Evaluating the utility of daily speech assessments for monitoring depression symptoms.

Objective: Depression is a common mental health disorder and a major public health concern, significantly interfering with the lives of those affected. The complex clinical presentation of depression complicates symptom assessments. Day-to-day fluctuations of depression symptoms within an individual bring an additional barrier, since infrequent testing may not reveal symptom fluctuation. Digital measures such as speech can facilitate daily objective symptom evaluation. Here, we evaluated the effectiveness of daily speech assessment in characterizing speech fluctuations in the context of depression symptoms, which can be completed remotely, at a low cost and with relatively low administrative resources. Methods: Community volunteers (N = 16) completed a daily speech assessment, using the Winterlight Speech App, and Patient Health Questionnaire-9 (PHQ-9) for 30 consecutive business days. We calculated 230 acoustic and 290 linguistic features from individual's speech and investigated their relationship to depression symptoms at the intra-individual level through repeated measures analyses. Results: We observed that depression symptoms were linked to linguistic features, such as less frequent use of dominant and positive words. Greater depression symptomatology was also significantly correlated with acoustic features: reduced variability in speech intensity and increased jitter. Conclusions: Our findings support the feasibility of using acoustic and linguistic features as a measure of depression symptoms and propose daily speech assessment as a tool for better characterization of symptom fluctuations.

OViTAD: Optimized Vision Transformer to Predict Various Stages of Alzheimer's Disease Using Resting-State fMRI and Structural MRI Data.

Advances in applied machine learning techniques for neuroimaging have encouraged scientists to implement models to diagnose brain disorders such as Alzheimer's disease at early stages. Predicting the exact stage of Alzheimer's disease is challenging; however, complex deep learning techniques can precisely manage this. While successful, these complex architectures are difficult to interrogate and computationally expensive. Therefore, using novel, simpler architectures with more efficient pattern extraction capabilities, such as transformers, is of interest to neuroscientists. This study introduced an optimized vision transformer architecture to predict the group membership by separating healthy adults, mild cognitive impairment, and Alzheimer's brains within the same age group (>75 years) using resting-state functional (rs-fMRI) and structural magnetic resonance imaging (sMRI) data aggressively preprocessed by our pipeline. Our optimized architecture, known as OViTAD is currently the sole vision transformer-based end-to-end pipeline and outperformed the existing transformer models and most state-of-the-art solutions. Our model achieved F1-scores of 97%±0.0 and 99.55%±0.39 from the testing sets for the rs-fMRI and sMRI modalities in the triple-class prediction experiments. Furthermore, our model reached these performances using 30% fewer parameters than a vanilla transformer. Furthermore, the model was robust and repeatable, producing similar estimates across three runs with random data splits (we reported the averaged evaluation metrics). Finally, to challenge the model, we observed how it handled increasing noise levels by inserting varying numbers of healthy brains into the two dementia groups. Our findings suggest that optimized vision transformers are a promising and exciting new approach for neuroimaging applications, especially for Alzheimer's disease prediction.

Screening for Generalized Anxiety Disorder From Acoustic and Linguistic Features of Impromptu Speech: Prediction Model Evaluation Study.

