Search

In human neuroscience, machine learning models can be used to discover lower-dimensional neural representations relevant to behavior. However, these models often require large datasets and can be overfit with the small sample sizes typical in neuroimaging. To address this, we developed the Task-Relevant Autoencoder via Classifier Enhancement (TRACE) to extract behaviorally relevant representations. When tested against standard autoencoders and principal component analysis, TRACE showed up to 12% increased classification accuracy and 56% improvement in discovering task-relevant representations using fMRI data from ventral temporal cortex (VTC) of 59 subjects, highlighting its potential for behavioral data.Machine learning models applications also extend to predictive modeling and pattern discovery in modern biology. However, these models often fail to generalize across different datasets due to statistical differences. This issue also exists in neuroscience, where data are collected across various laboratories using different experimental setups. Domain adaptation can align statistical distributions across datasets, enabling model transfer and mitigating overfitting issues. In the second chapter we discussed domain adaptation in the context of small-scale, heterogeneous biological data, outlining its benefits, challenges, and key methodologies. We advocate for integrating domain adaptation techniques into computational biology, with further customized developments.Building on these insights, we used DA for understanding brain region interactions during visual processing. We examine the ventral temporal cortex (VTC) and prefrontal cortex (PFC) using Domain Adaptive Task-Relevant Autoencoding via Classifier Enhancement (DATRACE) to explore shared neural representations. DATRACE leverages domain adaptation techniques within an encoder-decoder architecture to predict voxel activities from a shared latent space, in order to ensure relevance for object recognition tasks. P

Both model-driven and data-driven approaches can be used to analyse fMRI data on group level but they have their respective advantages and disadvantages. A model-driven approach such as the GLM requires prior knowledge of the BOLD contrast while offering a simple way of performing inference. In contrast, ICA is a data-driven approach that instead takes full advantage of the data but in which it is harder to make meaningful inference. A method for combining GLM and ICA on group level utilising their respective strengths is therefore tested. The performance of the method highly relies on ICA’s ability to estimate accurate sources. Violation of ICA’s assumptions can potentially affect this ability and this aspect is investigated to get a broader understanding. The method is tested on both simulated and real fMRI data consisting of subjects with schizophrenia and healthy controls. The simulation shows high power but a higher false positive rate than expected. The method is able to find the brain regions that are typically active during a spatial working memory task, such as the posterior parietal cortex and the dorsolateral prefrontal cortex. It is further found that ICA performs especially poorly when the sources are uniformly distributed on the unit sphere.

Resting state functional magnetic resonance imaging (rsfMRI) provides researchers and clinicians with a powerful tool to examine functional connectivity across large-scale brain networks, with ever-increasing applications to the study of neurological disorders, such as traumatic brain injury (TBI). While rsfMRI holds unparalleled promise in systems neurosciences, its acquisition and analytical methodology across research groups is variable, resulting in a literature that is challenging to integrate and interpret. The focus of this narrative review is to address the primary methodological issues including investigator decision points in the application of rsfMRI to study the consequences of TBI. As part of the ENIGMA Brain Injury working group, we have collaborated to identify a minimum set of recommendations that are designed to produce results that are reliable, harmonizable, and reproducible for the TBI imaging research community. Part one of this review provides the results of a literature search of current rsfMRI studies of TBI, highlighting key design considerations and data processing pipelines. Part two outlines seven data acquisition, processing, and analysis recommendations with the goal of maximizing study reliability and between-site comparability, while preserving investigator autonomy. Part three summarizes new directions and opportunities for future rsfMRI studies in TBI patients. The goal is to galvanize the TBI community to gain consensus for a set of rigorous and reproducible methods, and to increase analytical transparency and data sharing to address the reproducibility crisis in the field.

Both model-driven and data-driven approaches can be used to analyse fMRI data on group level but they have their respective advantages and disadvantages. A model-driven approach such as the GLM requires prior knowledge of the BOLD contrast while offering a simple way of performing inference. In contrast, ICA is a data-driven approach that instead takes full advantage of the data but in which it is harder to make meaningful inference. A method for combining GLM and ICA on group level utilising their respective strengths is therefore tested. The performance of the method highly relies on ICA’s ability to estimate accurate sources. Violation of ICA’s assumptions can potentially affect this ability and this aspect is investigated to get a broader understanding. The method is tested on both simulated and real fMRI data consisting of subjects with schizophrenia and healthy controls. The simulation shows high power but a higher false positive rate than expected. The method is able to find the brain regions that are typically active during a spatial working memory task, such as the posterior parietal cortex and the dorsolateral prefrontal cortex. It is further found that ICA performs especially poorly when the sources are uniformly distributed on the unit sphere.

Light has many non-image-forming functions including modulation of pupil size and stimulation of alertness and cognition. Part of these non-image-forming effects may be mediated by the brainstem locus coeruleus. The processing of sensory inputs can be associated with a transient pupil dilation that is likely driven in part by the phasic activity of the locus coeruleus. In the present study, we aimed to characterise the task-evoked pupil response associated with auditory inputs under different light levels and across two cognitive tasks. We continuously monitored the pupil of 20 young healthy participants (mean [SD] 24.05?[4.0]?years; 14 women) whilst they completed an attentional and an emotional auditory task whilst exposed to repeated 30-40-s blocks of light interleaved with darkness periods. Blocks could either consist of monochromatic orange light (0.16?melanopic equivalent daylight illuminance (EDI) lux) or blue-enriched white light of three different levels [37, 92, 190?melanopic EDI lux; 6500?K]. For the analysis, 15 and then 14 participants were included in the attentional and emotional tasks, respectively. Generalised linear mixed models showed a significant main effect of light level on the task-evoked pupil responses triggered by the attentional and emotional tasks (p?=?0.0001). The impact of light was different for the target versus non-target stimulus of the attentional task but was not different for the emotional and neutral stimulus of the emotional task. There is a smaller sustained pupil size during brighter light blocks but, a higher light level triggers a stronger task-evoked pupil response to auditory stimulation, presumably through the recruitment of the locus coeruleus.

