral and sulcal fMRI signals are learned and predicted, and the prediction performance is adopted to demonstrate the functional difference between gyri and sulci. By using the Human Connectome Project (HCP) fMRI data and macaque brain fMRI data, an average of 83% and 90% classification accuracy has been achieved to separate gyral/sulcal HCP task fMRI signals at the population and individual subject level, respectively; 81% and 86% classification accuracy for resting state fMRI signals at the group and individual subject level, respectively. In addition, 78% classification accuracy has been achieved to separate gyral/sulcal resting state fMRI signals in macaque brains. Importantly, further analysis reveals that the discriminative features that are learned by CNNs to differentiate gyral/sulcal fMRI signals can be meaningfully interpreted, thus unveiling the fundamental functional difference between cortical gyri and sulci. That is, gyri are more global functional integration centers with simpler lower frequency signal components, while sulci are more local processing units with more complex higher frequency signal components.

Newton methods can be applied in many supervised learning approaches. However, for large-scale data, the use of the whole Hessian matrix can be time-consuming. Recently, subsampled Newton methods have been proposed to reduce the computational time by using only a subset of data for calculating an approximation of the Hessian matrix. Unfortunately, we find that in some situations, the running speed is worse than the standard Newton method because cheaper but less accurate search directions are used. In this work, we propose some novel techniques to improve the existing subsampled Hessian Newton method. The main idea is to solve a two-dimensional subproblem per iteration to adjust the search direction to better minimize the second-order approximation of the function value. We prove the theoretical convergence of the proposed method. Experiments on logistic regression, linear SVM, maximum entropy, and deep networks indicate that our techniques significantly reduce the running time of the subsampled Hessian Newton method. The resulting algorithm becomes a compelling alternative to the standard Newton method for large-scale data classification.

The task of segmenting cell nuclei and cytoplasm in pap smear images is one of the most challenging tasks in automated cervix cytological analysis due to specifically the presence of overlapping cells. This paper introduces a multi-pass fast watershed-based method (MPFW) to segment both nucleus and cytoplasm from large cell masses of overlapping cervical cells in three watershed passes. The first pass locates the nuclei with barrier-based watershed on the gradient-based edge map of a pre-processed image. The next pass segments the isolated, touching, and partially overlapping cells with a watershed transform adapted to the cell shape and location. The final pass introduces mutual iterative watersheds separately applied to each nucleus in the largely overlapping clusters to estimate the cell shape. In MPFW, the line-shaped contours of the watershed cells are deformed with ellipse fitting and contour adjustment to give a better representation of cell shapes. The performance of the proposed method has been evaluated using synthetic, real extended depth-of-field, and multi-layers cervical cytology images provided by the first and second overlapping cervical cytology image segmentation challenges in ISBI 2014 and ISBI 2015. The experimental results demonstrate superior performance of the proposed MPFW in terms of segmentation accuracy, detection rate, and time complexity, compared with recent peer methods.

In Alzheimer’s research, Mild Cognitive Impairment (MCI) is an important intermediate stage between normal aging and Alzheimer’s. How to distinguish MCI samples that finally convert to AD from those do not is an essential problem in the prevention and diagnosis of Alzheimer’s. Traditional methods use various classification models to distinguish MCI converters from non-converters, while the performance is usually limited by the small number of available data. Moreover, previous methods only use the data at baseline time for training but ignore the longitudinal information at other time points along the disease progression. To tackle with these problems, we propose a novel deep learning framework that uncovers the temporal correlation structure between adjacent time points in the disease progression. We also construct a generative framework to learn the inherent data distribution so as to produce more reliable data to strengthen the training process. Extensive experiments on the ADNI cohort validate the superiority of our model.

Alzheimer’s disease (AD) is a degenerative brain disease that affects millions of people around the world. As populations in the United States and worldwide age, the prevalence of Alzheimer’s disease will only increase. In turn, the social and financial costs of AD will create a difficult environment for many families and caregivers across the globe. By combining genetic information, brain scans, and clinical data, gathered over time through the Alzheimer’s Disease Neuroimaging Initiative (ADNI), we propose a new Joint High-Order Multi-Modal Multi-Task Feature Learning method to predict the cognitive performance and diagnosis of patients with and without AD.