BACKGROUND: Frequent interaction with mental health professionals is required to screen, diagnose, and track mental health disorders. However, high costs and insufficient access can make frequent interactions difficult. The ability to assess a mental health disorder passively and at frequent intervals could be a useful complement to the conventional treatment. It may be possible to passively assess clinical symptoms with high frequency by characterizing speech alterations collected using personal smartphones or other wearable devices. The association between speech features and mental health disorders can be leveraged as an objective screening tool. OBJECTIVE: This study aimed to evaluate the performance of a model that predicts the presence of generalized anxiety disorder (GAD) from acoustic and linguistic features of impromptu speech on a larger and more generalizable scale than prior studies did. METHODS: A total of 2000 participants were recruited, and they participated in a single web-based session. They completed the Generalized Anxiety Disorder-7 item scale assessment and provided an impromptu speech sample in response to a modified version of the Trier Social Stress Test. We used the linguistic and acoustic features that were found to be associated with anxiety disorders in previous studies along with demographic information to predict whether participants fell above or below the screening threshold for GAD based on the Generalized Anxiety Disorder-7 item scale threshold of 10. Separate models for each sex were also evaluated. We reported the mean area under the receiver operating characteristic (AUROC) from a repeated 5-fold cross-validation to evaluate the performance of the models. RESULTS: A logistic regression model using only acoustic and linguistic speech features achieved a significantly greater prediction accuracy than a random model did (mean AUROC 0.57, SD 0.03; P<.001). When separately assessing samples from female participants, we observed a mean AUROC of 0.55 (SD 0.05; P=.01). The model constructed from the samples from male participants achieved a mean AUROC of 0.57 (SD 0.07; P=.002). The mean AUROC increased to 0.62 (SD 0.03; P<.001) on the all-sample data set when demographic information (age, sex, and income) was included, indicating the importance of demographics when screening for anxiety disorders. The performance also increased for the female sample to a mean of 0.62 (SD 0.04; P<.001) when using demographic information (age and income). An increase in performance was not observed when demographic information was added to the model constructed from the male samples. CONCLUSIONS: A logistic regression model using acoustic and linguistic speech features, which have been suggested to be associated with anxiety disorders in prior studies, can achieve above-random accuracy for predicting GAD. Importantly, the addition of basic demographic variables further improves model performance, suggesting a role for speech and demographic information to be used as automated, objective screeners of GAD.

Who does what to whom? graph representations of action-predication in speech relate to psychopathological dimensions of psychosis.

Graphical representations of speech generate powerful computational measures related to psychosis. Previous studies have mostly relied on structural relations between words as the basis of graph formation, i.e., connecting each word to the next in a sequence of words. Here, we introduced a method of graph formation grounded in semantic relationships by identifying elements that act upon each other (action relation) and the contents of those actions (predication relation). Speech from picture descriptions and open-ended narrative tasks were collected from a cross-diagnostic group of healthy volunteers and people with psychotic or non-psychotic disorders. Recordings were transcribed and underwent automated language processing, including semantic role labeling to identify action and predication relations. Structural and semantic graph features were computed using static and dynamic (moving-window) techniques. Compared to structural graphs, semantic graphs were more strongly correlated with dimensional psychosis symptoms. Dynamic features also outperformed static features, and samples from picture descriptions yielded larger effect sizes than narrative responses for psychosis diagnoses and symptom dimensions. Overall, semantic graphs captured unique and clinically meaningful information about psychosis and related symptom dimensions. These features, particularly when derived from semi-structured tasks using dynamic measurement, are meaningful additions to the repertoire of computational linguistic methods in psychiatry.

Acoustic and Linguistic Features of Impromptu Speech and Their Association With Anxiety: Validation Study.

BACKGROUND: The measurement and monitoring of generalized anxiety disorder requires frequent interaction with psychiatrists or psychologists. Access to mental health professionals is often difficult because of high costs or insufficient availability. The ability to assess generalized anxiety disorder passively and at frequent intervals could be a useful complement to conventional treatment and help with relapse monitoring. Prior work suggests that higher anxiety levels are associated with features of human speech. As such, monitoring speech using personal smartphones or other wearable devices may be a means to achieve passive anxiety monitoring. OBJECTIVE: This study aims to validate the association of previously suggested acoustic and linguistic features of speech with anxiety severity. METHODS: A large number of participants (n=2000) were recruited and participated in a single web-based study session. Participants completed the Generalized Anxiety Disorder 7-item scale assessment and provided an impromptu speech sample in response to a modified version of the Trier Social Stress Test. Acoustic and linguistic speech features were a priori selected based on the existing speech and anxiety literature, along with related features. Associations between speech features and anxiety levels were assessed using age and personal income as covariates. RESULTS: Word count and speaking duration were negatively correlated with anxiety scores (r=-0.12; P<.001), indicating that participants with higher anxiety scores spoke less. Several acoustic features were also significantly (P<.05) associated with anxiety, including the mel-frequency cepstral coefficients, linear prediction cepstral coefficients, shimmer, fundamental frequency, and first formant. In contrast to previous literature, second and third formant, jitter, and zero crossing rate for the z score of the power spectral density acoustic features were not significantly associated with anxiety. Linguistic features, including negative-emotion words, were also associated with anxiety (r=0.10; P<.001). In addition, some linguistic relationships were sex dependent. For example, the count of words related to power was positively associated with anxiety in women (r=0.07; P=.03), whereas it was negatively associated with anxiety in men (r=-0.09; P=.01). CONCLUSIONS: Both acoustic and linguistic speech measures are associated with anxiety scores. The amount of speech, acoustic quality of speech, and gender-specific linguistic characteristics of speech may be useful as part of a system to screen for anxiety, detect relapse, or monitor treatment.

