Bayesian network analysis reveals alterations to default mode network connectivity in individuals at risk for Alzheimer's disease

Changes in Brain Functional and Effective Connectivity After Treatment for Breast Cancer and Implications for Intervention Targets

Background: Patients with breast cancer frequently report cognitive impairment both during and after completion of therapy. Evidence suggests that cancer-related cognitive impairments are related to widespread neural network dysfunction. The default mode network (DMN) is a large conserved network that plays a critical role in integrating the functions of various neural systems. Disruption of the network may play a key role in the development of cognitive impairment. Methods: We compared neuroimaging and neurocognitive data from 43 newly diagnosed primary breast cancer patients (mean age = 48, standard deviation [SD] = 8.9 years) and 50 frequency-matched healthy female controls (mean age = 50, SD = 10 years) before treatment and 1 year after treatment completion. Functional and effective connectivity measures of the DMN were obtained using graph theory and Bayesian network analysis methods, respectively. Results: Compared with healthy females, the breast cancer group displayed higher global efficiency and path length post-treatment (p < 0.03, corrected). Breast cancer survivors showed significantly lower performance on measures of verbal memory, attention, and verbal fluency (p < 0.05) at both time points. Within the DMN, local brain network organization, as measured by edge-betweenness centralities, was significantly altered in the breast cancer group compared with controls at both time points (p < 0.0001, corrected), with several connections showing a significant group-by-time effect (p < 0.003, corrected). Effective connectivity demonstrated significantly altered patterns of neuronal coupling in patients with breast cancer (p < 0.05). Significant correlations were seen between hormone blockade therapy, radiation therapy, chemotherapy cycles, memory, and verbal fluency test and edge-betweenness centralities. Discussion: This pattern of altered network organization in the default mode is believed to result in reduced network efficiency and disrupted communication. Subregions of the DMN, the orbital prefrontal cortex and posterior memory network, appear to be at the center of this disruption and this could inform future interventions. Impact statement This prospective study is the first to investigate how post-treatment changes in functional and effective connectivity in the regions of default mode network are related to cancer therapy and measures of memory and verbal learning in breast cancer patients. We demonstrate that the interactions between treatment, brain connectivity, and neurocognitive outcomes coalesce around a subgroup of brain structures in the orbital frontal and parietal lobe. This would suggest that interventions that target these regions may improve neurocognitive outcomes in breast cancer survivors.

Dissociable Functional Brain Networks Associated With Apathy in Subcortical Ischemic Vascular Disease and Alzheimer’s Disease

Few studies have investigated differences in functional connectivity (FC) between patients with subcortical ischemic vascular disease (SIVD) and Alzheimer’s disease (AD), especially in relation to apathy. Therefore, the aim of this study was to compare apathy-related FC changes among patients with SIVD, AD, and cognitively normal subjects. The SIVD group had the highest level of apathy as measured using the Apathy Evaluation Scale-clinician version (AES). Dementia staging, volume of white matter hyperintensities (WMH), and the Beck Depression Inventory were the most significant clinical predictors for apathy. Group-wise comparisons revealed that the SIVD patients had the worst level of “Initiation” by factor analysis of the AES. FCs from four resting state networks (RSNs) were compared, and the connectograms at the level of intra- and inter-RSNs revealed dissociable FC changes, shared FC in the dorsal attention network, and distinct FC in the salient network across SIVD and AD. Neuronal correlates for “Initiation” deficits that underlie apathy were explored through a regional-specific approach, which showed that the right inferior frontal gyrus, left middle frontal gyrus, and left anterior insula were the critical hubs. These findings broaden the disconnection theory by considering the effect of FC interactions across multiple RSNs on apathy formation.

Corrected score methods for estimating Bayesian networks with error-prone nodes

Motivated by inferring cellular signaling networks using noisy flow cytometry data, we develop procedures to draw inference for Bayesian networks based on error-prone data. Two methods for inferring causal relationships between nodes in a network are proposed based on penalized estimation methods that account for measurement error and encourage sparsity. We discuss consistency of the proposed network estimators and develop an approach for selecting the tuning parameter in the penalized estimation methods. Empirical studies are carried out to compare the proposed methods with a naive method that ignores measurement error. Finally, we apply these methods to infer signaling networks using single cell flow cytometry data.