Motivation: Imaging genetics is an emerging field that studies the influence of genetic variation on brain structure and function. The major task is to examine the association between genetic markers such as single-nucleotide polymorphisms (SNPs) and quantitative traits (QTs) extracted from neuroimaging data. The complexity of these datasets has presented critical bioinformatics challenges that require new enabling tools. Sparse canonical correlation analysis (SCCA) is a bi-multivariate technique used in imaging genetics to identify complex multi-SNP–multi-QT associations. However, most of the existing SCCA algorithms are designed using the soft thresholding method, which assumes that the input features are independent from one another. This assumption clearly does not hold for the imaging genetic data. In this article, we propose a new knowledge-guided SCCA algorithm (KG-SCCA) to overcome this limitation as well as improve learning results by incorporating valuable prior knowledge. Results: The proposed KG-SCCA method is able to model two types of prior knowledge: one as a group structure (e.g. linkage disequilibrium blocks among SNPs) and the other as a network structure (e.g. gene co-expression network among brain regions). The new model incorporates these prior structures by introducing new regularization terms to encourage weight similarity between grouped or connected features. A new algorithm is designed to solve the KG-SCCA model without imposing the independence constraint on the input features. We demonstrate the effectiveness of our algorithm with both synthetic and real data. For real data, using an Alzheimer’s disease (AD) cohort, we examine the imaging genetic associations between all SNPs in the APOE gene (i.e. top AD gene) and amyloid deposition measures among cortical regions (i.e. a major AD hallmark). In comparison with a widely used SCCA implementation, our KG-SCCA algorithm produces not only improved cross-validation performances but also biologically meaningful results. Availability: Software is freely available on request. Contact: shenli@iu.edu

Most existing array comparative genomic hybridization (array CGH) data processing methods and evaluation models assumed that the probability density function (pdf) of noise in array CGH data is a Gaussian distribution. However, in practice, such noise distribution is peaky and heavy-tailed. Therefore, a Gaussian pdf is not adequate to approximate the noise in array CGH data and hence introduces wrong detections of chromosomal aberrations and leads misunderstanding on disease pathogenesis. A more accurate and sufficient model of noise in array CGH data is necessary and beneficial to the detection of DNA copy number variations. We analyze the real array CGH data from different platforms and show that the distribution of noise in array CGH data is fitted very well by generalized Gaussian distribution (GGD). Based on our new noise model, we propose a novel array CGH processing method combining the advantages of both the smoothing and segmentation approaches. The new method uses generalized Gaussian bivariate shrinkage function and one-directional derivative wavelet scalogram in generalized Gaussian noise. In the smoothing step, with the new generalized Gaussian noise model, we derive the heavy-tailed noise suppression algorithm in stationary wavelet domain. In the segmentation step, the 1D Gaussian derivative wavelet scalogram is employed to detect break points. Both real and simulated array CGH data with different noises (such as Gaussian noise, GGD noise, and real noise) are used in our experiments. We demonstrate that our new method outperforms other state-of-the-art methods, in terms of both root mean squared errors and receiver operating characteristic curves.

Blind source separation (BSS) is commonly used in functional magnetic resonance imaging (fMRI) data analysis. Recently, BSS models based on restricted Boltzmann machine (RBM), one of the building blocks of deep learning models, have been shown to improve brain network identification compared to conventional single matrix factorization models such as independent component analysis (ICA). These models, however, trained RBM on fMRI volumes, and are hence challenged by model complexity and limited training set. In this article, we propose to apply RBM to fMRI time courses instead of volumes for BSS. The proposed method not only interprets fMRI time courses explicitly to take advantages of deep learning models in latent feature learning but also substantially reduces model complexity and increases the scale of training set to improve training efficiency. Our experimental results based on Human Connectome Project (HCP) datasets demonstrated the superiority of the proposed method over ICA and the one that applied RBM to fMRI volumes in identifying task-related components, resulted in more accurate and specific representations of task-related activations. Moreover, our method separated out components representing intermixed effects between task events, which could reflect inherent interactions among functionally connected brain regions. Our study demonstrates the value of RBM in mining complex structures embedded in large-scale fMRI data and its potential as a building block for deeper models in fMRI data analysis.

In aging research, morphological age of tissue helps to characterize the effects of aging on different individuals. While currently manual evaluations are used to estimate morphological ages under microscopy, such operation is difficult and subjective due to the complex visual characteristics of tissue images. In this paper, we propose an automated method to quantify morphological ages of tissues from microscopy images. We design a new sparse representation method, namely dual discriminative local coding (DDLC), that classifies the tissue images into different chronological ages. DDLC in- corporates discriminative distance learning and dual-level local coding into the basis model of locality-constrained linear coding thus achieves higher discriminative capability. The morphological age is then computed based on the classification scores. We conducted our study using the publicly avail- able terminal bulb aging database that has been commonly used in existing microscopy imaging research. To represent these images, we also design a highly descriptive descriptor that combines several complementary texture features extracted at two scales. Experimental results show that our method achieves significant improvement in age classification when compared to the existing approaches and other popular classifiers. We also present promising results in quantification of morphological ages.

Alzheimer’s Disease (AD), a severe type of neurodegenerative disorder with progressive impairment of learning and memory, has threatened the health of millions of people. How to recognize AD at early stage is crucial. Multiple models have been presented to predict cognitive impairments by means of neuroimaging data. However, traditional models did not employ the valuable longitudinal information along the progression of the disease. In this paper, we proposed a novel longitudinal feature learning model to simultaneously uncover the interrelations among different cognitive measures at different time points and utilize such interrelated structures to enhance the learning of associations between imaging features and prediction tasks. Moreover, we adopted Schatten p-norm to identify the interrelation structures existing in the low-rank subspace. Empirical results on the ADNI cohort demonstrated promising performance of our model.