Three Dimensions of Association Link Migraine Symptoms and Functional Connectivity.

Migraine is a heterogeneous disorder with variable symptoms and responsiveness to therapy. Because of previous analytic shortcomings, variance in migraine symptoms has been inconsistently related to brain function. In the current analysis, we used data from two sites (n = 143, male and female humans), and performed canonical correlation analysis, relating resting-state functional connectivity (RSFC) with a broad range of migraine symptoms, ranging from headache characteristics to sleep abnormalities. This identified three dimensions of covariance between symptoms and RSFC. The first dimension related to headache intensity, headache frequency, pain catastrophizing, affect, sleep disturbances, and somatic abnormalities, and was associated with frontoparietal and dorsal attention network connectivity, both of which are major cognitive networks. Additionally, RSFC scores from this dimension, both the baseline value and the change from baseline to postintervention, were associated with responsiveness to mind-body therapy. The second dimension was related to an inverse association between pain and anxiety, and to default mode network connectivity. The final dimension was related to pain catastrophizing, and salience, sensorimotor, and default mode network connectivity. In addition to performing canonical correlation analysis, we evaluated the current clustering of migraine patients into episodic and chronic subtypes, and found no evidence to support this clustering. However, when using RSFC scores from the three significant dimensions, we identified a novel clustering of migraine patients into four biotypes with unique functional connectivity patterns. These findings provide new insight into individual variability in migraine, and could serve as the foundation for novel therapies that take advantage of migraine heterogeneity.SIGNIFICANCE STATEMENT Using a large multisite dataset of migraine patients, we identified three dimensions of multivariate association between symptoms and functional connectivity. This analysis revealed neural networks that relate to all measured symptoms, but also to specific symptom ensembles, such as patient propensity to catastrophize painful events. Using these three dimensions, we found four biotypes of migraine informed by clinical and neural variation together. Such findings pave the way for precision medicine therapy for migraine.

Natural Language Processing as an Emerging Tool to Detect Late-Life Depression.

Late-life depression (LLD) is a major public health concern. Despite the availability of effective treatments for depression, barriers to screening and diagnosis still exist. The use of current standardized depression assessments can lead to underdiagnosis or misdiagnosis due to subjective symptom reporting and the distinct cognitive, psychomotor, and somatic features of LLD. To overcome these limitations, there has been a growing interest in the development of objective measures of depression using artificial intelligence (AI) technologies such as natural language processing (NLP). NLP approaches focus on the analysis of acoustic and linguistic aspects of human language derived from text and speech and can be integrated with machine learning approaches to classify depression and its severity. In this review, we will provide rationale for the use of NLP methods to study depression using speech, summarize previous research using NLP in LLD, compare findings to younger adults with depression and older adults with other clinical conditions, and discuss future directions including the use of complementary AI strategies to fully capture the spectrum of LLD.

Brain and Physiological Markers of Autonomic Function Are Associated With Treatment-Related Improvements in Self-Reported Autonomic Dysfunction in Veterans With Gulf War Illness: An Exploratory Pilot Study.

