Multiple Holdouts With Stability: Improving the Generalizability of Machine Learning Analyses of Brain-Behavior Relationships

Interpretable and integrative deep learning for discovering brain-behaviour associations

Recent advances highlight the limitations of classification strategies in machine learning that rely on a single data source for understanding, diagnosing and predicting psychiatric syndromes. Moreover, approaches based solely on clinician labels often fail to capture the complexity and variability of these conditions. Recent research underlines the importance of considering multiple dimensions that span across different psychiatric syndromes. These developments have led to more comprehensive approaches to studying psychiatric conditions that incorporate diverse data sources such as imaging, genetics, and symptom reports. Multi-view unsupervised learning frameworks, particularly deep learning models, present promising solutions for integrating and analysing complex datasets. Such models contain generative capabilities which facilitate the exploration of relationships between different data views. In this study, we propose a robust framework for interpreting these models that combines digital avatars with stability selection to assess these relationships. We apply this framework to the Healthy Brain Network cohort which includes clinical behavioural scores and brain imaging features, uncovering a consistent set of brain-behaviour interactions. These associations link cortical measurements obtained from structural MRI with clinical reports evaluating psychiatric symptoms. Our framework effectively identifies relevant and stable associations, even with incomplete datasets, while isolating variability of interest from confounding factors.

Estimating correlations in low-reliability settings with constrained hierarchical models

It is popular to study individual differences in cognition with experimental tasks, and the main goal of such approaches is to analyze the pattern of correlations across a battery of tasks and measures. One difficulty is that experimental tasks are often low in reliability as effects are small relative to trial-by-trial variability. Consequently, it remains difficult to accurately estimate correlations. One approach that seems attractive is hierarchical modeling where trial-by-trial variability and variability across conditions, tasks, and individuals are modeled separately. Here we show that hierarchical models may reduce the error in estimating correlations up to 43%, but only if substantive constraint is imposed. The approach here is Bayesian, and we develop novel Bayesian hierarchical factor models for experiments where trials are nested in conditions, tasks, and individuals. The prior on covariances across tasks can either be unconstrained, in which there is little error reduction, or constrained, in which there is substantial error reduction. The constraints are: (1) There is a low-dimension factor structure underlying the covariation across tasks, and (2) all loadings are non-negative leading to a positive manifold on correlations. We argue that both of these assumptions are reasonable in cognitive domains, and that with them, researchers may profitably use hierarchical models to estimate correlations across tasks in low-reliability settings.

Elevated Activity in Left Homologous Music Circuits Is Inhibitory for Music Perception but Mediated by Structure-Function Coupling

AIMS

Previous studies suggested that structural and functional connectivity of right frontotemporal circuits associate with music perception. Emerging evidences demonstrated that structure-function coupling is important for cognition and may allow for a more sensitive investigation of brain-behavior association, while we know little about the relationship between structure-function coupling and music perception.

METHODS

We collected multimodal neuroimaging data from 106 participants and measured their music perception by Montreal Battery of Evaluation of Amusia (MBEA). Then we computed structure-function coupling, amplitude of low-frequency fluctuation (ALFF), gray matter volume (GMV), and structural/functional degree centrality (DC) and utilized support vector regression algorithm to build their relationship with MBEA score.

RESULTS

We found structure-function coupling, rather than GMV, ALFF, or DC, contributed to predict MBEA score. Left middle frontal gyrus (L.MFG), bilateral inferior temporal gyrus, and right insula were the most predictive ROIs for MBEA score. Mediation analysis revealed structure-function coupling of L.MFG, a region that is homologous to typical music circuits, fully mediated the negative link between ALFF of L.MFG and MBEA score.

CONCLUSION

Structure-function coupling is more effective when explaining variation in music perception. Our findings provide further understanding for the neural basis of music and have implications for cognitive causes of amusia.

Association between clinical features and decreased degree centrality and variability in dynamic functional connectivity in the obsessive-compulsive disorder

Neuroimaging studies have indicated widespread brain structural and functional disruptions in patients with obsessive-compulsive disorder (OCD). However, the underlying mechanism of these changes remains unclear. A total of 45 patients with OCD and 42 healthy controls (HC) were enrolled. The study investigated local degree centrality (DC) abnormalities and employed abnormal regions of DC as seeds to investigate variability in dynamic functional connectivity (dFC) in the whole brain using a sliding window approach to analyze resting-state functional magnetic resonance imaging. The relationship between abnormal DC and dFC as well as the clinical features of OCD were examined using correlation analysis. Our findings suggested decreased DC in the bilateral thalamus, bilateral precuneus, and bilateral cuneus in OCD patients and a nominally negative correlation between the DC value in the thalamus and illness severity measured using the Yale-Brown Obsessive Compulsive Scale (Y-BOCS). In addition, seed-based dFC analysis showed that compared to measurements in the HC, the patients had decreased dFC variability between the left thalamus and the left cuneus and right lingual gyrus, and between the bilateral cuneus and bilateral postcentral gyrus, and a nominally positive correlation between the duration of illness and dFC variability between the left cuneus and left postcentral gyrus. These results indicated that OCD patients had decreased hub importance in the bilateral thalamus and cuneus throughout the entire brain. This reduction was associated with impaired coupling with dynamic function in the visual cortex and sensorimotor network and provided novel insights into the neurophysiological mechanisms underlying OCD.