We present a lesion detection and characterization method for (18)F-fluorodeoxyglucose positron emission tomography-computed tomography (FDG PET-CT) images of the thorax in the evaluation of patients with primary nonsmall cell lung cancer (NSCLC) with regional nodal disease. Lesion detection can be difficult due to low contrast between lesions and normal anatomical structures. Lesion characterization is also challenging due to similar spatial characteristics between the lung tumors and abnormal lymph nodes. To tackle these problems, we propose a context driven approximation (CDA) method. There are two main components of our method. First, a sparse representation technique with region-level contexts was designed for lesion detection. To discriminate low-contrast data with sparse representation, we propose a reference consistency constraint and a spatial consistent constraint. Second, a multi-atlas technique with image-level contexts was designed to represent the spatial characteristics for lesion characterization. To accommodate inter-subject variation in a multi-atlas model, we propose an appearance constraint and a similarity constraint. The CDA method is effective with a simple feature set, and does not require parametric modeling of feature space separation. The experiments on a clinical FDG PET-CT dataset show promising performance improvement over the state-of-the-art.

Array comparative genomic hybridization (aCGH) is a newly introduced method for the detection of copy number abnormalities associated with human diseases with special focus on cancer. Specific patterns in DNA copy number variations (CNVs) can be associated with certain disease types and can facilitate prognosis and progress monitoring of the disease. Machine learning techniques have been used to model the problem of tissue typing as a classification problem. Feature selection is an important part of the classification process, because many biological features are not related to the diseases and confuse the classification tasks. Multiple feature selection methods have been proposed in the different domains where classification has been applied. In this work, we will present a new feature selection method based on structured sparsity-inducing norms to identify the informative aCGH biomarkers which can help us classify different disease subtypes. To validate the performance of the proposed method, we experimentally compare it with existing feature selection methods on four publicly available aCGH data sets. In all empirical results, the proposed sparse learning based feature selection method consistently outperforms other related approaches. More important, we carefully investigate the aCGH biomarkers selected by our method, and the biological evidences in literature strongly support our results.

Background Bioimage classification is a fundamental problem for many important biological studies that require accurate cell phenotype recognition, subcellular localization, and histopathological classification. In this paper, we present a new bioimage classification method that can be generally applicable to a wide variety of classification problems. We propose to use a high-dimensional multi-modal descriptor that combines multiple texture features. We also design a novel subcategory discriminant transform (SDT) algorithm to further enhance the discriminative power of descriptors by learning convolution kernels to reduce the within-class variation and increase the between-class difference. Results We evaluate our method on eight different bioimage classification tasks using the publicly available IICBU 2008 database. Each task comprises a separate dataset, and the collection represents typical subcellular, cellular, and tissue level classification problems. Our method demonstrates improved classification accuracy (0.9 to 9%) on six tasks when compared to state-of-the-art approaches. We also find that SDT outperforms the well-known dimension reduction techniques, with for example 0.2 to 13% improvement over linear discriminant analysis. Conclusions We present a general bioimage classification method, which comprises a highly descriptive visual feature representation and a learning-based discriminative feature transformation algorithm. Our evaluation on the IICBU 2008 database demonstrates improved performance over the state-of-the-art for six different classification tasks.

Protein interactions are central to all the biological processes and structural scaffolds in living organisms, because they orchestrate a number of cellular processes such as metabolic pathways and immunological recognition. Several high-throughput methods, for example, yeast two-hybrid system and mass spectrometry method, can help determine protein interactions, which, however, suffer from high false-positive rates. Moreover, many protein interactions predicted by one method are not supported by another. Therefore, computational methods are necessary and crucial to complete the interactome expeditiously. In this work, we formulate the problem of predicting protein interactions from a new mathematical perspective--sparse matrix completion, and propose a novel nonnegative matrix factorization (NMF)-based matrix completion approach to predict new protein interactions from existing protein interaction networks. Through using manifold regularization, we further develop our method to integrate different biological data sources, such as protein sequences, gene expressions, protein structure information, etc. Extensive experimental results on four species, Saccharomyces cerevisiae, Drosophila melanogaster, Homo sapiens, and Caenorhabditis elegans, have shown that our new methods outperform related state-of-the-art protein interaction prediction methods.