Spectral dynamic causal modelling of resting-state fMRI: an exploratory study relating effective brain connectivity in the default mode network to genetics

We conduct an imaging genetics study to explore how effective brain connectivity in the default mode network (DMN) may be related to genetics within the context of Alzheimer's disease and mild cognitive impairment. We develop an analysis of longitudinal resting-state functional magnetic resonance imaging (rs-fMRI) and genetic data obtained from a sample of 111 subjects with a total of 319 rs-fMRI scans from the Alzheimer's Disease Neuroimaging Initiative (ADNI) database. A Dynamic Causal Model (DCM) is fit to the rs-fMRI scans to estimate effective brain connectivity within the DMN and related to a set of single nucleotide polymorphisms (SNPs) contained in an empirical disease-constrained set which is obtained out-of-sample from 663 ADNI subjects having only genome-wide data. We relate longitudinal effective brain connectivity estimated using spectral DCM to SNPs using both linear mixed effect (LME) models as well as function-on-scalar regression (FSR). In both cases we implement a parametric bootstrap for testing SNP coefficients and make comparisons with p-values obtained from asymptotic null distributions. In both networks at an initial q-value threshold of 0.1 no effects are found. We report on exploratory patterns of associations with relatively high ranks that exhibit stability to the differing assumptions made by both FSR and LME.

Resting State Abnormalities of the Default Mode Network in Mild Cognitive Impairment: A Systematic Review and Meta-Analysis

BACKGROUND

Large-scale brain networks such as the default mode network (DMN) are often disrupted in Alzheimer's disease (AD). Numerous studies have examined DMN functional connectivity in those with mild cognitive impairment (MCI), a presumed AD precursor, to discover a biomarker of AD risk. Prior reviews were qualitative or limited in scope or approach.

OBJECTIVE

We aimed to systematically and quantitatively review DMN resting state fMRI studies comparing MCI and healthy comparison (HC) groups.

METHODS

PubMed was searched for relevant articles. Study characteristics were abstracted and the number of studies showing no group difference or hyper- versus hypo-connnectivity in MCI was tallied. A voxel-wise (ES-SDM) meta-analysis was conducted to identify regional group differences.

RESULTS

Qualitatively, our review of 57 MCI versus HC comparisons suggests substantial inconsistency; 9 showed no group difference, 8 showed MCI > HC and 22 showed HC > MCI across the brain, and 18 showed regionally-mixed directions of effect. The meta-analysis of 31 studies revealed areas of significant hypo- and hyper-connectivity in MCI, including hypoconnectivity in the posterior cingulate cortex/precuneus (z = -3.1, p < 0.0001). Very few individual studies, however, showed patterns resembling the meta-analytic results. Methodological differences did not appear to explain inconsistencies.

CONCLUSIONS

The pattern of altered resting DMN function or connectivity in MCI is complex and variable across studies. To date, no index of DMN connectivity qualifies as a useful biomarker of MCI or risk for AD. Refinements to MCI diagnosis, including other biological markers, or longitudinal studies of progression to AD, might identify DMN alterations predictive of AD risk.

The fronto-insular cortex causally mediates the default-mode and central-executive networks to contribute to individual cognitive performance in healthy elderly

The triple network model that consists of the default-mode network (DMN), central-executive network (CEN), and salience network (SN) has been suggested as a powerful paradigm for investigation of network mechanisms underlying various cognitive functions and brain disorders. A crucial hypothesis in this model is that the fronto-insular cortex (FIC) in the SN plays centrally in mediating interactions between the networks. Using a machine learning approach based on independent component analysis and Bayesian network (BN), this study characterizes the directed connectivity architecture of the triple network and examines the role of FIC in connectivity of the model. Data-driven exploration shows that the FIC initiates influential connections to all other regions to globally control the functional dynamics of the triple network. Moreover, stronger BN connectivity between the FIC and regions of the DMN and the CEN, as well as the increased outflow connections from the FIC are found to predict individual performance in memory and executive tasks. In addition, the posterior cingulate cortex in the DMN was also confirmed as an inflow hub that integrates information converging from other areas. Collectively, the results highlight the central role of FIC in mediating the activity of large-scale networks, which is crucial for individual cognitive function.

