Dynamic time warping outperforms Pearson correlation in detecting atypical functional connectivity in autism spectrum disorders

Studying time-resolved functional connectivity via communication theory: on the complementary nature of phase synchronization and sliding window Pearson correlation

Time-resolved functional connectivity (trFC) assesses the time-resolved coupling between brain regions using functional magnetic resonance imaging (fMRI) data. This study aims to compare two techniques used to estimate trFC, to investigate their similarities and differences when applied to fMRI data. These techniques are the sliding window Pearson correlation (SWPC), an amplitude-based approach, and phase synchronization (PS), a phase-based technique. To accomplish our objective, we used resting-state fMRI data from the Human Connectome Project (HCP) with 827 subjects (repetition time: 0.7s) and the Function Biomedical Informatics Research Network (fBIRN) with 311 subjects (repetition time: 2s), which included 151 schizophrenia patients and 160 controls. Our simulations reveal distinct strengths in two connectivity methods: SWPC captures high-magnitude, low-frequency connectivity, while PS detects low-magnitude, high-frequency connectivity. Stronger correlations between SWPC and PS align with pronounced fMRI oscillations. For fMRI data, higher correlations between SWPC and PS occur with matched frequencies and smaller SWPC window sizes (~30s), but larger windows (~88s) sacrifice clinically relevant information. Both methods identify a schizophrenia-associated brain network state but show different patterns: SWPC highlights low anti-correlations between visual, subcortical, auditory, and sensory-motor networks, while PS shows reduced positive synchronization among these networks. In sum, our findings underscore the complementary nature of SWPC and PS, elucidating their respective strengths and limitations without implying the superiority of one over the other.

Extracting interpretable signatures of whole-brain dynamics through systematic comparison

The brain's complex distributed dynamics are typically quantified using a limited set of manually selected statistical properties, leaving the possibility that alternative dynamical properties may outperform those reported for a given application. Here, we address this limitation by systematically comparing diverse, interpretable features of both intra-regional activity and inter-regional functional coupling from resting-state functional magnetic resonance imaging (rs-fMRI) data, demonstrating our method using case-control comparisons of four neuropsychiatric disorders. Our findings generally support the use of linear time-series analysis techniques for rs-fMRI case-control analyses, while also identifying new ways to quantify informative dynamical fMRI structures. While simple statistical representations of fMRI dynamics performed surprisingly well (e.g., properties within a single brain region), combining intra-regional properties with inter-regional coupling generally improved performance, underscoring the distributed, multifaceted changes to fMRI dynamics in neuropsychiatric disorders. The comprehensive, data-driven method introduced here enables systematic identification and interpretation of quantitative dynamical signatures of multivariate time-series data, with applicability beyond neuroimaging to diverse scientific problems involving complex time-varying systems.

Assessing Predictive Ability of Dynamic Time Warping Functional Connectivity for ASD Classification

Functional connectivity MRI (fcMRI) is a technique used to study the functional connectedness of distinct regions of the brain by measuring the temporal correlation between their blood oxygen level-dependent (BOLD) signals. fcMRI is typically measured with the Pearson correlation (PC), which assumes that there is no lag between time series. Dynamic time warping (DTW) is an alternative measure of similarity between time series that is robust to such time lags. We used PC fcMRI data and DTW fcMRI data as predictors in machine learning models for classifying autism spectrum disorder (ASD). When combined with dimension reduction techniques, such as principal component analysis, functional connectivity estimated with DTW showed greater predictive ability than functional connectivity estimated with PC. Our results suggest that DTW fcMRI can be a suitable alternative measure that may be characterizing fcMRI in a different, but complementary, way to PC fcMRI that is worth continued investigation. In studying different variants of cross validation (CV), our results suggest that, when it is necessary to tune model hyperparameters and assess model performance at the same time, a K-fold CV nested within leave-one-out CV may be a competitive contender in terms of performance and computational speed, especially when sample size is not large.