With the advances of neuroimaging techniques and genome sequences understanding, the phenotype and genotype data have been utilized to study the brain diseases (known as imaging genetics). One of the most important topics in image genetics is to discover the genetic basis of phenotypic markers and their associations. In such studies, the linear regression models have been playing an important role by providing interpretable results. However, due to their modeling characteristics, it is limited to effectively utilize inherent information among the phenotypes and genotypes, which are helpful for better understanding their associations. In this work, we propose a structured sparse low-rank regression method to explicitly consider the correlations within the imaging phenotypes and the genotypes simultaneously for Brain-Wide and Genome-Wide Association (BW-GWA) study. Specifically, we impose the low-rank constraint as well as the structured sparse constraint on both phenotypes and phenotypes. By using the Alzheimer’s Disease Neuroimaging Initiative (ADNI) dataset, we conducted experiments of predicting the phenotype data from genotype data and achieved performance improvement by 12.75 % on average in terms of the root-mean-square error over the state-of-the-art methods.

As a neurodegenerative disorder, the Alzheimer's disease (AD) status can be characterized by the progressive impairment of memory and other cognitive functions. Thus, it is an important topic to use neuroimaging measures to predict cognitive performance and track the progression of AD. Many existing cognitive performance prediction methods employ the regression models to associate cognitive scores to neuroimaging measures, but these methods do not take into account the interconnected structures within imaging data and those among cognitive scores. To address this problem, we propose a novel multi-task learning model for minimizing the k smallest singular values to uncover the underlying low-rank common subspace and jointly analyze all the imaging and clinical data. The effectiveness of our method is demonstrated by the clearly improved prediction performances in all empirical AD cognitive scores prediction cases.

Motivation: Staining the mRNA of a gene via in situ hybridization (ISH) during the development of a Drosophila melanogaster embryo delivers the detailed spatio-temporal patterns of the gene expression. Many related biological problems such as the detection of co-expressed genes, co-regulated genes and transcription factor binding motifs rely heavily on the analysis of these image patterns. To provide the text-based pattern searching for facilitating related biological studies, the images in the Berkeley Drosophila Genome Project (BDGP) study are annotated with developmental stage term and anatomical ontology terms manually by domain experts. Due to the rapid increase in the number of such images and the inevitable bias annotations by human curators, it is necessary to develop an automatic method to recognize the developmental stage and annotate anatomical terms. Results: In this article, we propose a novel computational model for jointly stage classification and anatomical terms annotation of Drosophila gene expression patterns. We propose a novel Tri-Relational Graph (TG) model that comprises the data graph, anatomical term graph, developmental stage term graph, and connect them by two additional graphs induced from stage or annotation label assignments. Upon the TG model, we introduce a Preferential Random Walk (PRW) method to jointly recognize developmental stage and annotate anatomical terms by utilizing the interrelations between two tasks. The experimental results on two refined BDGP datasets demonstrate that our joint learning method can achieve superior prediction results on both tasks than the state-of-the-art methods. Availability: http://ranger.uta.edu/%7eheng/Drosophila/ Contact: heng@uta.edu

Computational tools for the analysis of complex biological networks are lacking in human connectome research. Especially, how to discover the brain network patterns shared by a group of subjects is a challenging computational neuroscience problem. Although some single graph clustering methods can be extended to solve the multi-graph cases, the discovered network patterns are often imbalanced, e.g. isolated points. To address these problems, we propose a novel indicator constrained and balanced multi-graph normalized cut method to identify the connectome module patterns from the connectivity brain networks of the targeted subject group. We evaluated our method by analyzing the weighted fiber connectivity networks.

Motivation: Structured sparse canonical correlation analysis (SCCA) models have been used to identify imaging genetic associations. These models either use group lasso or graph-guided fused lasso to conduct feature selection and feature grouping simultaneously. The group lasso based methods require prior knowledge to define the groups, which limits the capability when prior knowledge is incomplete or unavailable. The graph-guided methods overcome this drawback by using the sample correlation to define the constraint. However, they are sensitive to the sign of the sample correlation, which could introduce undesirable bias if the sign is wrongly estimated. Results: We introduce a novel SCCA model with a new penalty, and develop an efficient optimization algorithm. Our method has a strong upper bound for the grouping effect for both positively and negatively correlated features. We show that our method performs better than or equally to three competing SCCA models on both synthetic and real data. In particular, our method identifies stronger canonical correlations and better canonical loading patterns, showing its promise for revealing interesting imaging genetic associations. Availability and implementation: The Matlab code and sample data are freely available at http://www.iu.edu/∼shenlab/tools/angscca/. Contact: shenli@iu.edu Supplementary information: Supplementary data are available at Bioinformatics online.