Voxelwise Multivariate Analysis of Brain-Psychosocial Associations in Adolescents Reveals 6 Latent Dimensions of Cognition and Psychopathology

BACKGROUND

Adolescence heralds the onset of considerable psychopathology, which may be conceptualized as an emergence of altered covariation between symptoms and brain measures. Multivariate methods can detect such modes of covariation or latent dimensions, but none specifically relating to psychopathology have yet been found using population-level structural brain data. Using voxelwise (instead of parcellated) brain data may strengthen latent dimensions' brain-psychosocial relationships, but this creates computational challenges.

METHODS

We obtained voxelwise gray matter density and psychosocial variables from the baseline (ages 9-10 years) Adolescent Brain Cognitive Development (ABCD) Study cohort (N = 11,288) and employed a state-of-the-art segmentation method, sparse partial least squares, and a rigorous machine learning framework to prevent overfitting.

RESULTS

We found 6 latent dimensions, 4 of which pertain specifically to mental health. The mental health dimensions were related to overeating, anorexia/internalizing, oppositional symptoms (all ps < .002) and attention-deficit/hyperactivity disorder symptoms (p = .03). Attention-deficit/hyperactivity disorder was related to increased and internalizing symptoms related to decreased gray matter density in dopaminergic and serotonergic midbrain areas, whereas oppositional symptoms were related to increased gray matter in a noradrenergic nucleus. Internalizing symptoms were related to increased and oppositional symptoms to reduced gray matter density in the insular, cingulate, and auditory cortices. Striatal regions featured strongly, with reduced caudate nucleus gray matter in attention-deficit/hyperactivity disorder and reduced putamen gray matter in oppositional/conduct problems. Voxelwise gray matter density generated stronger brain-psychosocial correlations than brain parcellations.

CONCLUSIONS

Voxelwise brain data strengthen latent dimensions of brain-psychosocial covariation, and sparse multivariate methods increase their psychopathological specificity. Internalizing and externalizing symptoms are associated with opposite gray matter changes in similar cortical and subcortical areas.

Brain Functional Connectome Defines a Transdiagnostic Dimension Shared by Cognitive Function and Psychopathology in Preadolescents

BACKGROUND

Cognitive function and general psychopathology are two important classes of human behavior dimensions that are individually related to mental disorders across diagnostic categories. However, whether these two transdiagnostic dimensions are linked to common or distinct brain networks that convey resilience or risk for the development of psychiatric disorders remains unclear.

METHODS

The current study is a longitudinal investigation with 11,875 youths from the Adolescent Brain Cognitive Development (ABCD) Study at ages 9 to 10 years at the onset of the study. A machine learning approach based on canonical correlation analysis was used to identify latent dimensional associations of the resting-state functional connectome with multidomain behavioral assessments including cognitive functions and psychopathological measures. For the latent resting-state functional connectivity factor showing a robust behavioral association, its ability to predict psychiatric disorders was assessed using 2-year follow-up data, and its genetic association was evaluated using twin data from the same cohort.

RESULTS

A latent functional connectome pattern was identified that showed a strong and generalizable association with the multidomain behavioral assessments (5-fold cross-validation: ρ = 0.68-0.73 for the training set [n = 5096]; ρ = 0.56-0.58 for the test set [n = 1476]). This functional connectome pattern was highly heritable (h2 = 74.42%, 95% CI: 56.76%-85.42%), exhibited a dose-response relationship with the cumulative number of psychiatric disorders assessed concurrently and at 2 years post-magnetic resonance imaging scan, and predicted the transition of diagnosis across disorders over the 2-year follow-up period.

CONCLUSIONS

These findings provide preliminary evidence for a transdiagnostic connectome-based measure that underlies individual differences in the development of psychiatric disorders during early adolescence.

Lower motor performance is linked with poor sleep quality, depressive symptoms, and grey matter volume alterations

Motor performance (MP) is essential for functional independence and well-being, particularly in later life. However, the relationship between behavioural aspects such as sleep quality and depressive symptoms, which contribute to MP, and the underlying structural brain substrates of their interplay remains unclear. This study used three population-based cohorts of younger and older adults (n=1,950) from the Human Connectome Project-Young Adult (HCP-YA), HCP-Aging (HCP-A), and enhanced Nathan Kline Institute-Rockland sample (eNKI-RS). Several canonical correlation analyses were computed within a machine learning framework to assess the associations between each of the three domains (sleep quality, depressive symptoms, grey matter volume (GMV)) and MP. The HCP-YA analyses showed progressively stronger associations between MP and each domain: depressive symptoms (unexpectedly positive, r=0.13, SD=0.06), sleep quality (r=0.17, SD=0.05), and GMV (r=0.19, SD=0.06). Combining sleep and depressive symptoms significantly improved the canonical correlations (r=0.25, SD=0.05), while the addition of GMV exhibited no further increase (r=0.23, SD=0.06). In young adults, better sleep quality, mild depressive symptoms, and GMV of several brain regions were associated with better MP. This was conceptually replicated in young adults from the eNKI-RS cohort. In HCP-Aging, better sleep quality, fewer depressive symptoms, and increased GMV were associated with MP. Robust multivariate associations were observed between sleep quality, depressive symptoms and GMV with MP, as well as age-related variations in these factors. Future studies should further explore these associations and consider interventions targeting sleep and mental health to test the potential effects on MP across the lifespan.

