89 research outputs found
Fusion of Multitask fMRI Data with Constrained Independent Vector Analysis
No full text2025 59th Annual Conference on Information Sciences and Systems (CISS) Baltimore, MD, 19-21 March 2025Functional magnetic resonance imaging (fMRI) is a widely used neuroimaging tool for investigating brain function. In multitask fMRI analysis, data fusion methods enable the integration of information across tasks to provide a comprehensive understanding of brain activity. Independent vector analysis (IVA) provides an attractive framework for data fusion as it enables datasets to fully interact with each other by maximizing statistical dependence across the datasets. IVA with multivariate Laplacian distribution IVA-L provides a good model match for fMRI analysis as fMRI signals often exhibit multivariate heavy-tailed distributions. However, IVA can benefit from incorporating prior information when available. This paper proposes a novel way for multitask fMRI data fusion by integrating prior information into an optimized IVA-L framework using a constrained cost function. The proposed method is applied to a multitask fMRI dataset comprising 271 subjects, successfully identifying task-related group differences between healthy controls and schizophrenia patients. Identified important functional areas include the caudate and thalamus during the sensory-motor task (SM), as well as the inferior parietal lobule, superior medial frontal gyrus, and inferior frontal gyrus during the auditory oddball (AOD) task. Additionally, this work highlights the importance of selecting a higher model order and allowing some components to remain unconstrained for the constrained IVA-L framework. These choices enhance the estimation performance and allow the algorithm to capture important information not included in the prior information.Emin Erdem Kumbasar and Hanlu Yang contrributed equally to this work. This work was supported in part by grants NSF 2316420, NIH R01MH118695, NIH R01MH123610, and NIH R01AG073949https://ieeexplore.ieee.org/abstract/document/1094469
Data-Driven Techniques for Inference in Large-Scale fMRI Datasets: Homogeneous Subgroup Identification and Multi-Subject Analysis
No full textThe availability of large-scale, open-source neuroimaging datasets has significantly expanded opportunities for brain research. By jointly analyzing multisubject data from such repositories, researchers can draw group inferences across cohorts, enhance our understanding of brain function, and identify potential biomarkers or subtypes of different disorders. Additionally, large-scale datasets facilitate the detection of subtle effects that may not be statistically discernible in smaller cohorts. However, analyzing such data poses challenges due to high dimensionality, inter-subject variability, and the computational demands of existing methods, which grow with dataset size. While these frameworks are designed to effectively capture subject differences, their computational cost increases as the number of subjects grows. This dissertation addresses these challenges by developing data-driven techniques that efficiently analyze large-scale datasets, extract meaningful and reproducible features, and optimize computational performance. Using multi-subject resting-state fMRI (rs-fMRI) as a case study, we demonstrate the effectiveness of these methods in applications such as homogeneous subgroup identification and biomarker detection. Our proposed techniques preserve subject variability while maintaining computational efficiency, enabling the identification of clinically meaningful subgroups and biomarkers from various large psychiatric cohorts. We begin by introducing foundational concepts in fMRI data analysis and commonly used techniques such as blind source separation (BSS) and joint BSS (JBSS) methods. We provide studies of subgroup identification from multi-subject rs-fMRI data, highlighting the advantages of JBSS techniques in preserving subject variability. We propose to model the cross-functional network information as a multiplex network and enhance the subgroup identification performance by taking the multi-dimensional information into account. To address computational complexity limitations of the current JBSS methods, we develop methods that enhance computational efficiency while preserving subject variability. These techniques position data-driven and model-driven approaches as two ends of a spectrum, seeking an optimal balance in between by either flexible constraint selection schemes or a representative coreset strategy. Furthermore, we extend the coreset concept to higher-dimensional data, developing an efficient tensor-based method for complex fMRI research tasks such as dynamic functional network analysis over time. We conclude by summarizing our proposed large-scale data analysis techniques and providing guidelines for selecting appropriate methods based on specific research needs
POCS Augmented CycleGAN for MR Image Reconstruction
No full textTraditional Magnetic Resonance Imaging (MRI) reconstruction methods, which may be highly time-consuming and sensitive to noise, heavily depend on solving nonlinear optimization problems. By contrast, deep learning (DL)-based reconstruction methods do not need any explicit analytical data model and are robust to noise due to its large data-based training, which both make DL a versatile tool for fast and high-fidelity MR image reconstruction. While DL can be performed completely independently of traditional methods, it can, in fact, benefit from incorporating these established methods to achieve better results. To test this hypothesis, we proposed a hybrid DL-based MR image reconstruction method, which combines two state-of-the-art deep learning networks, U-Net and Generative Adversarial Network with Cycle loss (CycleGAN), with a traditional data reconstruction method: Projection Onto Convex Sets (POCS). Experiments were then performed to evaluate the method by comparing it to several existing state-of-the-art methods. Our results demonstrate that the proposed method outperformed the current state-of-the-art in terms of higher peak signal-to-noise ratio (PSNR) and higher Structural Similarity Index (SSIM).Electrical and Computer Engineerin
A036: Research on the Impact of Physical Dance on Body Dissatisfaction Among Female College Students
No full textBackground/Purpose: With the exacerbation of social phenomena such as internal competition and labeling, mental health issues among university students, especially anxiety and depression among female students, are becoming increasingly prominent. “Body Dissatisfaction”, recognized as a key factor influencing mental health, has garnered widespread attention from various sectors of society. This study aimed to investigate the impact of physical dance on body dissatisfaction among female college students, with the goal of fostering positive body image and promoting healthy exercise habits.