Amortization Transformer for Brain Effective Connectivity Estimation from fMRI Data

Using machine learning methods to estimate brain effective connectivity networks from functional magnetic resonance imaging (fMRI) data has garnered significant attention in the fields of neuroinformatics and bioinformatics. However, existing methods usually require retraining the model for each subject, which ignores the knowledge shared across subjects. In this paper, we propose a novel framework for estimating effective connectivity based on an amortization transformer, named AT-EC. In detail, AT-EC first employs an amortization transformer to model the dynamics of fMRI time series and infer brain effective connectivity across different subjects, which can train an amortized model that leverages the shared knowledge from different subjects. Then, an assisted learning mechanism based on functional connectivity is designed to assist the estimation of the brain effective connectivity network. Experimental results on both simulated and real-world data demonstrate the efficacy of our method.

Integrating Economic Theory, Domain Knowledge, and Social Knowledge into Hybrid Sentiment Models for Predicting Crude Oil Markets

For several decades, sentiment analysis has been considered a key indicator for assessing market mood and predicting future price changes. Accurately predicting commodity markets requires an understanding of fundamental market dynamics such as the interplay between supply and demand, which are not considered in standard affective models. This paper introduces two domain-specific affective models, CrudeBERT and CrudeBERT+, that adapt sentiment analysis to the crude oil market by incorporating economic theory with common knowledge of the mentioned entities and social knowledge extracted from Google Trends. To evaluate the predictive capabilities of these models, comprehensive experiments were conducted using dynamic time warping to identify the model that best approximates WTI crude oil futures price movements. The evaluation included news headlines and crude oil prices between January 2012 and April 2021. The results show that CrudeBERT+ outperformed RavenPack, BERT, FinBERT, and early CrudeBERT models during the 9-year evaluation period and within most of the individual years that were analyzed. The success of the introduced domain-specific affective models demonstrates the potential of integrating economic theory with sentiment analysis and external knowledge sources to improve the predictive power of financial sentiment analysis models. The experiments also confirm that CrudeBERT+ has the potential to provide valuable insights for decision-making in the crude oil market.

ERP evidence of heightened attentional response to visual stimuli in migraine headache disorders

New findings from migraine studies have indicated that this common headache disorder is associated with anomalies in attentional processing. In tandem with the previous explorations, this study will provide evidence to show that visual attention is impacted by migraine headache disorders. 43 individuals were initially recruited in the migraine group and 33 people with non-migraine headache disorders were in the control group. The event-related potentials (ERP) of the participants were calculated using data from a visual oddball paradigm task. By analyzing the N200 and P300 ERP components, migraineurs, as compared to controls, had an exaggerated oddball response showing increased amplitude in N200 and P300 difference scores for the oddball vs. standard, while the latencies of the two components remained the same in the migraine and control groups. We then looked at two classifications of migraine with and without aura compared to non-migraine controls. One-Way ANOVA analysis of the two migraine groups and the non-migraine control group showed that the different level of N200 and P300 amplitude mean scores was greater between migraineurs without aura and the control group while these components’ latency remained the same relatively in the three groups. Our results give more neurophysiological support that people with migraine headaches have altered processing of visual attention.

Identification of Alzheimer’s Disease Progression Stages Using Topological Measures of Resting-State Functional Connectivity Networks: A Comparative Study

Resting-state functional magnetic resonance imaging (rs-fMRI) has been widely employed to examine brain functional connectivity (FC) alterations in various neurological disorders. At present, various computational methods have been proposed to estimate connectivity strength between different brain regions, as the edge weight of FC networks. However, little is known about which model is more sensitive to Alzheimer's disease (AD) progression. This study comparatively characterized topological properties of rs-FC networks constructed with Pearson correlation (PC), dynamic time warping (DTW), and group information guided independent component analysis (GIG-ICA), aimed at investigating the sensitivity and effectivity of these methods in differentiating AD stages. A total of 54 subjects from Alzheimer's Disease Neuroimaging Initiative (ANDI) database, divided into healthy control (HC), mild cognition impairment (MCI), and AD groups, were included in this study. Network-level (global efficiency and characteristic path length) and nodal (clustering coefficient) metrics were used to capture groupwise difference across HC, MCI, and AD groups. The results showed that almost no significant differences were found according to global efficiency and characteristic path length. However, in terms of clustering coefficient, 52 brain parcels sensitive to AD progression were identified in rs-FC networks built with GIG-ICA, much more than PC (6 parcels) and DTW (3 parcels). This indicates that GIG-ICA is more sensitive to AD progression than PC and DTW. The findings also confirmed that the AD-linked FC alterations mostly appeared in temporal, cingulate, and angular areas, which might contribute to clinical diagnosis of AD. Overall, this study provides insights into the topological properties of rs-FC networks over AD progression, suggesting that FC strength estimation of FC networks cannot be neglected in AD-related graph analysis.