Multivariate brain-behaviour associations in psychiatric disorders

Mapping brain-behaviour associations is paramount to understand and treat psychiatric disorders. Standard approaches involve investigating the association between one brain and one behavioural variable (univariate) or multiple variables against one brain/behaviour feature ('single' multivariate). Recently, large multimodal datasets have propelled a new wave of studies that leverage on 'doubly' multivariate approaches capable of parsing the multifaceted nature of both brain and behaviour simultaneously. Within this movement, canonical correlation analysis (CCA) and partial least squares (PLS) emerge as the most popular techniques. Both seek to capture shared information between brain and behaviour in the form of latent variables. We provide an overview of these methods, review the literature in psychiatric disorders, and discuss the main challenges from a predictive modelling perspective. We identified 39 studies across four diagnostic groups: attention deficit and hyperactive disorder (ADHD, k = 4, N = 569), autism spectrum disorders (ASD, k = 6, N = 1731), major depressive disorder (MDD, k = 5, N = 938), psychosis spectrum disorders (PSD, k = 13, N = 1150) and one transdiagnostic group (TD, k = 11, N = 5731). Most studies (67%) used CCA and focused on the association between either brain morphology, resting-state functional connectivity or fractional anisotropy against symptoms and/or cognition. There were three main findings. First, most diagnoses shared a link between clinical/cognitive symptoms and two brain measures, namely frontal morphology/brain activity and white matter association fibres (tracts between cortical areas in the same hemisphere). Second, typically less investigated behavioural variables in multivariate models such as physical health (e.g., BMI, drug use) and clinical history (e.g., childhood trauma) were identified as important features. Finally, most studies were at risk of bias due to low sample size/feature ratio and/or in-sample testing only. We highlight the importance of carefully mitigating these sources of bias with an exemplar application of CCA.

Limited generalizability of multivariate brain-based dimensions of child psychiatric symptoms

Multivariate machine learning techniques are a promising set of tools for identifying complex brain-behavior associations. However, failure to replicate results from these methods across samples has hampered their clinical relevance. Here we aimed to delineate dimensions of brain functional connectivity that are associated with child psychiatric symptoms in two large and independent cohorts: the Adolescent Brain Cognitive Development (ABCD) Study and the Generation R Study (total n = 6935). Using sparse canonical correlations analysis, we identified two brain-behavior dimensions in ABCD: attention problems and aggression/rule-breaking behaviors. Importantly, out-of-sample generalizability of these dimensions was consistently observed in ABCD, suggesting robust multivariate brain-behavior associations. Despite this, out-of-study generalizability in Generation R was limited. These results highlight that the degrees of generalizability can vary depending on the external validation methods employed as well as the datasets used, emphasizing that biomarkers will remain elusive until models generalize better in true external settings.

On the stability of canonical correlation analysis and partial least squares with application to brain-behavior associations

Associations between datasets can be discovered through multivariate methods like Canonical Correlation Analysis (CCA) or Partial Least Squares (PLS). A requisite property for interpretability and generalizability of CCA/PLS associations is stability of their feature patterns. However, stability of CCA/PLS in high-dimensional datasets is questionable, as found in empirical characterizations. To study these issues systematically, we developed a generative modeling framework to simulate synthetic datasets. We found that when sample size is relatively small, but comparable to typical studies, CCA/PLS associations are highly unstable and inaccurate; both in their magnitude and importantly in the feature pattern underlying the association. We confirmed these trends across two neuroimaging modalities and in independent datasets with n ≈ 1000 and n = 20,000, and found that only the latter comprised sufficient observations for stable mappings between imaging-derived and behavioral features. We further developed a power calculator to provide sample sizes required for stability and reliability of multivariate analyses. Collectively, we characterize how to limit detrimental effects of overfitting on CCA/PLS stability, and provide recommendations for future studies.

Covariance patterns between sleep health domains and distributed intrinsic functional connectivity

Sleep health is both conceptually and operationally a composite concept containing multiple domains of sleep. In line with this, high dependence and interaction across different domains of sleep health encourage a transition in sleep health research from categorical to dimensional approaches that integrate neuroscience and sleep health. Here, we seek to identify the covariance patterns between multiple sleep health domains and distributed intrinsic functional connectivity by applying a multivariate approach (partial least squares). This multivariate analysis reveals a composite sleep health dimension co-varying with connectivity patterns involving the attentional and thalamic networks and which appear relevant at the neuromolecular level. These findings are further replicated and generalized to several unseen independent datasets. Critically, the identified sleep-health related connectome shows diagnostic potential for insomnia disorder. These results together delineate a potential brain connectome biomarker for sleep health with high potential for clinical translation.