Method: This study employed a combination of literature review, experimental methods, interviews, field observations, questionnaire surveys, and mathematical statistics. Three measurement tools, namely the Brief Negative Body Image Scale (BNBIS), the Physical Self-Perception Profile (PSPP), and the Body Image States Scale (BISS), were utilized. A total of 45 female college students with no background in sports dance and no history of major illnesses were selected for an 8-week experimental teaching intervention in physical dance. Data were collected and organized before, during, and after the experiment, with descriptive statistics and paired-sample t-tests conducted using SPSS 26.0 to analyze the data required for this paper.
Results: The experimental results revealed a significant improvement among female college students in terms of physical fitness, body condition, and physical attractiveness (P \u3c 0.05). There was a highly significant improvement in overall dissatisfaction factors (P \u3c 0.01), despite the relatively subtle changes observed in dissatisfaction factors related to height and weight (P \u3e 0.05), indicating a positive trend. Furthermore, there was a significant elevation in the level of body image states post-intervention (P \u3c 0.05).
Conclusion: The research highlights the effectiveness of an 8-week moderate-intensity physical dance elective course in reducing body dissatisfaction among female college students, thereby positively impacting their physical and mental well-being. This intervention not only broadens the spectrum of sports activities conducive to improving body image but also encourages a holistic approach to self-care and fosters a positive attitude towards physical activity. However, the study recognizes the limitation of its sample size and emphasizes the need for future research to expand samples and implement stringent experimental controls. The findings of this study carry significant theoretical and practical implications, advocating for the integration of physical dance elective courses into university curricula to promote healthy body images and enhance overall well-being among female students
Large-Scale Independent Vector Analysis (IVA-G) via Coresets
No full textJoint blind source separation (JBSS) involves the factorization of multiple matrices, i.e. “datasets”, into “sources” that are statistically dependent across datasets and independent within datasets. Despite this usefulness for analyzing multiple datasets, JBSS methods suffer from considerable computational costs and are typically intractable for hundreds or thousands of datasets. To address this issue, we present a methodology for how a subset of the datasets can be used to perform efficient JBSS over the full set. We motivate two such methods: a numerical extension of independent vector analysis (IVA) with the multivariate Gaussian model (IVA-G), and a recently proposed analytic method resembling generalized joint diagonalization (GJD). We derive nonidentifiability conditions for both methods, and then demonstrate how one can significantly improve these methods’ generalizability by an efficient representative subset selection method. This involves selecting a coreset (a weighted subset) that minimizes a measure of discrepancy between the statistics of the coreset and the full set. Using simulated and real functional magnetic resonance imaging (fMRI) data, we demonstrate significant scalability and source separation advantages of our “coreIVA-G” method vs. other JBSS methods.https://ieeexplore.ieee.org/document/10798966/authors#author
Mode Coresets for Efficient, Interpretable Tensor Decompositions: An Application to Feature Selection in fMRI Analysis
No full textGeneralizations of matrix decompositions to multidimensional arrays, called tensor decompositions, are simple yet powerful methods for analyzing datasets in the form of tensors. These decompositions model a data tensor as a sum of rank-1 tensors, whose factors provide uses for a myriad of applications. Given the massive sizes of modern datasets, an important challenge is how well computational complexity scales with the data, balanced with how well decompositions approximate the data. Many efficient methods exploit a small subset of the tensor’s elements, representing most of the tensor’s variation via a basis over the subset. These methods’ efficiencies are often due to their randomized natures; however, deterministic methods can provide better approximations, and can perform feature selection, highlighting a meaningful subset that well-represents the entire tensor. In this paper, we introduce an efficient subset-based form of the Tucker decomposition, by selecting coresets from the tensor modes such that the resulting core tensor can well-approximate the full tensor. Furthermore, our method enables a