scholarly journals Consecutive Independence and Correlation Transform for Multimodal Data Fusion: Discovery of One-to-Many Associations in Structural and Functional Imaging Data

2021 ◽  
Vol 11 (18) ◽  
pp. 8382
Author(s):  
Chunying Jia ◽  
Mohammad Abu Baker Siddique Akhonda ◽  
Yuri Levin-Schwartz ◽  
Qunfang Long ◽  
Vince D. Calhoun ◽  
...  

Brain signals can be measured using multiple imaging modalities, such as magnetic resonance imaging (MRI)-based techniques. Different modalities convey distinct yet complementary information; thus, their joint analyses can provide valuable insight into how the brain functions in both healthy and diseased conditions. Data-driven approaches have proven most useful for multimodal fusion as they minimize assumptions imposed on the data, and there are a number of methods that have been developed to uncover relationships across modalities. However, none of these methods, to the best of our knowledge, can discover “one-to-many associations”, meaning one component from one modality is linked with more than one component from another modality. However, such “one-to-many associations” are likely to exist, since the same brain region can be involved in multiple neurological processes. Additionally, most existing data fusion methods require the signal subspace order to be identical for all modalities—a severe restriction for real-world data of different modalities. Here, we propose a new fusion technique—the consecutive independence and correlation transform (C-ICT) model—which successively performs independent component analysis and independent vector analysis and is uniquely flexible in terms of the number of datasets, signal subspace order, and the opportunity to find “one-to-many associations”. We apply C-ICT to fuse diffusion MRI, structural MRI, and functional MRI datasets collected from healthy controls (HCs) and patients with schizophrenia (SZs). We identify six interpretable triplets of components, each of which consists of three associated components from the three modalities. Besides, components from these triplets that show significant group differences between the HCs and SZs are identified, which could be seen as putative biomarkers in schizophrenia.

Author(s):  
Wuyi Wang ◽  
Simon Zhornitsky ◽  
Sheng Zhang ◽  
Chiang-shan R. Li

AbstractPreclinical studies have implicated noradrenergic (NA) dysfunction in cocaine addiction. In particular, the NA system plays a central role in motivated behavior and may partake in the regulation of craving and drug use. Yet, human studies of the NA system are scarce, likely hampered by the difficulty in precisely localizing the locus coeruleus (LC). Here, we used neuromelanin imaging to localize the LC and quantified LC neuromelanin signal (NMS) intensity in 44 current cocaine users (CU; 37 men) and 59 nondrug users (NU; 44 men). We also employed fMRI to investigate cue-induced regional responses and LC functional connectivities, as quantified by generalized psychophysiological interaction (gPPI), in CU. Imaging data were processed by published routines and the findings were evaluated with a corrected threshold. We examined how these neural measures were associated with chronic cocaine craving, as assessed by the Cocaine Craving Questionnaire (CCQ). Compared to NU, CU demonstrated higher LC NMS for all probabilistic thresholds defined of 50–90% of the peak. In contrast, NMS of the ventral tegmental area/substantia nigra (VTA/SN) did not show significant group differences. Drug as compared to neutral cues elicited higher activations of many cortical and subcortical regions, none of which were significantly correlated with CCQ score. Drug vs. neutral cues also elicited “deactivation” of bilateral parahippocampal gyri (PHG) and PHG gPPI with a wide array of cortical and subcortical regions, including the ventral striatum and, with small volume correction, the LC. Less deactivation of the PHG (r = 0.40, p = 0.008) and higher PHG-LC gPPI (r = 0.44, p = 0.003) were positively correlated with the CCQ score. In contrast, PHG-VTA/SN connectivity did not correlate with the CCQ score. Together, chronic cocaine exposure may induce higher NMS intensity, suggesting neurotoxic effects on the LC. The correlation of cue-elicited PHG LC connectivity with CCQ score suggests a noradrenergic correlate of chronic cocaine craving. Potentially compensating for memory functions as in neurodegenerative conditions, cue-elicited PHG LC circuit connectivity plays an ill-adaptive role in supporting cocaine craving.


2018 ◽  
Author(s):  
Danilo Bzdok ◽  
Denis Engemann ◽  
Olivier Grisel ◽  
Gaël Varoquaux ◽  
Bertrand Thirion

AbstractIn the 20th century many advances in biological knowledge and evidence-based medicine were supported by p-values and accompanying methods. In the beginning 21st century, ambitions towards precision medicine put a premium on detailed predictions for single individuals. The shift causes tension between traditional methods used to infer statistically significant group differences and burgeoning machine-learning tools suited to forecast an individual’s future. This comparison applies the linear model for identifying significant contributing variables and for finding the most predictive variable sets. In systematic data simulations and common medical datasets, we explored how statistical inference and pattern recognition can agree and diverge. Across analysis scenarios, even small predictive performances typically coincided with finding underlying significant statistical relationships. However, even statistically strong findings with very low p-values shed little light on their value for achieving accurate prediction in the same dataset. More complete understanding of different ways to define ‘important’ associations is a prerequisite for reproducible research findings that can serve to personalize clinical care.