Motivation: Peaks are the key information in mass spectrometry (MS) which has been increasingly used to discover diseases-related proteomic patterns. Peak detection is an essential step for MS-based proteomic data analysis. Recently, several peak detection algorithms have been proposed. However, in these algorithms, there are three major deficiencies: (i) because the noise is often removed, the true signal could also be removed; (ii) baseline removal step may get rid of true peaks and create new false peaks; (iii) in peak quantification step, a threshold of signal-to-noise ratio (SNR) is usually used to remove false peaks; however, noise estimations in SNR calculation are often inaccurate in either time or wavelet domain. In this article, we propose new algorithms to solve these problems. First, we use bivariate shrinkage estimator in stationary wavelet domain to avoid removing true peaks in denoising step. Second, without baseline removal, zero-crossing lines in multi-scale of derivative Gaussian wavelets are investigated with mixture of Gaussian to estimate discriminative parameters of peaks. Third, in quantification step, the frequency, SD, height and rank of peaks are used to detect both high and small energy peaks with robustness to noise. Results: We propose a novel Gaussian Derivative Wavelet (GDWavelet) method to more accurately detect true peaks with a lower false discovery rate than existing methods. The proposed GDWavelet method has been performed on the real Surface-Enhanced Laser Desorption/Ionization Time-Of-Flight (SELDI-TOF) spectrum with known polypeptide positions and on two synthetic data with Gaussian and real noise. All experimental results demonstrate that our method outperforms other commonly used methods. The standard receiver operating characteristic (ROC) curves are used to evaluate the experimental results. Availability: http://ranger.uta.edu/∼heng/MS/GDWavelet.html or http://www.naaan.org/nhanguyen/archive.htm Contact: heng@uta.edu

Conventional computational approaches for protein function prediction usually predict one function at a time, fundamentally. As a result, the protein functions are treated as separate target classes. However, biological processes are highly correlated in reality, which makes multiple functions assigned to a protein not independent. Therefore, it would be beneficial to make use of function category correlations when predicting protein functions. In this article, we propose a novel Maximization of Data-Knowledge Consistency (MDKC) approach to exploit function category correlations for protein function prediction. Our approach banks on the assumption that two proteins are likely to have large overlap in their annotated functions if they are highly similar according to certain experimental data. We first establish a new pairwise protein similarity using protein annotations from knowledge perspective. Then by maximizing the consistency between the established knowledge similarity upon annotations and the data similarity upon biological experiments, putative functions are assigned to unannotated proteins. Most importantly, function category correlations are gracefully incorporated into our learning objective through the knowledge similarity. Comprehensive experimental evaluations on the Saccharomyces cerevisiae species have demonstrated promising results that validate the performance of our methods.

In this paper, we propose a new Locality-constrained Subcluster Representation Ensemble (LSRE) model, to classify high-resolution computed tomography (HRCT) images of interstitial lung diseases (ILDs). Medical images normally exhibit large intra-class variation and inter-class ambiguity in the feature space. Modelling of feature space separation between different classes is thus problematic and this affects the classification performance. Our LSRE model tackles this issue in an ensemble classification construct. The image set is first partitioned into subclusters based on spectral clustering with approximation-based affinity matrix. Basis representations of the test image are then generated with sparse approximation from the subclusters. These basis representations are finally fused with approximation- and distribution-based weights to classify the test image. Our experimental results on a large HRCT database show good performance improvement over existing popular classifiers.

Medical images typically exhibit complex feature space distributions due to high intra-class variation and inter-class ambiguity. Monolithic classification models are often problematic. In this study, we propose a novel Large Margin Local Estimate (LMLE) method for medical image classification. In the first step, the reference images are subcategorized, and local estimates of the test image are computed based on the reference subcategories. In the second step, the local estimates are fused in a large margin model to derive the similarity level between the test image and the reference images, and the test image is classified accordingly. For evaluation, the LMLE method is applied to classify image patches of different interstitial lung disease (ILD) patterns on high-resolution computed tomography (HRCT) images. We demonstrate promising performance improvement over the state-of-the-art.

Brain imaging genetics is an emergent research field where the association between genetic variations such as single nucleotide polymorphisms (SNPs) and neuroimaging quantitative traits (QTs) is evaluated. Sparse canonical correlation analysis (SCCA) is a bi-multivariate analysis method that has the potential to reveal complex multi-SNP-multi-QT associations. Most existing SCCA algorithms are designed using the soft threshold strategy, which assumes that the features in the data are independent from each other. This independence assumption usually does not hold in imaging genetic data, and thus inevitably limits the capability of yielding optimal solutions. We propose a novel structure-aware SCCA (denoted as S2CCA) algorithm to not only eliminate the independence assumption for the input data, but also incorporate group-like structure in the model. Empirical comparison with a widely used SCCA implementation, on both simulated and real imaging genetic data, demonstrated that S2CCA could yield improved prediction performance and biologically meaningful findings.