Sampling inequalities affect generalization of neuroimaging-based diagnostic classifiers in psychiatry

BACKGROUND

The development of machine learning models for aiding in the diagnosis of mental disorder is recognized as a significant breakthrough in the field of psychiatry. However, clinical practice of such models remains a challenge, with poor generalizability being a major limitation.

METHODS

Here, we conducted a pre-registered meta-research assessment on neuroimaging-based models in the psychiatric literature, quantitatively examining global and regional sampling issues over recent decades, from a view that has been relatively underexplored. A total of 476 studies (n = 118,137) were included in the current assessment. Based on these findings, we built a comprehensive 5-star rating system to quantitatively evaluate the quality of existing machine learning models for psychiatric diagnoses.

RESULTS

A global sampling inequality in these models was revealed quantitatively (sampling Gini coefficient (G) = 0.81, p < .01), varying across different countries (regions) (e.g., China, G = 0.47; the USA, G = 0.58; Germany, G = 0.78; the UK, G = 0.87). Furthermore, the severity of this sampling inequality was significantly predicted by national economic levels (β =  - 2.75, p < .001, R²_adj = 0.40; r =  - .84, 95% CI: - .41 to - .97), and was plausibly predictable for model performance, with higher sampling inequality for reporting higher classification accuracy. Further analyses showed that lack of independent testing (84.24% of models, 95% CI: 81.0-87.5%), improper cross-validation (51.68% of models, 95% CI: 47.2-56.2%), and poor technical transparency (87.8% of models, 95% CI: 84.9-90.8%)/availability (80.88% of models, 95% CI: 77.3-84.4%) are prevailing in current diagnostic classifiers despite improvements over time. Relating to these observations, model performances were found decreased in studies with independent cross-country sampling validations (all p < .001, BF₁₀ > 15). In light of this, we proposed a purpose-built quantitative assessment checklist, which demonstrated that the overall ratings of these models increased by publication year but were negatively associated with model performance.

CONCLUSIONS

Together, improving sampling economic equality and hence the quality of machine learning models may be a crucial facet to plausibly translating neuroimaging-based diagnostic classifiers into clinical practice.

Functional connectivity signatures of major depressive disorder: machine learning analysis of two multicenter neuroimaging studies

The promise of machine learning has fueled the hope for developing diagnostic tools for psychiatry. Initial studies showed high accuracy for the identification of major depressive disorder (MDD) with resting-state connectivity, but progress has been hampered by the absence of large datasets. Here we used regular machine learning and advanced deep learning algorithms to differentiate patients with MDD from healthy controls and identify neurophysiological signatures of depression in two of the largest resting-state datasets for MDD. We obtained resting-state functional magnetic resonance imaging data from the REST-meta-MDD (N = 2338) and PsyMRI (N = 1039) consortia. Classification of functional connectivity matrices was done using support vector machines (SVM) and graph convolutional neural networks (GCN), and performance was evaluated using 5-fold cross-validation. Features were visualized using GCN-Explainer, an ablation study and univariate t-testing. The results showed a mean classification accuracy of 61% for MDD versus controls. Mean accuracy for classifying (non-)medicated subgroups was 62%. Sex classification accuracy was substantially better across datasets (73-81%). Visualization of the results showed that classifications were driven by stronger thalamic connections in both datasets, while nearly all other connections were weaker with small univariate effect sizes. These results suggest that whole brain resting-state connectivity is a reliable though poor biomarker for MDD, presumably due to disease heterogeneity as further supported by the higher accuracy for sex classification using the same methods. Deep learning revealed thalamic hyperconnectivity as a prominent neurophysiological signature of depression in both multicenter studies, which may guide the development of biomarkers in future studies.

Characterization of diffusion magnetic resonance imaging revealing relationships between white matter disconnection and behavioral disturbances in mild cognitive impairment: a systematic review

White matter disconnection is the primary cause of cognition and affection abnormality in mild cognitive impairment (MCI). Adequate understanding of behavioral disturbances, such as cognition and affection abnormality in MCI, can help to intervene and slow down the progression of Alzheimer's disease (AD) promptly. Diffusion MRI is a non-invasive and effective technique for studying white matter microstructure. This review searched the relevant papers published from 2010 to 2022. Sixty-nine studies using diffusion MRI for white matter disconnections associated with behavioral disturbances in MCI were screened. Fibers connected to the hippocampus and temporal lobe were associated with cognition decline in MCI. Fibers connected to the thalamus were associated with both cognition and affection abnormality. This review summarized the correspondence between white matter disconnections and behavioral disturbances such as cognition and affection, which provides a theoretical basis for the future diagnosis and treatment of AD.