Linking Inter-Individual Variability in Functional Brain Connectivity to Cognitive Ability in Elderly Individuals

Increasing evidence suggests that functional brain connectivity is an important determinant of cognitive aging. However, the fundamental concept of inter-individual variations in functional connectivity in older individuals is not yet completely understood. It is essential to evaluate the extent to which inter-individual variability in connectivity impacts cognitive performance at an older age. In the current study, we aimed to characterize individual variability of functional connectivity in the elderly and to examine its significance to individual cognition. We mapped inter-individual variability of functional connectivity by analyzing whole-brain functional connectivity magnetic resonance imaging data obtained from a large sample of cognitively normal older adults. Our results demonstrated a gradual increase in variability in primary regions of the visual, sensorimotor, and auditory networks to specific subcortical structures, particularly the hippocampal formation, and the prefrontal and parietal cortices, which largely constitute the default mode and fronto-parietal networks, to the cerebellum. Further, the inter-individual variability of the functional connectivity correlated significantly with the degree of cognitive relevance. Regions with greater connectivity variability demonstrated more connections that correlated with cognitive performance. These results also underscored the crucial function of the long-range and inter-network connections in individual cognition. Thus, individual connectivity-cognition variability mapping findings may provide important information for future research on cognitive aging and neurocognitive diseases.

Disrupted Causal Connectivity Anchored in the Posterior Cingulate Cortex in Amnestic Mild Cognitive Impairment

Amnestic mild cognitive impairment (aMCI) is a transitional stage between normal cognitive aging and Alzheimer’s disease. Previous studies have found that neuronal activity and functional connectivity impaired in many functional networks, especially in the default mode network (DMN), which is related to significantly impaired cognitive and memory functions in aMCI patients. However, few studies have focused on the effective connectivity of the DMN and its subsystems in aMCI patients. The posterior cingulate cortex (PCC) is considered a crucial region in connectivity of the DMN and its key subsystem. In this study, using the coefficient Granger causality analysis approach and using the PCC as the region of interest, we explored changes in the DMN and its subsystems in effective connectivity with other brain regions as well as in correlations among them in 16 aMCI patients and 15 age-matched cognitively normal elderly. Results showed decreased effective connectivity from PCC to whole brain in the left prefrontal cortex, the left medial temporal lobe (MTL), the left fusiform gyrus (FG), and the left cerebellar hemisphere, meanwhile, right temporal lobe showed increased effective connectivity from PCC to the whole brain in aMCI patients compared with normal control. In addition, compared with the normal controls, increased effective connectivity of the whole brain to the PCC in aMCI patients was found in the right thalamus, left medial temporal lobe, left FG, and left cerebellar hemisphere. Compared with the normal controls, no reduced effective connectivity was found in any brain regions from the whole brain to the PCC in aMCI patients. The reduced effective connectivity of the PCC to left MTL showed negative correlation trend with neuropsychological tests (Auditory Verbal Learning Test-immediate recall and clock drawing test) in aMCI patients. Our study shows that aMCI patients have abnormalities in effective connectivity within the PCC-centered DMN network and its posterior subsystems as well as in the cerebellar hemisphere and thalamus. Abnormal integration of networks may be related to cognitive and memory impairment and compensation mechanisms in aMCI patients.

Mild Cognitive Impairment Is Not "Mild" at All in Altered Activation of Episodic Memory Brain Networks: Evidence from ALE Meta-Analysis

The present study conducted a quantitative meta-analysis aiming at assessing consensus across the functional neuroimaging studies of episodic memory in individuals with amnestic mild cognitive impairment (aMCI) and elucidating consistent activation patterns. An activation likelihood estimation (ALE) was conducted on the functional neuroimaging studies of episodic encoding and retrieval in aMCI individuals published up to March 31, 2015. Analyses covered 24 studies, which yielded 770 distinct foci. Compared to healthy controls, aMCI individuals showed statistically significant consistent activation differences in a widespread episodic memory network, not only in the bilateral medial temporal lobe and prefrontal cortex, but also in the angular gyrus, precunes, posterior cingulate cortex, and even certain more basic structures. The present ALE meta-analysis revealed that the abnormal patterns of widespread episodic memory network indicated that individuals with aMCI may not be completely "mild" in nature.