There have been several recent studies that used sparse representation for fMRI signal analysis and activation detection based on the assumption that each voxel's fMRI signal is linearly composed of sparse components. Previous studies have employed sparse coding to model functional networks in various modalities and scales. These prior contributions inspired the exploration of whether/how sparse representation can be used to identify functional networks in a voxel-wise way and on the whole brain scale. This paper presents a novel, alternative methodology of identifying multiple functional networks via sparse representation of whole-brain task-based fMRI signals. Our basic idea is that all fMRI signals within the whole brain of one subject are aggregated into a big data matrix, which is then factorized into an over-complete dictionary basis matrix and a reference weight matrix via an effective online dictionary learning algorithm. Our extensive experimental results have shown that this novel methodology can uncover multiple functional networks that can be well characterized and interpreted in spatial, temporal and frequency domains based on current brain science knowledge. Importantly, these well-characterized functional network components are quite reproducible in different brains. In general, our methods offer a novel, effective and unified solution to multiple fMRI data analysis tasks including activation detection, de-activation detection, and functional network identification.

Neuroimaging markers have been widely used to predict the cognitive functions relevant to the progression of Alzheimer's disease (AD). Most previous studies identify the imaging markers without considering the brain structural correlations between neuroimaging measures. However, many neuroimaging markers interrelate and work together to reveal the cognitive functions, such that these relevant markers should be selected together as the phenotypic markers. To solve this problem, in this paper, we propose a novel network constrained feature selection (NCFS) model to identify the neuroimaging markers guided by the structural brain network, which is constructed by the sparse representation method such that the interrelations between neuroimaging features are encoded into probabilities. Our new methods are evaluated by the MRI and AV45-PET data from ADNI-GO and ADNI-2 (Alzheimer's Disease Neuroimaging Initiative). In all cognitive function prediction tasks, our new NCFS method outperforms other state-of-the-art regression approaches. Meanwhile, we show that the new method can select the correlated imaging markers, which are ignored by the competing approaches.

The performance of automatic lesion detection is often affected by the intra- and inter-subject feature variations of lesions and normal anatomical structures. In this work, we propose a similarity-guided sparse representation method for image patch labeling, with three aspects of similarity information modeling, to reduce the chance that the best reconstruction of a feature vector does not provide the correct classification. Based on this classification model, we then design a new approach for detecting lesions in positron emission tomography computed tomography (PET-CT) images. The approach works well with simple image features, and the proposed sparse representation model is effectively applied for both detection of all lesions and characterization of lung tumors and abnormal lymph nodes. The experiments show promising performance improvement over the state-of-the-art.

Many recent scientific efforts have been devoted to constructing the human connectome using Diffusion Tensor Imaging (DTI) data for understanding the large-scale brain networks that underlie higher-level cognition in human. However, suitable computational network analysis tools are still lacking in human connectome research. To address this problem, we propose a novel multi-graph min-max cut model to detect the consistent network modules from the brain connectivity networks of all studied subjects. A new multi-graph MinMax cut model is introduced to solve this challenging computational neuroscience problem and the efficient optimization algorithm is derived. In the identified connectome module patterns, each network module shows similar connectivity patterns in all subjects, which potentially associate to specific brain functions shared by all subjects. We validate our method by analyzing the weighted fiber connectivity networks. The promising empirical results demonstrate the effectiveness of our method.

In-vivo parcellation of the cerebral cortex via non-invasive neuroimaging data has been in active research for years. A variety of model-driven and/or data-driven computational approaches have been proposed to parcellate the cortex. However, two fundamental common issues in these parcellation methodologies are the features or attributes used to define boundaries between cortical regions and the establishment of correspondences of the parcellated regions across different brains. This paper uses a novel DTI-derived fiber shape feature for the parcellation of cortical gyrus into fine-granularity segments. The gyral parcellation is formulated and solved as a surface vertex clustering problem, in which fiber shape feature similarity is used to define the distances between vertices. Then, we designed and applied a novel multi-view spectral clustering algorithm to group the vertices into group-wise consistent gyral segments across different brains. The experimental results showed that the precentral and postcentral gyrus, as two test-beds, can be consistently parcellated into 10 segments on both hemispheres across different subjects. Evaluation studies using benchmark task-based fMRI and cortical landmarks demonstrated the effectiveness and validity of the proposed methods.

Motivation: Imaging genetic studies typically focus on identifying single-nucleotide polymorphism (SNP) markers associated with imaging phenotypes. Few studies perform regression of SNP values on phenotypic measures for examining how the SNP values change when phenotypic measures are varied. This alternative approach may have a potential to help us discover important imaging genetic associations from a different perspective. In addition, the imaging markers are often measured over time, and this longitudinal profile may provide increased power for differentiating genotype groups. How to identify the longitudinal phenotypic markers associated to disease sensitive SNPs is an important and challenging research topic. Results: Taking into account the temporal structure of the longitudinal imaging data and the interrelatedness among the SNPs, we propose a novel ‘task-correlated longitudinal sparse regression’ model to study the association between the phenotypic imaging markers and the genotypes encoded by SNPs. In our new association model, we extend the widely used ℓ2,1-norm for matrices to tensors to jointly select imaging markers that have common effects across all the regression tasks and time points, and meanwhile impose the trace-norm regularization onto the unfolded coefficient tensor to achieve low rank such that the interrelationship among SNPs can be addressed. The effectiveness of our method is demonstrated by both clearly improved prediction performance in empirical evaluations and a compact set of selected imaging predictors relevant to disease sensitive SNPs. Availability: Software is publicly available at: http://ranger.uta.edu/%7eheng/Longitudinal/ Contact: heng@uta.edu or shenli@inpui.edu