Multivariate brain-based dimensions of child psychiatric problems: degrees of generalizability

Multivariate machine learning techniques are a promising set of tools for identifying complex brain-behavior associations. However, failure to replicate results from these methods across samples has hampered their clinical relevance. This study aimed to delineate dimensions of brain functional connectivity that are associated with child psychiatric symptoms in two large and independent cohorts: the Adolescent Brain Cognitive Development (ABCD) Study and the Generation R Study (total n =8,605). Using sparse canonical correlations analysis, we identified three brain-behavior dimensions in ABCD: attention problems, aggression and rule-breaking behaviors, and withdrawn behaviors. Importantly, out-of-sample generalizability of these dimensions was consistently observed in ABCD, suggesting robust multivariate brain-behavior associations. Despite this, out-of-study generalizability in Generation R was limited. These results highlight that the degree of generalizability can vary depending on the external validation methods employed as well as the datasets used, emphasizing that biomarkers will remain elusive until models generalize better in true external settings.

Patterns of connectome variability in autism across five functional activation tasks: findings from the LEAP project

BACKGROUND

Autism spectrum disorder (autism) is a complex neurodevelopmental condition with pronounced behavioral, cognitive, and neural heterogeneities across individuals. Here, our goal was to characterize heterogeneity in autism by identifying patterns of neural diversity as reflected in BOLD fMRI in the way individuals with autism engage with a varied array of cognitive tasks.

METHODS

All analyses were based on the EU-AIMS/AIMS-2-TRIALS multisite Longitudinal European Autism Project (LEAP) with participants with autism (n = 282) and typically developing (TD) controls (n = 221) between 6 and 30 years of age. We employed a novel task potency approach which combines the unique aspects of both resting state fMRI and task-fMRI to quantify task-induced variations in the functional connectome. Normative modelling was used to map atypicality of features on an individual basis with respect to their distribution in neurotypical control participants. We applied robust out-of-sample canonical correlation analysis (CCA) to relate connectome data to behavioral data.

RESULTS

Deviation from the normative ranges of global functional connectivity was greater for individuals with autism compared to TD in each fMRI task paradigm (all tasks p < 0.001). The similarity across individuals of the deviation pattern was significantly increased in autistic relative to TD individuals (p < 0.002). The CCA identified significant and robust brain-behavior covariation between functional connectivity atypicality and autism-related behavioral features.

CONCLUSIONS

Individuals with autism engage with tasks in a globally atypical way, but the particular spatial pattern of this atypicality is nevertheless similar across tasks. Atypicalities in the tasks originate mostly from prefrontal cortex and default mode network regions, but also speech and auditory networks. We show how sophisticated modeling methods such as task potency and normative modeling can be used toward unravelling complex heterogeneous conditions like autism.

Deriving psychiatric symptom-based biomarkers from multivariate relationships between psychophysiological and biochemical measures

Identification of biomarkers for psychiatric disorders remains very challenging due to substantial symptom heterogeneity and diagnostic comorbidity, limiting the ability to map symptoms to underlying neurobiology. Dimensional symptom clusters, such as anhedonia, hyperarousal, etc., are complex and arise due to interactions of a multitude of complex biological relationships. The primary aim of the current investigation was to use multi-set canonical correlation analysis (mCCA) to derive biomarkers (biochemical, physiological) linked to dimensional symptoms across the anxiety and depressive spectrum. Active-duty service members (N = 2,592) completed standardized depression, anxiety and posttraumatic stress questionnaires and several psychophysiological and biochemical assays. Using this approach, we identified two phenotype associations between distinct physiological and biological phenotypes. One was characterized by symptoms of dysphoric arousal (anhedonia, anxiety, hypervigilance) which was associated with low blood pressure and startle reactivity. This finding is in line with previous studies suggesting blunted physiological reactivity is associated with subpopulations endorsing anxiety with comorbid depressive features. A second phenotype of anxious fatigue (high anxiety and reexperiencing/avoidance symptoms coupled with fatigue) was associated with elevated blood levels of norepinephrine and the inflammatory marker C-reactive protein in conjunction with high blood pressure. This second phenotype may describe populations in which inflammation and high sympathetic outflow might contribute to anxious fatigue. Overall, these findings support the growing consensus that distinct neuropsychiatric symptom patterns are associated with differential physiological and blood-based biological profiles and highlight the potential of mCCA to reveal important psychiatric symptom biomarkers from several psychophysiological and biochemical measures.

A cross-cohort replicable and heritable latent dimension linking behaviour to multi-featured brain structure

Identifying associations between interindividual variability in brain structure and behaviour requires large cohorts, multivariate methods, out-of-sample validation and, ideally, out-of-cohort replication. Moreover, the influence of nature vs nurture on brain-behaviour associations should be analysed. We analysed associations between brain structure (grey matter volume, cortical thickness, and surface area) and behaviour (spanning cognition, emotion, and alertness) using regularized canonical correlation analysis and a machine learning framework that tests the generalisability and stability of such associations. The replicability of brain-behaviour associations was assessed in two large, independent cohorts. The load of genetic factors on these associations was analysed with heritability and genetic correlation. We found one heritable and replicable latent dimension linking cognitive-control/executive-functions and positive affect to brain structural variability in areas typically associated with higher cognitive functions, and with areas typically associated with sensorimotor functions. These results revealed a major axis of interindividual behavioural variability linking to a whole-brain structural pattern.