novel feature selection scheme unlike other methods for tensor data. We introduce methods for random and deterministic coresets, minimizing error via a measure of discrepancy between the coreset and full tensor. We perform the decompositions on simulated data, and perform on real-world fMRI data to demonstrate our method’s feature selection ability. We demonstrate that compared with other similar decomposition methods, our methods can typically better approximate the tensor with comparably low computational complexities.This work was supported in part by NSF under Grant 2316420; in part by NIH under Grant R01MH118695, Grant R01MH123610, and Grant R01AG073949; in part by the Computational Hardware used is part of the University of Maryland, Baltimore County (UMBC) High Performance Computing Facility (HPCF) funded by the U.S. NSF through the MRI and SCREMS Programs under Grant CNS-0821258, Grant CNS-1228778, Grant OAC-1726023, and Grant DMS-0821311; and in part by UMBC.https://ieeexplore.ieee.org/document/10798430/authors#author
The Threshold Temperature and Lag Effects on Daily Excess Mortality in Harbin, China: A Time Series Analysis
No full textAbstract 3779: Evaluation of efficacy of an RNA aptamer toward non-small cell lung cancer
No full textDeep Learning Method Based on Spectral Characteristic Rein-Forcement for the Extraction of Winter Wheat Planting Area in Complex Agricultural Landscapes
No full textWinter wheat is one of the most important food crops in the world. Remote sensing technology can be used to obtain the spatial distribution and planting area of winter wheat in a timely and accurate manner, which is of great significance for agricultural management. Influenced by the growth conditions of winter wheat, the planting structures of the northern and southern regions differ significantly. Therefore, in this study, the spectral and phenological characteristics of winter wheat were analyzed in detail, and four red-edge vegetation indices (NDVI, NDRE, SRre, and CIred-edge) were included after band analysis to enhance the ability of the characteristics to extract winter wheat. These indices were combined with a deep convolutional neural network (CNN) model to achieve intelligent extraction of the winter wheat planting area in a countable number of complex agricultural landscapes. Using this method, GF-6 WFV and Sentinel-2A remote sensing data were used to obtain full coverage of the region to evaluate the geographical environment differences. This spectral characteristic enhancement method combined with a CNN could extract the winter wheat data well for both data sources, with average overall accuracies of 94.01 and 93.03%, respectively. This study proposes a method for fast and accurate extraction of winter wheat in complex agricultural landscapes that can provide decision support for national and local intelligent agricultural construction. Thus, our study has important application value and practical significance
Investigation on the relationship between renal NONO expression, fibrosis and prognosis in diabetic nephropathy
No full textBackgroundRenal interstitial fibrosis (RIF) is an important manifestation of Diabetic nephropathy (DN) progression. Non-POU domain containing octamer-binding protein (NONO) is crucial in fibrosis in cardiovascular diseases, but its role in DN fibrosis remains unclear. This study explores the expression of NONO in DN and its correlation with Matrix Metalloproteinase-9 (MMP-9, as an important regulator of fibrosis), renal fibrosis, and prognosis.MethodsForty patients with type 2 diabetes mellitus (T2DM) with pathologically confirmed DN were included, divided into early DN group (n=20) and late DN group (n=20). 6 normal renal tissue as control group. HE, Masson staining, immunohistochemical staining and Immunofluorescence double staining were performed. The correlation between NONO expression levels and MMP-9 as well as clinical pathological data was analyzed. Cox regression analysis and Kaplan-Meier survival curves were used to evaluate the relationship between renal tissue NONO expression levels and DN prognosis.ResultsCompared with control group, NONO expression levels in renal tissues of DN patient were increased, and the late DN group was higher than the early DN group (P<0.05). NONO and MMP-9 expression were positively correlated with multiple clinical and Fibrosis-related pathological indicators, and NONO expression was positively correlated with MMP-9(P<0.05). Patients with high renal NONO expression had lower kidney progression-free survival rates.ConclusionsNONO expression levels correlate positively with MMP-9, collagen and renal damage indicators in renal tissues of DN patients. High NONO expression is linked to poor renal prognosis in DN. NONO may contribute to renal tissue fibrosis in DN by regulating MMP-9 levels
- …