Motivation: Recent advances in brain imaging and high-throughput genotyping techniques enable new approaches to study the influence of genetic and anatomical variations on brain functions and disorders. Traditional association studies typically perform independent and pairwise analysis among neuroimaging measures, cognitive scores and disease status, and ignore the important underlying interacting relationships between these units. Results: To overcome this limitation, in this article, we propose a new sparse multimodal multitask learning method to reveal complex relationships from gene to brain to symptom. Our main contributions are three-fold: (i) introducing combined structured sparsity regularizations into multimodal multitask learning to integrate multidimensional heterogeneous imaging genetics data and identify multimodal biomarkers; (ii) utilizing a joint classification and regression learning model to identify disease-sensitive and cognition-relevant biomarkers; (iii) deriving a new efficient optimization algorithm to solve our non-smooth objective function and providing rigorous theoretical analysis on the global optimum convergency. Using the imaging genetics data from the Alzheimer's Disease Neuroimaging Initiative database, the effectiveness of the proposed method is demonstrated by clearly improved performance on predicting both cognitive scores and disease status. The identified multimodal biomarkers could predict not only disease status but also cognitive function to help elucidate the biological pathway from gene to brain structure and function, and to cognition and disease. Availability: Software is publicly available at: http://ranger.uta.edu/%7eheng/multimodal/ Contact: heng@uta.edu; shenli@iupui.edu

Traditional neuroimaging studies in Alzheimer’s disease (AD) typically employ independent and pairwise analyses between multimodal data, which treat imaging biomarkers, cognitive measures, and disease status as isolated units. To enhance mechanistic understanding of AD, in this paper, we conduct a new study for identifying imaging biomarkers that are associated with both cognitive measures and AD. To achieve this goal, we propose a new sparse joint classification and regression method. The imaging biomarkers identified by our method are AD-sensitive and cognition-relevant and can help reveal complex relationships among brain structure, cognition and disease status. Using the imaging and cognition data from Alzheimer’s Disease Neuroimaging Initiative database, the effectiveness of the proposed method is demonstrated by clearly improved performance on predicting both cognitive scores and disease status.

Motivation: Solenoid proteins are emerging as a protein class with properties intermediate between structured and intrinsically unstructured proteins. Containing repeating structural units, solenoid proteins are expected to share sequence similarities. However, in many cases, the sequence similarities are weak and non-detectable. Moreover, solenoids can be degenerated and widely vary in the number of units. So that it is difficult to detect them. Recently, several solenoid repeats detection methods have been proposed, such as self-alignment of the sequence, spectral analysis and discrete Fourier transform of sequence. Although these methods have shown good performance on certain data sets, they often fail to detect repeats with weak similarities. In this article, we propose a new approach to recognize solenoid repeats and non-solenoid proteins using stationary wavelet packet transform (SWPT). Our method associates with three advantages: (i) naturally representing five main factors of protein structure and properties by wavelet analysis technique; (ii) extracting novel wavelet features that can capture hidden components from solenoid sequence similarities and distinguish them from global proteins; (iii) obtaining statistics features that capture repeating motifs of solenoid proteins. Results: Our method analyzes the characteristics of amino acid sequence in both spectral and temporal domains using SWPT. Both global and local information of proteins are captured by SWPT coefficients. We obtain and integrate wavelet-based features and statistics-based features of amino acid sequence to improve the classification task. Our proposed method is evaluated by comparing to state-of-the-art methods such as HHrepID and REPETITA. The experimental results show that our algorithm consistently outperforms them in areas under ROC curve. At the same false positive rate, the sensitivity of our WAVELET method is higher than other methods. Availability: http://www.naaan.org/anvo/Software/Software.htm Contact: anphuocnhu.vo@mavs.uta.edu

Array-based comparative genomic hybridization (aCGH) has merged as a highly efficient technique for the detection of chromosomal imbalances. Characteristics of these DNA copy number aberrations provide the insights into cancer, and they are useful for the diagnostic and therapy strategies. In this article, we propose a statistical bivariate model for aCGH data in the stationary wavelet packet transform (SWPT) and apply this bivariate shrinkage estimator into the aCGH smoothing study. Because our new dependent Laplacian bivariate shrinkage estimator covers the dependency between wavelet coefficients and the shift invariant SWPT results include both low- and high-frequency information, our dependent Laplacian bivariate shrinkage estimator based SWPT method (named as SWPT-LaBi) has fundamental advantages to solve aCGH data smoothing problem compared to other methods. In our experiments, two standard evaluation methods, the Root Mean Squared Error (RMSE) and the Receiver Operating Characteristic (ROC) curve, are calculated to demonstrate the performance of our method. In all experimental results, our SWPT-LaBi method outperforms the previous most commonly used aCGH smoothing algorithms on both synthetic data and real data. Meantime, we also propose a new synthetic data generation method for aCGH smoothing algorithms evaluation. In our new data model, the noise from real aCGH data is extracted and used to improve synthetic data generation.