Canonical Correlation Analysis and Partial Least Squares for Identifying Brain-Behavior Associations: A Tutorial and a Comparative Study

Canonical correlation analysis (CCA) and partial least squares (PLS) are powerful multivariate methods for capturing associations across 2 modalities of data (e.g., brain and behavior). However, when the sample size is similar to or smaller than the number of variables in the data, standard CCA and PLS models may overfit, i.e., find spurious associations that generalize poorly to new data. Dimensionality reduction and regularized extensions of CCA and PLS have been proposed to address this problem, yet most studies using these approaches have some limitations. This work gives a theoretical and practical introduction into the most common CCA/PLS models and their regularized variants. We examine the limitations of standard CCA and PLS when the sample size is similar to or smaller than the number of variables. We discuss how dimensionality reduction and regularization techniques address this problem and explain their main advantages and disadvantages. We highlight crucial aspects of the CCA/PLS analysis framework, including optimizing the hyperparameters of the model and testing the identified associations for statistical significance. We apply the described CCA/PLS models to simulated data and real data from the Human Connectome Project and Alzheimer's Disease Neuroimaging Initiative (both of n > 500). We use both low- and high-dimensionality versions of these data (i.e., ratios between sample size and variables in the range of ∼1-10 and ∼0.1-0.01, respectively) to demonstrate the impact of data dimensionality on the models. Finally, we summarize the key lessons of the tutorial.

Research on Discrete Dynamic Modeling of Learner Behavior Analysis in English Teaching

The current English teaching mode focuses on the traditional offline teaching and online teaching. In order to solve the problems that some students are inefficient and cannot teach students according to their aptitude in the teaching process, this paper uses the big data analysis strategy based on a neural network algorithm. This paper studies the discrete dynamic modeling method of learner behavior analysis in English teaching. Firstly, it summarizes the current situation of English teaching and the research status of the hybrid application of discrete dynamic modeling technology. Secondly, combined with English teaching content and teaching objectives, through the analysis of various data of students' learning behavior, this paper evaluates students' English teaching quality from five aspects that affect the students' English teaching quality and puts forward a personalized English teaching quality evaluation model based on discrete dynamic modeling technology and learners' behavior analysis. Finally, through the practical teaching application in a university, the feasibility of the discrete dynamic English teaching model is verified. The results show that compared with the current innovative English teaching methods based on a dynamic iterative decision algorithm, the personalized discrete dynamic English teaching model based on learner behavior analysis significantly improves the quality of English teaching and students' academic performance.

Study Protocol of tDCS Based Pain Modulation in Head and Neck Cancer Patients Under Chemoradiation Therapy Condition: An fNIRS-EEG Study

Background

Multiple therapeutic strategies have been adopted to reduce pain, odynophagia, and oral mucositis in head and neck cancer patients. Among them, transcranial direct current stimulation (tDCS) represents a unique analgesic modality. However, the details of tDCS mechanisms in pain treatment are still unclear.

Aims

(1) to study the analgesic effects of a protocol that encompassed supervised-remote and in-clinic tDCS sessions applied in head and neck patients undergoing chemoradiation therapy; (2) to explore the underlining brain mechanisms of such modulation process, using a novel protocol that combined functional near-infrared spectroscopy (fNIRS), and electroencephalograph (EEG), two distinct neuroimaging methods that bring information regarding changes in the hemodynamic as well as in the electrical activity of the brain, respectively.

Methods

This proof-of-concept study was performed on two subjects. The study protocol included a 7-week-long tDCS stimulation procedure, a pre-tDCS baseline session, and two post-tDCS follow-up sessions. Two types of tDCS devices were used. One was used in the clinical setting and the other remotely. Brain imaging was obtained in weeks 1, 2, 5, 7, 8, and after 1 month.

Results

The protocol implemented was safe and reliable. Preliminary results of the fNIRS analysis in weeks 2 and 7 showed a decrease in functional connections between the bilateral prefrontal cortex (PFC) and the primary sensory cortex (S1) (p < 0.05, FDR corrected). Changes in EEG power spectra were found in the PFC when comparing the seventh with the first week of tDCS.

Conclusion

The protocol combining remote and in-clinic administered tDCS and integrated fNIRS and EEG to evaluate the brain activity is feasible. The preliminary results suggest that the mechanisms of tDCS in reducing the pain of head and neck cancer patients may be related to its effects on the connections between the S1 and the PFC.

Linking interindividual variability in brain structure to behaviour

What are the brain structural correlates of interindividual differences in behaviour? More than a decade ago, advances in structural MRI opened promising new avenues to address this question. The initial wave of research then progressively led to substantial conceptual and methodological shifts, and a replication crisis unveiled the limitations of traditional approaches, which involved searching for associations between local measurements of neuroanatomy and behavioural variables in small samples of healthy individuals. Given these methodological issues and growing scepticism regarding the idea of one-to-one mapping of psychological constructs to brain regions, new perspectives emerged. These not only embrace the multivariate nature of brain structure-behaviour relationships and promote generalizability but also embrace the representation of the relationships between brain structure and behavioural data by latent dimensions of interindividual variability. Here, we examine the past and present of the study of brain structure-behaviour associations in healthy populations and address current challenges and open questions for future investigations.