Learning Discriminative Bayesian Networks from High-Dimensional Continuous Neuroimaging Data

Due to its causal semantics, Bayesian networks (BN) have been widely employed to discover the underlying data relationship in exploratory studies, such as brain research. Despite its success in modeling the probability distribution of variables, BN is naturally a generative model, which is not necessarily discriminative. This may cause the ignorance of subtle but critical network changes that are of investigation values across populations. In this paper, we propose to improve the discriminative power of BN models for continuous variables from two different perspectives. This brings two general discriminative learning frameworks for Gaussian Bayesian networks (GBN). In the first framework, we employ Fisher kernel to bridge the generative models of GBN and the discriminative classifiers of SVMs, and convert the GBN parameter learning to Fisher kernel learning via minimizing a generalization error bound of SVMs. In the second framework, we employ the max-margin criterion and build it directly upon GBN models to explicitly optimize the classification performance of the GBNs. The advantages and disadvantages of the two frameworks are discussed and experimentally compared. Both of them demonstrate strong power in learning discriminative parameters of GBNs for neuroimaging based brain network analysis, as well as maintaining reasonable representation capacity. The contributions of this paper also include a new Directed Acyclic Graph (DAG) constraint with theoretical guarantee to ensure the graph validity of GBN.

A Triple Network Connectivity Study of Large-Scale Brain Systems in Cognitively Normal APOE4 Carriers

The triple network model, consisting of the central executive network (CEN), salience network (SN) and default mode network (DMN), has been recently employed to understand dysfunction in core networks across various disorders. Here we used the triple network model to investigate the large-scale brain networks in cognitively normal apolipoprotein e4 (APOE4) carriers who are at risk of Alzheimer’s disease (AD). To explore the functional connectivity for each of the three networks and the effective connectivity among them, we evaluated 17 cognitively normal individuals with a family history of AD and at least one copy of the APOE4 allele and compared the findings to those of 12 individuals who did not carry the APOE4 gene or have a family history of AD, using independent component analysis (ICA) and Bayesian network (BN) approach. Our findings indicated altered within-network connectivity that suggests future cognitive decline risk, and preserved between-network connectivity that may support their current preserved cognition in the cognitively normal APOE4 allele carriers. The study provides novel sights into our understanding of the risk factors for AD and their influence on the triple network model of major psychopathology.

Classification of patients with MCI and AD from healthy controls using directed graph measures of resting-state fMRI

Brain network alterations in patients with Alzheimer's disease (AD) has been the subject of much investigation, but the biological mechanisms underlying these alterations remain poorly understood. Here, we aim to identify the changes in brain networks in patients with AD and mild cognitive impairment (MCI), and provide an accurate algorithm for classification of these patients from healthy control subjects (HC) by using a graph theoretical approach and advanced machine learning methods. Multivariate Granger causality analysis was performed on resting-state functional magnetic resonance imaging (rs-fMRI) data of 34 AD, 89 MCI, and 45 HC to calculate various directed graph measures. The graph measures were used as the original feature set for the machine learning algorithm. Filter and wrapper feature selection methods were applied to the original feature set to select an optimal subset of features. An accuracy of 93.3% was achieved for classification of AD, MCI, and HC using the optimal features and the naïve Bayes classifier. We also performed a hub node analysis and found that the number of hubs in HC, MCI, and AD were 12, 10, and 9, respectively, suggesting that patients with AD experience disturbance of critical communication areas in their brain network as AD progresses. The findings of this study provide insight into the neurophysiological mechanisms underlying MCI and AD. The proposed classification method highlights the potential of directed graph measures of rs-fMRI data for identification of the early stage of AD.

Progressive Bidirectional Age-Related Changes in Default Mode Network Effective Connectivity across Six Decades

The default mode network (DMN) is a set of regions that is tonically engaged during the resting state and exhibits task-related deactivation that is readily reproducible across a wide range of paradigms and modalities. The DMN has been implicated in numerous disorders of cognition and, in particular, in disorders exhibiting age-related cognitive decline. Despite these observations, investigations of the DMN in normal aging are scant. Here, we used blood oxygen level dependent (BOLD) functional magnetic resonance imaging (fMRI) acquired during rest to investigate age-related changes in functional connectivity of the DMN in 120 healthy normal volunteers comprising six, 20-subject, decade cohorts (from 20–29 to 70–79). Structural equation modeling (SEM) was used to assess age-related changes in inter-regional connectivity within the DMN. SEM was applied both using a previously published, meta-analytically derived, node-and-edge model, and using exploratory modeling searching for connections that optimized model fit improvement. Although the two models were highly similar (only 3 of 13 paths differed), the sample demonstrated significantly better fit with the exploratory model. For this reason, the exploratory model was used to assess age-related changes across the decade cohorts. Progressive, highly significant changes in path weights were found in 8 (of 13) paths: four rising, and four falling (most changes were significant by the third or fourth decade). In all cases, rising paths and falling paths projected in pairs onto the same nodes, suggesting compensatory increases associated with age-related decreases. This study demonstrates that age-related changes in DMN physiology (inter-regional connectivity) are bidirectional, progressive, of early onset and part of normal aging.