BACKGROUND: Gulf War Illness (GWI) is a poorly understood condition characterized by a constellation of mood, cognitive, and physical symptoms. A growing body of evidence demonstrates autonomic nervous system (ANS) dysfunction. Few published treatment studies exist for GWI. METHOD: We recently completed a randomized controlled trial comparing a 10-week group yoga intervention to 10-week group cognitive behavioral therapy (CBT) for veterans with GWI. Here, we present exploratory data on ANS biomarkers of treatment response from a small pilot exploratory neurophysiological add-on study (n = 13) within that larger study. RESULTS: Findings suggest that veterans with GWI receiving either yoga or CBT for pain improved following treatment and that changes in biological ANS-especially for the yoga group-moved in the direction of healthy profiles: lower heart rate, higher square root of the mean squared differences between successive R-R intervals (RMSSD), greater parasympathetic activation/dominance (increased high-frequency heart rate variability [HF-HRV], decreased low-frequency/high-frequency [LF/HF] ratio), reduced right amygdala volume, and stronger amygdala-default mode/amygdala-salience network connectivity, both immediately posttreatment and at 6-month follow-up. Biological mechanisms of CBT appeared to underlie improvements in more psychologically loaded symptoms such as self-reported fatigue and energy. Higher tonic arousal and/or more sympathetic dominance (higher skin conductance, lower RMSSD, lower HF-HRV, higher LF/HF ratio) pretreatment predicted greater treatment-related improvements in self-reported ANS for both the yoga and CBT group. CONCLUSION: These exploratory pilot data provide preliminary support for the suggestion that treatment (yoga, CBT) is associated with improvements in both biological and self-reported ANS dysfunctions in GWI. The major limitation for these findings is the small sample size. Larger and more controlled studies are needed to replicate these findings and directly compare biomarkers of yoga versus CBT.

Association between Anterior Cingulate Neurochemical Concentration and Individual Differences in Hypnotizability.

Hypnosis is the oldest form of Western psychotherapy and a powerful evidence-based treatment for numerous disorders. Hypnotizability is variable between individuals; however, it is a stable trait throughout adulthood, suggesting that neurophysiological factors may underlie hypnotic responsiveness. One brain region of particular interest in functional neuroimaging studies of hypnotizability is the anterior cingulate cortex (ACC). Here, we examined the relationships between the neurochemicals, GABA, and glutamate, in the ACC and hypnotizability in healthy individuals. Participants underwent a magnetic resonance imaging (MRI) session, whereby T1-weighted anatomical and MEGA-PRESS spectroscopy scans were acquired. Voxel placement over the ACC was guided by a quantitative meta-analysis of functional neuroimaging studies of hypnosis. Hypnotizability was assessed using the Hypnotic Induction Profile (HIP), and self-report questionnaires to assess absorption (TAS), dissociation (DES), and negative affect were completed. ACC GABA concentration was positively associated with HIP scores such that the higher the GABA concentration, the more hypnotizable an individual. An exploratory analysis of questionnaire subscales revealed a negative relationship between glutamate and the absorption and imaginative involvement subscale of the DES. These results provide a putative neurobiological basis for individual differences in hypnotizability and can inform our understanding of treatment response to this growing psychotherapeutic tool.

Altered structural brain network topology in chronic migraine.

Despite its prevalence and high disease burden, the pathophysiological mechanisms underlying chronic migraine (CM) are not well understood. As CM is a complex disorder associated with a range of sensory, cognitive, and affective comorbidities, examining structural network disruption may provide additional insights into CM symptomology beyond studies of focal brain regions. Here, we compared structural interconnections in patients with CM (n = 52) and healthy controls (HC) (n = 48) using MRI measures of cortical thickness and subcortical volume combined with graph theoretical network analyses. The analysis focused on both local (nodal) and global measures of topology to examine network integration, efficiency, centrality, and segregation. Our results indicated that patients with CM had altered global network properties that were characterized as less integrated and efficient (lower global and local efficiency) and more highly segregated (higher transitivity). Patients also demonstrated aberrant local network topology that was less integrated (higher path length), less central (lower closeness centrality), less efficient (lower local efficiency) and less segregated (lower clustering). These network differences not only were most prominent in the limbic and insular cortices but also occurred in frontal, temporal, and brainstem regions, and occurred in the absence of group differences in focal brain regions. Taken together, examining structural correlations between brain areas may be a more sensitive means to detect altered brain structure and understand CM symptomology at the network level. These findings contribute to an increased understanding of structural connectivity in CM and provide a novel approach to potentially track and predict the progression of migraine disorders.This study is registered on ClinicalTrials.gov (Identifier: NCT03304886).

Clinical Features Contributing to Cortical Thickness Changes in Chronic Migraine - A Pilot Study.