2020 ◽  
Author(s):  
Patrick Wehrli ◽  
Wojciech Michno ◽  
Laurent Guerard ◽  
Julia Fernandez-Rodriguez ◽  
Anders Bergh ◽  
...  

<p>Imaging mass spectrometry (IMS) is a powerful tool for spatially-resolved chemical analysis and thereby offers novel perspectives for applications in biology and medicine. The understanding of chemically complex systems, such as biological tissues, benefits from the combination of multiple imaging modalities contributing with complementary molecular information. Effective analysis and interpretation of multimodal IMS data is challenging and requires both, precise alignment and combination of the imaging data as well as suitable statistical analysis methods to identify cross-modal correlations. Commonly applied IMS data analysis methods include qualitative comparative analysis where cross-modal interpretation is subject to human judgement; Workflows that incorporate image registration procedures are usually applied for co-representing data rather than to mine data across modalities. </p><p>Here, we present an IMS-based, histology-driven strategy for comprehensive interrogation of biological tissues by spatial chemometrics. Our workflow implements a 1+1-evolutionary image registration method enabling direct correlation of chemical information across multiple modalities at single pixel resolution. Comprehensive multimodal imaging data were evaluated using a novel approach based on orthogonal multiblock component analysis (OnPLS). Finally, we present a novel image fusion method by implementing consecutively acquired pathological staining data to enhance histological interpretation.</p><p>We demonstrate the method’s potential in two biomedical applications where trimodal matrix-assisted laser desorption/ionization (MALDI) IMS delineates pathology associated co-localization patterns of lipids and proteins in (1) a transgenic Alzheimer’s disease (AD) mouse model, and in (2) a human xenograft rat model of prostate cancer. The presented image analysis paradigm allows to comprehensively interrogate complex biological systems with single pixel resolution at cellular length scales.</p>


Author(s):  
Timo Mäkelä ◽  
Quoc-Cuong Pham ◽  
Patrick Clarysse ◽  
Jyrki Lötjönen ◽  
Kirsi Lauerma ◽  
...  

Molecules ◽  
2019 ◽  
Vol 24 (12) ◽  
pp. 2301 ◽  
Author(s):  
Liqun Kuang ◽  
Deyu Zhao ◽  
Jiacheng Xing ◽  
Zhongyu Chen ◽  
Fengguang Xiong ◽  
...  

Recent research of persistent homology in algebraic topology has shown that the altered network organization of human brain provides a promising indicator of many neuropsychiatric disorders and neurodegenerative diseases. However, the current slope-based approach may not accurately characterize changes of persistent features over graph filtration because such curves are not strictly linear. Moreover, our previous integrated persistent feature (IPF) works well on an rs-fMRI cohort while it has not yet been studied on metabolic brain networks. To address these issues, we propose a novel univariate network measurement, kernel-based IPF (KBI), based on the prior IPF, to quantify the difference between IPF curves. In our experiments, we apply the KBI index to study fluorodeoxyglucose positron emission tomography (FDG-PET) imaging data from 140 subjects with Alzheimer’s disease (AD), 280 subjects with mild cognitive impairment (MCI), and 280 healthy normal controls (NC). The results show the disruption of network integration in the progress of AD. Compared to previous persistent homology-based measures, as well as other standard graph-based measures that characterize small-world organization and modular structure, our proposed network index KBI possesses more significant group difference and better classification performance, suggesting that it may be used as an effective preclinical AD imaging biomarker.


2020 ◽  
Vol 10 (1) ◽  
Author(s):  
Michele Allegra ◽  
Elena Facco ◽  
Francesco Denti ◽  
Alessandro Laio ◽  
Antonietta Mira

Abstract One of the founding paradigms of machine learning is that a small number of variables is often sufficient to describe high-dimensional data. The minimum number of variables required is called the intrinsic dimension (ID) of the data. Contrary to common intuition, there are cases where the ID varies within the same data set. This fact has been highlighted in technical discussions, but seldom exploited to analyze large data sets and obtain insight into their structure. Here we develop a robust approach to discriminate regions with different local IDs and segment the points accordingly. Our approach is computationally efficient and can be proficiently used even on large data sets. We find that many real-world data sets contain regions with widely heterogeneous dimensions. These regions host points differing in core properties: folded versus unfolded configurations in a protein molecular dynamics trajectory, active versus non-active regions in brain imaging data, and firms with different financial risk in company balance sheets. A simple topological feature, the local ID, is thus sufficient to achieve an unsupervised segmentation of high-dimensional data, complementary to the one given by clustering algorithms.


NeuroImage ◽  
2010 ◽  
Vol 51 (1) ◽  
pp. 123-134 ◽  
Author(s):  
Jing Sui ◽  
Tülay Adali ◽  
Godfrey Pearlson ◽  
Honghui Yang ◽  
Scott R. Sponheim ◽  
...  

NeuroImage ◽  
2016 ◽  
Vol 134 ◽  
pp. 486-493 ◽  
Author(s):  
Yuri Levin-Schwartz ◽  
Yang Song ◽  
Peter J. Schreier ◽  
Vince D. Calhoun ◽  
Tülay Adalı

Sign in / Sign up

Export Citation Format

Share Document