A Complex Systems Approach to Causal Discovery in Psychiatry

Conventional research methodologies and data analytic approaches in psychiatric research are unable to reliably infer causal relations without experimental designs, or to make inferences about the functional properties of the complex systems in which psychiatric disorders are embedded. This article describes a series of studies to validate a novel hybrid computational approach–the Complex Systems-Causal Network (CS-CN) method–designed to integrate causal discovery within a complex systems framework for psychiatric research. The CS-CN method was first applied to an existing dataset on psychopathology in 163 children hospitalized with injuries (validation study). Next, it was applied to a much larger dataset of traumatized children (replication study). Finally, the CS-CN method was applied in a controlled experiment using a ‘gold standard’ dataset for causal discovery and compared with other methods for accurately detecting causal variables (resimulation controlled experiment). The CS-CN method successfully detected a causal network of 111 variables and 167 bivariate relations in the initial validation study. This causal network had well-defined adaptive properties and a set of variables was found that disproportionally contributed to these properties. Modeling the removal of these variables resulted in significant loss of adaptive properties. The CS-CN method was successfully applied in the replication study and performed better than traditional statistical methods, and similarly to state-of-the-art causal discovery algorithms in the causal detection experiment. The CS-CN method was validated, replicated, and yielded both novel and previously validated findings related to risk factors and potential treatments of psychiatric disorders. The novel approach yields both fine-grain (micro) and high-level (macro) insights and thus represents a promising approach for complex systems-oriented research in psychiatry.

Bayesian graphical network analyses reveal complex biological interactions specific to Alzheimer's disease

With different approaches to finding prognostic or diagnostic biomarkers for Alzheimer's disease (AD), many studies pursue only brief lists of biomarkers or disease specific pathways, potentially dismissing information from groups of correlated biomarkers. Using a novel Bayesian graphical network method, with data from the Australian Imaging, Biomarkers and Lifestyle (AIBL) study of aging, the aim of this study was to assess the biological connectivity between AD associated blood-based proteins. Briefly, three groups of protein markers (18, 37, and 48 proteins, respectively) were assessed for the posterior probability of biological connection both within and between clinical classifications. Clinical classification was defined in four groups: high performance healthy controls (hpHC), healthy controls (HC), participants with mild cognitive impairment (MCI), and participants with AD. Using the smaller group of proteins, posterior probabilities of network similarity between clinical classifications were very high, indicating no difference in biological connections between groups. Increasing the number of proteins increased the capacity to separate both hpHC and HC apart from the AD group (0 for complete separation, 1 for complete similarity), with posterior probabilities shifting from 0.89 for the 18 protein group, through to 0.54 for the 37 protein group, and finally 0.28 for the 48 protein group. Using this approach, we identified beta-2 microglobulin (β2M) as a potential master regulator of multiple proteins across all classifications, demonstrating that this approach can be used across many data sets to identify novel insights into diseases like AD.

Directed progression brain networks in Alzheimer's disease: properties and classification

This article introduces a new approach in brain connectomics aimed at characterizing the temporal spread in the brain of pathologies like Alzheimer's disease (AD). The main instrument is the development of "directed progression networks" (DPNets), wherein one constructs directed edges between nodes based on (weakly) inferred directions of the temporal spreading of the pathology. This stands in contrast to many previously studied brain networks where edges represent correlations, physical connections, or functional progressions. In addition, this is one of a few studies showing the value of using directed networks in the study of AD. This article focuses on the construction of DPNets for AD using longitudinal cortical thickness measurements from magnetic resonance imaging data. The network properties are then characterized, providing new insights into AD progression, as well as novel markers for differentiating normal cognition (NC) and AD at the group level. It also demonstrates the important role of nodal variations for network classification (i.e., the significance of standard deviations, not just mean values of nodal properties). Finally, the DPNets are utilized to classify subjects based on their global network measures using a variety of data-mining methodologies. In contrast to most brain networks, these DPNets do not show high clustering and small-world properties.