OBJECTIVES: The objectives of this cross-sectional pilot study were threefold: to identify regions of cortical thickness that differentiate chronic migraine (CM) from controls, to assess group differences in interregional cortical thickness covariance, and to determine group differences in associations between clinical variables and cortical thickness. BACKGROUND: Cortical thickness alterations in relation to clinical features have not been adequately explored in CM. Assessment of this relationship can be useful to describe cortical substrates for disease progression in migraine and to identify clinical variables that warrant management emphasis. METHODS: Thirty CM cases (mean age 40 years; male-to-female 1:4) and 30 sex-matched healthy controls (mean age 40 years) were enrolled. Participants completed self-administered and standardized questionnaires assessing headache-related clinical features and common psychological comorbidities. T1-weighted brain images were acquired on a 3T MRI. A whole-brain cortical thickness analysis was performed. Additionally, correlations between all brain regions were assessed to examine interregional cortical thickness covariance. Interactions were analyzed to identify clinical variables that were significantly associated with cortical thickness. RESULTS: The whole brain cortical thickness analysis revealed no significant differences between CM patients and controls. However, significant associations between clinical features and cortical thickness were observed for the patients only. These associations included the right superior temporal sulcus (R2  = 0.72, P = .001) and the right insula (R2  = 0.71, P = .002) with distinct clinical variables ie, longer history of CM, posttraumatic stress disorder (PTSD), sleep quality, pain self-efficacy, and somatic symptoms. Higher interregional cortical covariance was found in CM compared to controls (OR = 3.1, CI 2.10-4.56, P < .0001), such that cortical thickness between regions tended to be more correlated in patients, particularly in the temporal and frontal lobes. CONCLUSION: CM patients have significantly greater cortical covariance compared to controls. Cortical thickness in CM patients was predominantly accounted for by CM duration, PTSD, and poor sleep quality, while improved pain self-efficacy buffered cortical thickness. While it is important to address all CM features and comorbidities, it may be useful to emphasize optimizing the management of certain clinical features that contribute to cortical abnormalities including managing PTSD, early management to shorten duration of CM, and improving pain self-efficacy and sleep quality.

MCADNNet: Recognizing Stages of Cognitive Impairment through Efficient Convolutional fMRI and MRI Neural Network Topology Models.

Mild cognitive impairment (MCI) represents the intermediate stage between normal cerebral aging and dementia associated with Alzheimer's disease (AD). Early diagnosis of MCI and AD through artificial intelligence has captured considerable scholarly interest; researchers hope to develop therapies capable of slowing or halting these processes. We developed a state-of-the-art deep learning algorithm based on an optimized convolutional neural network (CNN) topology called MCADNNet that simultaneously recognizes MCI, AD, and normally aging brains in adults over the age of 75 years, using structural and functional magnetic resonance imaging (fMRI) data. Following highly detailed preprocessing, four-dimensional (4D) fMRI and 3D MRI were decomposed to create 2D images using a lossless transformation, which enables maximum preservation of data details. The samples were shuffled and subject-level training and testing datasets were completely independent. The optimized MCADNNet was trained and extracted invariant and hierarchical features through convolutional layers followed by multi-classification in the last layer using a softmax layer. A decision-making algorithm was also designed to stabilize the outcome of the trained models. To measure the performance of classification, the accuracy rates for various pipelines were calculated before and after applying the decision-making algorithm. Accuracy rates of 99.77% 0.36% and 97.5% 1.16% were achieved for MRI and fMRI pipelines, respectively, after applying the decision-making algorithm. In conclusion, a cutting-edge and optimized topology called MCADNNet was designed and preceded a preprocessing pipeline; this was followed by a decision-making step that yielded the highest performance achieved for simultaneous classification of the three cohorts examined.

Comparison of Diffusion-Weighted MRI Reconstruction Methods for Visualization of Cranial Nerves in Posterior Fossa Surgery.