Estimation of Pb Content Using Reflectance Spectroscopy in Farmland Soil near Metal Mines, Central China

The contamination of farmlands with hazardous metals from mining puts the safety of agricultural commodities at risk. For remediation, it is crucial to map the spatial distribution of contaminated soil. Typical sampling-based procedures are time-consuming and labor-intensive. The use of visible, near-infrared, and short-wave infrared reflectance (VNIR-SWIR) spectroscopy to detect soil heavy metal pollution is an alternative. With the aim of investigating a methodology of detecting the most sensitive bands using VNIR-SWIR spectra to find lead (Pb) anomalies in agriculture soil near mining activities, the area in Xiaoqinling Mountain, downstream from a series of active gold mines, was selected to test the feasibility of utilizing VNIR-SWIR spectroscopy to map soil Pb. A total of 115 soil samples were collected for laboratory Pb analysis and spectral measurement. Partial least squares regression (PLSR) was adopted to estimate the soil Pb content by building the prediction model, and the model was optimized by finding the optimal number of bands involved. The spatial distribution of Pb concentration was mapped using the ordinary kriging (OK) interpolation method. This study found that five spectral bands (522 nm, 1668 nm, 2207 nm, 2296 nm, and 2345 nm) were sensitive to soil Pb content. The optimized prediction model’s coefficient of determination (R2), residual prediction deviation (RPD), and root mean square error (RMSE) were 0.711, 1.860, and 0.711 ln(mg/kg), respectively. Additionally, the result of OK interpolation was convincing and accurate (R2 = 0.775, RMSE = 0.328 ln(mg/kg)), comparing maps from estimated and ground truth data. This study proves that it is feasible to use VNIR-SWIR spectral data for in situ estimation of the soil Pb content.

Dynamic Time Warping Identifies Functionally Distinct fMRI Resting State Cortical Networks Specific to VTA and SNc: A Proof of Concept

Functional connectivity (FC) is determined by similarity between functional magnetic resonance imaging (fMRI) signals from distinct brain regions. However, traditional FC analyses ignore temporal phase differences. Here, we addressed this limitation, using dynamic time warping (DTW) within a machine-learning framework, to study cortical FC patterns of 2 spatially adjacent but functionally distinct subcortical regions, namely Substantia Nigra Pars Compacta (SNc) and ventral tegmental area (VTA). We evaluate: 1) the influence of pair of brain regions considered, 2) the influence of warping window sizes, 3) the classification efficacy of DTW, and 4) the uniqueness of features identified. Whole brain 7 Tesla resting state fMRI scans from 81 healthy participants were used. FC between 2 subcortical regions of interests (ROIs) and 360 cortical parcels were computed using: 1) Pearson correlations (PCs), 2) dynamic time-warped PCs (DTW-PC). The separability of SNc-cortical and VTA-cortical network was validated on 40 participants and tested on the remaining 41, using a support vector machine (SVM). The SVM separated the SNc-cortical versus VTA-cortical network with 74.39 and 97.56% test accuracy using PC and DTW-PC, respectively. SVM-recursive feature elimination yielded 20 DTW-PC features that most strongly contributed to the separation of the networks and revealed novel VTA versus SNc preferential connections (P < 0.05, Bonferroni-Holm corrected).