Construct validation of a DCM for resting state fMRI

Recently, there has been a lot of interest in characterising the connectivity of resting state brain networks. Most of the literature uses functional connectivity to examine these intrinsic brain networks. Functional connectivity has well documented limitations because of its inherent inability to identify causal interactions. Dynamic causal modelling (DCM) is a framework that allows for the identification of the causal (directed) connections among neuronal systems — known as effective connectivity. This technical note addresses the validity of a recently proposed DCM for resting state fMRI – as measured in terms of their complex cross spectral density – referred to as spectral DCM. Spectral DCM differs from (the alternative) stochastic DCM by parameterising neuronal fluctuations using scale free (i.e., power law) forms, rendering the stochastic model of neuronal activity deterministic. Spectral DCM not only furnishes an efficient estimation of model parameters but also enables the detection of group differences in effective connectivity, the form and amplitude of the neuronal fluctuations or both. We compare and contrast spectral and stochastic DCM models with endogenous fluctuations or state noise on hidden states. We used simulated data to first establish the face validity of both schemes and show that they can recover the model (and its parameters) that generated the data. We then used Monte Carlo simulations to assess the accuracy of both schemes in terms of their root mean square error. We also simulated group differences and compared the ability of spectral and stochastic DCMs to identify these differences. We show that spectral DCM was not only more accurate but also more sensitive to group differences. Finally, we performed a comparative evaluation using real resting state fMRI data (from an open access resource) to study the functional integration within default mode network using spectral and stochastic DCMs.

•

This paper provides construct validation of spectral DCM against stochastic DCM.

•

Spectral DCM is shown to be more accurate than stochastic DCM in terms of root mean square error.

•

Spectral DCM is shown to be more sensitive at identifying group differences.

Bayesian networks in neuroscience: a survey

Bayesian networks are a type of probabilistic graphical models lie at the intersection between statistics and machine learning. They have been shown to be powerful tools to encode dependence relationships among the variables of a domain under uncertainty. Thanks to their generality, Bayesian networks can accommodate continuous and discrete variables, as well as temporal processes. In this paper we review Bayesian networks and how they can be learned automatically from data by means of structure learning algorithms. Also, we examine how a user can take advantage of these networks for reasoning by exact or approximate inference algorithms that propagate the given evidence through the graphical structure. Despite their applicability in many fields, they have been little used in neuroscience, where they have focused on specific problems, like functional connectivity analysis from neuroimaging data. Here we survey key research in neuroscience where Bayesian networks have been used with different aims: discover associations between variables, perform probabilistic reasoning over the model, and classify new observations with and without supervision. The networks are learned from data of any kind-morphological, electrophysiological, -omics and neuroimaging-, thereby broadening the scope-molecular, cellular, structural, functional, cognitive and medical- of the brain aspects to be studied.

Bayesian network analysis revealed the connectivity difference of the default mode network from the resting-state to task-state

Functional magnetic resonance imaging (fMRI) studies have converged to reveal the default mode network (DMN), a constellation of regions that display co-activation during resting-state but co-deactivation during attention-demanding tasks in the brain. Here, we employed a Bayesian network (BN) analysis method to construct a directed effective connectivity model of the DMN and compared the organizational architecture and interregional directed connections under both resting-state and task-state. The analysis results indicated that the DMN was consistently organized into two closely interacting subsystems in both resting-state and task-state. The directed connections between DMN regions, however, changed significantly from the resting-state to task-state condition. The results suggest that the DMN intrinsically maintains a relatively stable structure whether at rest or performing tasks but has different information processing mechanisms under varied states.

Study of the contribution of the state of water to the gel properties of a photocrosslinked polyacrylic acid hydrogel using magnetic resonance imaging

Photocrosslinked polyacrylic acid (PAA-HEMA) hydrogels are a promising candidate for use in dermatological patch adhesives. To gain further knowledge about the properties of this gel, we investigated the T1 relaxation time and the diffusion coefficient (D) of water in the hydrogels using magnetic resonance (MR) imaging. Hydrogels with different formulations and process factors were prepared and tested. The observed data were analyzed by ANOVA, which clarified the mode of action of the formulation and process factors based on these MR parameters. Various gel properties (i.e., gel fraction, swelling capacity, gel strength, and water-retention ability) were also measured, followed by a Bayesian network (BN) analysis. The BN allowed us to summarize well the relationships between the formulation and process factors, MR parameters, and gel properties. T1 was associated with the swelling and water-retention properties of the hydrogel, whereas D was associated with gel formation and gel strength. Furthermore, this study clarified that T1 and D mostly represented the hydration and water-compartmentalization effects of the hydrogel, respectively. In conclusion, the state of water seems to play an important role in the properties of the PAA-HEMA hydrogel.