Mass Spectrometry (MS) is increasingly being used to discover disease related proteomic patterns. The peak detection step is one of most important steps in the typical analysis of MS data. Recently, many new algorithms have been proposed to increase true position rate with low false position rate in peak detection. Most of them follow two approaches: one is denoising approach and the other one is decomposing approach. In the previous studies, the decomposition of MS data method shows more potential than the first one. In this paper, we propose a new method named GaborLocal which can detect more true peaks with a very low false position rate. The Gaussian local maxima is employed for peak detection, because it is robust to noise in signals. Moreover, the maximum rank of peaks is defined at the first time to identify peaks instead of using the signal-to-noise ratio and the Gabor filter is used to decompose the raw MS signal. We perform the proposed method on the real SELDI-TOF spectrum with known polypeptide positions. The experimental results demonstrate our method outperforms other common used methods in the receiver operating characteristic (ROC) curve.

Mass Spectrometry (MS) is increasingly being used to discover diseases-related proteomic patterns. The peak detection step is one of the most important steps in the typical analysis of MS data. Recently, many new algorithms have been proposed to increase true position rate with low false discovery rate in peak detection. Most of them follow two approaches: one is the denoising approach and the other is the decomposing approach. In the previous studies, the decomposition of MS data method shows more potential than the first one. In this paper, we propose two novel methods, named GaborLocal and GaborEnvelop, both of which can detect more true peaks with a lower false discovery rate than previous methods. We employ the method of Gaussian local maxima to detect peaks, because it is robust to noise in signals. A new approach, peak rank, is defined for the first time to identify peaks instead of using the signal-to-noise ratio. Meanwhile, the Gabor filter is used to amplify important information and compress noise in the raw MS signal. Moreover, we also propose the envelope analysis to improve the quantification of peaks and remove more false peaks. The proposed methods have been performed on the real SELDI-TOF spectrum with known polypeptide positions. The experimental results demonstrate that our methods outperform other commonly used methods in the Receiver Operating Characteristic (ROC) curve.

In this paper, we propose a novel surface matching algorithm for arbitrarily shaped but simply connected 3-D objects. The spherical harmonic (SPHARM) method is used to describe these 3-D objects, and a novel surface registration approach is presented. The proposed technique is applied to various applications of medical image analysis. The results are compared with those using the traditional method, in which the first-order ellipsoid is used for establishing surface correspondence and aligning objects. In these applications, our surface alignment method is demonstrated to be more accurate and flexible than the traditional approach. This is due in large part to the fact that a new surface parameterization is generated by a shortcut that employs a useful rotational property of spherical harmonic basis functions for a fast implementation. In order to achieve a suitable computational speed for practical applications, we propose a fast alignment algorithm that improves computational complexity of the new surface registration method from O(n3) to O(n2).

The spherical harmonic (SPHARM) description is a powerful surface modeling technique that can model arbitrarily shaped but simply connected 3D objects and has been used in many applications in medical imaging. Previous SPHARM techniques use the first order ellipsoid for establishing surface correspondence and aligning objects. However, this first order information may not be sufficient in many cases; a more general method for establishing surface correspondence would be to minimize the mean squared distance between two corresponding surfaces. In this paper, a new surface matching algorithm is proposed for 3D SPHARM models to achieve this goal. This algorithm employs a useful rotational property of spherical harmonic basis functions for a fast implementation. Applications of medical image analysis (e.g., spatio-temporal modeling of heart shape changes) are used to demonstrate this approach. Theoretical proofs and experimental results show that our approach is an accurate and flexible surface correspondence alignment method.

We propose a novel framework to predict pacing sites in the left ventricle (LV) of a heart and its result can be used to assist pacemaker implantation and programming in cardiac resynchronization therapy (CRT), a widely adopted therapy for heart failure patients. In a traditional CRT device deployment, pacing sites are selected without quantitative prediction. That runs the risk of suboptimal benefits. In this work, the spherical harmonic (SPHARM) description is employed to model the ventricular surfaces and a novel SPHARM-based surface correspondence approach is proposed to capture the ventricular wall motion. A hierarchical agglomerative clustering technique is applied to the time series of regional wall thickness to identify candidate pacing sites. Using clinical MRI data in our experiments, we demonstrate that the proposed framework can not only effectively identify suitable pacing sites, but also distinguish patients from normal subjects perfectly to help medical diagnosis and prognosis.