Decoding functional brain networks through graph measures in infancy: The case of emotional faces

Graph measures represent an optimal way to investigate neural networks' organization, yet their application is still limited in developmental samples. To uncover the organization of 7-month-old infants' functional brain networks during an emotional perception task, we combined a decoding technique (i.e., Principal Component Regression) to graph metrics computation. Nodes' Within Module Degree Z Score (WMDZ) was computed as a measure of modular organization, and we decoded networks' functional organizations across EEG alpha and theta bands in response to static and dynamic facial expressions of emotions. We found that infants' brain topological activity differentiates between static and dynamic emotional faces due to the involvement of visual streams and sensorimotor areas, as often observed in adults. Moreover, network invariances point toward an already present rudimental network structure tuned to face processing already at 7-months of age. Overall, our results affirm the fruitfulness of the application of graph measures in developmental samples, due to their flexibility and the wealth of information they provide over infants' networks functional organization.

A hierarchical Bayesian model to find brain-behaviour associations in incomplete data sets

Canonical Correlation Analysis (CCA) and its regularised versions have been widely used in the neuroimaging community to uncover multivariate associations between two data modalities (e.g., brain imaging and behaviour). However, these methods have inherent limitations: (1) statistical inferences about the associations are often not robust; (2) the associations within each data modality are not modelled; (3) missing values need to be imputed or removed. Group Factor Analysis (GFA) is a hierarchical model that addresses the first two limitations by providing Bayesian inference and modelling modality-specific associations. Here, we propose an extension of GFA that handles missing data, and highlight that GFA can be used as a predictive model. We applied GFA to synthetic and real data consisting of brain connectivity and non-imaging measures from the Human Connectome Project (HCP). In synthetic data, GFA uncovered the underlying shared and specific factors and predicted correctly the non-observed data modalities in complete and incomplete data sets. In the HCP data, we identified four relevant shared factors, capturing associations between mood, alcohol and drug use, cognition, demographics and psychopathological measures and the default mode, frontoparietal control, dorsal and ventral networks and insula, as well as two factors describing associations within brain connectivity. In addition, GFA predicted a set of non-imaging measures from brain connectivity. These findings were consistent in complete and incomplete data sets, and replicated previous findings in the literature. GFA is a promising tool that can be used to uncover associations between and within multiple data modalities in benchmark datasets (such as, HCP), and easily extended to more complex models to solve more challenging tasks.

Cardiac-induced cerebral pulsatility, brain structure, and cognition in middle and older-aged adults

Changes of cardiac-induced regional pulsatility can be associated with specific regions of brain volumetric changes, and these are related with cognitive alterations. Thus, mapping of cardiac pulsatility over the entire brain can be helpful to assess these relationships. A total of 108 subjects (age: 66.5 ± 8.4 years, 68 females, 52 healthy controls, 11 subjective cognitive decline, 17 impaired without complaints, 19 MCI and 9 AD) participated. The pulsatility map was obtained directly from resting-state functional MRI time-series data at 3T. Regional brain volumes were segmented from anatomical MRI. Multidomain neuropsychological battery was performed to test memory, language, attention and visuospatial construction. The Montreal Cognitive Assessment (MoCA) was also administered. The sparse partial least square (SPLS) method, which is desirable for better interpreting high-dimensional variables, was applied for the relationship between the entire brain voxels of pulsatility and 45 segmented brain volumes. A multiple holdout SPLS framework was used to optimize sparsity for assessing the pulsatility-volume relationship model and to test the reliability by fitting the models to 9 different splits of the data. We found statistically significant associations between subsets of pulsatility voxels and subsets of segmented brain volumes by rejecting the omnibus null hypothesis (any of 9 splits has p < 0.0056 (=0.05/9) with the Bonferroni correction). The pulsatility was positively associated with the lateral ventricle, choroid plexus, inferior lateral ventricle, and 3rd ventricle and negatively associated with hippocampus, ventral DC, and thalamus volumes for the first pulsatility-volume relationship. The pulsatility had an additional negative relationship with the amygdala and brain stem volumes for the second pulsatility-volume relationship. The spatial distribution of correlated pulsatility was observed in major feeding arteries to the brain regions, ventricles, and sagittal sinus. The indirect mediating pathways through the volumetric changes were statistically significant between the pulsatility and multiple cognitive measures (p < 0.01). Thus, the cerebral pulsatility, along with volumetric measurements, could be a potential marker for better understanding of pathophysiology and monitoring disease progression in age-related neurodegenerative disorders.

Conditional canonical correlation estimation based on covariates with random forests

MOTIVATION

Investigating the relationships between two sets of variables helps to understand their interactions and can be done with canonical correlation analysis (CCA). However, the correlation between the two sets can sometimes depend on a third set of covariates, often subject-related ones such as age, gender, or other clinical measures. In this case, applying CCA to the whole population is not optimal and methods to estimate conditional CCA, given the covariates, can be useful.