Diffusion-weighted imaging (DWI)-based tractography has gained increasing popularity as a method for detailed visualization of white matter (WM) tracts. Different imaging techniques, and more novel, advanced imaging methods provide significant WM structural detail. While there has been greater focus on improving tract visualization for larger WM pathways, the relative value of each method for cranial nerve reconstruction and how this methodology can assist surgical decision-making is still understudied. Images from 10 patients with posterior fossa tumors (4 male, mean age: 63.5), affecting either the trigeminal nerve (CN V) or the facial/vestibular complex (CN VII/VIII), were employed. Three distinct reconstruction methods [two tensor-based methods: single diffusion tensor tractography (SDT) (3D Slicer), eXtended streamline tractography (XST), and one fiber orientation distribution (FOD)-based method: streamline tractography using constrained spherical deconvolution (CSD)-derived estimates (MRtrix3)], were compared to determine which of these was best suited for use in a neurosurgical setting in terms of processing speed, anatomical accuracy, and accurate depiction of the relationship between the tumor and affected CN. Computation of the tensor map was faster when compared to the implementation of CSD to provide estimates of FOD. Both XST and CSD-based reconstruction methods tended to give more detailed representations of the projections of CN V and CN VII/VIII compared to SDT. These reconstruction methods were able to more accurately delineate the course of CN V and CN VII/VIII, differentiate CN V from the cerebellar peduncle, and delineate compression of CN VII/VIII in situations where SDT could not. However, CSD-based reconstruction methods tended to generate more invalid streamlines. XST offers the best combination of anatomical accuracy and speed of reconstruction of cranial nerves within this patient population. Given the possible anatomical limitations of single tensor models, supplementation with more advanced tensor-based reconstruction methods might be beneficial.

Structural Magnetic Resonance Imaging Can Identify Trigeminal System Abnormalities in Classical Trigeminal Neuralgia.

Classical trigeminal neuralgia (TN) is a chronic pain disorder that has been described as one of the most severe pains one can suffer. The most prevalent theory of TN etiology is that the trigeminal nerve is compressed at the root entry zone (REZ) by blood vessels. However, there is significant evidence showing a lack of neurovascular compression (NVC) for many cases of classical TN. Furthermore, a considerable number of patients who are asymptomatic have MR evidence of NVC. Since there is no validated animal model that reproduces the clinical features of TN, our understanding of TN pathology mainly comes from biopsy studies that have limitations. Sophisticated structural MRI techniques including diffusion tensor imaging provide new opportunities to assess the trigeminal nerves and CNS to provide insight into TN etiology and pathogenesis. Specifically, studies have used high-resolution structural MRI methods to visualize patterns of trigeminal nerve-vessel relationships and to detect subtle pathological features at the trigeminal REZ. Structural MRI has also identified CNS abnormalities in cortical and subcortical gray matter and white matter and demonstrated that effective neurosurgical treatment for TN is associated with a reversal of specific nerve and brain abnormalities. In conclusion, this review highlights the advanced structural neuroimaging methods that are valuable tools to assess the trigeminal system in TN and may inform our current understanding of TN pathology. These methods may in the future have clinical utility for the development of neuroimaging-based biomarkers of TN.

Journal Club: Exacerbation of headache during dihydroergotamine for chronic migraine does not alter outcome.

Transient headache exacerbation during IV dihydroergotamine (DHE) therapy of migraine may prompt clinicians to prematurely discontinue DHE therapy, potentially depriving patients of the full benefit of DHE infusion. In a recent Neurology® article, Eller et al. evaluated whether or not worsening headache during DHE infusion was associated with suboptimal medium-term headache outcomes.

Diffusivity signatures characterize trigeminal neuralgia associated with multiple sclerosis.

BACKGROUND: Trigeminal neuralgia secondary to multiple sclerosis (MS-TN) is a facial neuropathic pain syndrome similar to classic trigeminal neuralgia (TN). While TN is caused by neurovascular compression of the fifth cranial nerve (CN V), how MS-related demyelination correlates with pain in MS-TN is not understood. OBJECTIVES: We aim to examine diffusivities along CN V in MS-TN, TN, and controls in order to reveal differential neuroimaging correlates across groups. METHODS: 3T MR diffusion weighted, T1, T2 and FLAIR sequences were acquired for MS-TN, TN, and controls. Multi-tensor tractography was used to delineate CN V across cisternal, root entry zone (REZ), pontine and peri-lesional segments. Diffusion metrics including fractional anisotropy (FA), and radial (RD), axial (AD), and mean diffusivities (MD) were measured from each segment. RESULTS: CN V segments showed distinctive diffusivity patterns. The TN group showed higher FA in the cisternal segment ipsilateral to the side of pain, and lower FA in the ipsilateral REZ segment. The MS-TN group showed lower FA in the ipsilateral peri-lesional segments, suggesting differential microstructural changes along CN V in these conditions. CONCLUSIONS: The study demonstrates objective differences in CN V microstrucuture in TN and MS-TN using non-invasive neuroimaging. This represents a significant improvement in the methods currently available to study pain in MS.