RESULTS

We propose a new method called Random Forest with Canonical Correlation Analysis (RFCCA) to estimate the conditional canonical correlations between two sets of variables given subject-related covariates. The individual trees in the forest are built with a splitting rule specifically designed to partition the data to maximize the canonical correlation heterogeneity between child nodes. We also propose a significance test to detect the global effect of the covariates on the relationship between two sets of variables. The performance of the proposed method and the global significance test is evaluated through simulation studies that show it provides accurate canonical correlation estimations and well-controlled Type-1 error. We also show an application of the proposed method with EEG data.

AVAILABILITY

RFCCA is implemented in a freely available R package on CRAN (https://CRAN.R-project.org/package=RFCCA).

SUPPLEMENTARY INFORMATION

Supplementary material are available at Bioinformatics online.

Objective Assessments in Diagnoses and Treatment: A Proposed Change in Paradigm

For patients with psychiatric disorders, current diagnostic and treatment approaches are far from optimal. The clinical interview drives the standard approach-matching symptoms to diagnostic criteria-and results in standardized pharmacological and behavioral treatments, often, with inadequate outcome; but now, recent imaging advances can correlate behavioral assessments with brain function and measure them against normative databases to provide data critical for the reevaluation of patient diagnosis and treatment. This article addresses the data that support a redefinition of our current paradigm. We believe a neurobehavioral approach provides for more personalized treatment approaches unbound from classically defined diagnostic biases.

Probing the clinical and brain structural boundaries of bipolar and major depressive disorder

Bipolar disorder (BD) and major depressive disorder (MDD) have both common and distinct clinical features, that pose both conceptual challenges in terms of their diagnostic boundaries and practical difficulties in optimizing treatment. Multivariate machine learning techniques offer new avenues for exploring these boundaries based on clinical neuroanatomical features. Brain structural data were obtained at 3 T from a sample of 90 patients with BD, 189 patients with MDD, and 162 healthy individuals. We applied sparse partial least squares discriminant analysis (s-PLS-DA) to identify clinical and brain structural features that may discriminate between the two clinical groups, and heterogeneity through discriminative analysis (HYDRA) to detect patient subgroups with reference to healthy individuals. Two clinical dimensions differentiated BD from MDD (area under the curve: 0.76, P < 0.001); one dimension emphasized disease severity as well as irritability, agitation, anxiety and flight of ideas and the other emphasized mostly elevated mood. Brain structural features could not distinguish between the two disorders. HYDRA classified patients in two clusters that differed in global and regional cortical thickness, the distribution proportion of BD and MDD and positive family history of psychiatric disorders. Clinical features remain the most reliable discriminant attributed of BD and MDD depression. The brain structural findings suggests that biological partitions of patients with mood disorders are likely to lead to the identification of subgroups, that transcend current diagnostic divisions into BD and MDD and are more likely to be aligned with underlying genetic variation. These results set the foundation for future studies to enhance our understanding of brain-behavior relationships in mood disorders.

How to Determine the Early Warning Threshold Value of Meteorological Factors on Influenza through Big Data Analysis and Machine Learning

Infectious diseases are a major health challenge for the worldwide population. Since their rapid spread can cause great distress to the real world, in addition to taking appropriate measures to curb the spread of infectious diseases in the event of an outbreak, proper prediction and early warning before the outbreak of the threat of infectious diseases can provide an important basis for early and reasonable response by the government health sector, reduce morbidity and mortality, and greatly reduce national losses. However, if only traditional medical data is involved, it may be too late or too difficult to implement prediction and early warning of an infectious outbreak. Recently, medical big data has become a research hotspot and has played an increasingly important role in public health, precision medicine, and disease prediction. In this paper, we focus on exploring a prediction and early warning method for influenza with the help of medical big data. It is well known that meteorological conditions have an influence on influenza outbreaks. So, we try to find a way to determine the early warning threshold value of influenza outbreaks through big data analysis concerning meteorological factors. Results show that, based on analysis of meteorological conditions combined with influenza outbreak history data, the early warning threshold of influenza outbreaks could be established with reasonable high accuracy.

Associations between aversive learning processes and transdiagnostic psychiatric symptoms in a general population sample

Symptom expression in psychiatric conditions is often linked to altered threat perception, however how computational mechanisms that support aversive learning relate to specific psychiatric symptoms remains undetermined. We answer this question using an online game-based aversive learning task together with measures of common psychiatric symptoms in 400 subjects. We show that physiological symptoms of anxiety and a transdiagnostic compulsivity-related factor are associated with enhanced safety learning, as measured using a probabilistic computational model, while trait cognitive anxiety symptoms are associated with enhanced learning from danger. We use data-driven partial least squares regression to identify two separable components across behavioural and questionnaire data: one linking enhanced safety learning and lower estimated uncertainty to physiological anxiety, compulsivity, and impulsivity; the other linking enhanced threat learning and heightened uncertainty estimation to symptoms of depression and social anxiety. Our findings implicate aversive learning processes in the expression of psychiatric symptoms that transcend diagnostic boundaries.