Introduction: Approximately one in six people will experience an episode of major depressive disorder (MDD) in their lifetime. Effective treatment is hindered by subjective clinical decision-making and a lack of objective prognostic biomarkers. Functional MRI (fMRI) could provide such an objective measure but the majority of MDD studies has focused on static approaches, disregarding the rapidly changing nature of the brain. In this study, we aim to predict depression severity changes at 3 and 6 months using dynamic fMRI features. Methods: For our research, we acquired a longitudinal dataset of 32 MDD patients with fMRI scans acquired at baseline and clinical follow-ups 3 and 6 months later. Several measures were derived from an emotion face-matching fMRI dataset: activity in brain regions, static and dynamic functional connectivity between functional brain networks (FBNs) and two measures from a wavelet coherence analysis approach. All fMRI features were evaluated independently, with and without demographic and clinical parameters. Patients were divided into two classes based on changes in depression severity at both follow-ups. Results: The number of coherence clusters (nCC) between FBNs, reflecting the total number of interactions (either synchronous, anti-synchronous or causal), resulted in the highest predictive performance. The nCC-based classifier achieved 87.5% and 77.4% accuracy for the 3- and 6-months change in severity, respectively. Furthermore, regression analyses supported the potential of nCC for predicting depression severity on a continuous scale. The posterior default mode network (DMN), dorsal attention network (DAN) and two visual networks were the most important networks in the optimal nCC models. Reduced nCC was associated with a poorer depression course, suggesting deficits in sustained attention to and coping with emotion-related faces. An ensemble of classifiers with demographic, clinical and lead coherence features, a measu

Goal: Structural brain graphs are conventionally limited to defining nodes as gray matter regions from an atlas,with edges reflecting the density of axonal projections between pairs of nodes. Here we explicitly model the entire set of voxels within a brain mask as nodes of high-resolution, subject-specific graphs. Methods: We define the strength of local voxel-to-voxel connections using diffusion tensors and orientation distribution functions derived from diffusion MRI data. We study the graphs’ Laplacian spectral properties on data from the Human Connectome Project. We then assess the extent of inter-subject variability of the Laplacian eigenmodes via a procrustes validation scheme. Finally, we demonstrate the extent to which functional MRI data are shaped by the underlying anatomical structure via graph signal processing. Results: The graph Laplacian eigenmodes manifest highly resolved spatial profiles, reflecting distributed patterns that correspond to major white matter pathways. We show that the intrinsic dimensionality of the eigenspace of such high-resolution graphs is only a mere fraction of the graph dimensions. By projecting task and resting-state data on low frequency graph Laplacian eigenmodes, we show that brain activity can be well approximated by a small subset of low frequency components. Conclusions: The proposed graphs open new avenues in studying the brain, be it, by exploring their organisational properties via graph or spectral graph theory, or by treating them as the scaffold on which brain function is observed at the individual level.

Goal: Structural brain graphs are conventionally limited to defining nodes as gray matter regions from an atlas,with edges reflecting the density of axonal projections between pairs of nodes. Here we explicitly model the entire set of voxels within a brain mask as nodes of high-resolution, subject-specific graphs. Methods: We define the strength of local voxel-to-voxel connections using diffusion tensors and orientation distribution functions derived from diffusion MRI data. We study the graphs’ Laplacian spectral properties on data from the Human Connectome Project. We then assess the extent of inter-subject variability of the Laplacian eigenmodes via a procrustes validation scheme. Finally, we demonstrate the extent to which functional MRI data are shaped by the underlying anatomical structure via graph signal processing. Results: The graph Laplacian eigenmodes manifest highly resolved spatial profiles, reflecting distributed patterns that correspond to major white matter pathways. We show that the intrinsic dimensionality of the eigenspace of such high-resolution graphs is only a mere fraction of the graph dimensions. By projecting task and resting-state data on low frequency graph Laplacian eigenmodes, we show that brain activity can be well approximated by a small subset of low frequency components. Conclusions: The proposed graphs open new avenues in studying the brain, be it, by exploring their organisational properties via graph or spectral graph theory, or by treating them as the scaffold on which brain function is observed at the individual level.

Advances in technology have led to increasing use of neuroimaging methods, such as electroencephalography (EEG), magnetoencephalography (MEG) and functional MRI (fMRI), that allow researchers to study brain function in a non-invasive way. However, functional neuroimaging studies are frequently contaminated by noise, or artefact, present in the data, and therefore, methods to remove or minimise artefact are crucial for obtaining accurate, reproducible results. This thesis examines artefact reduction in two different functional neuroimaging modalities - fMRI, and simultaneous EEG-fMRI. Firstly, a systematic review of artefact reduction for EEG-fMRI is presented, which contains two sub-reviews: a review of artefact reduction methods available, and a review of artefact reduction methods used in contemporary studies. The first review successfully distils all published artefact reduction methods from a twenty-year period into clear recommendations for researchers using this imaging modality. The second review found that from EEG-fMRI papers published over a four-year period, most users were selecting one or two similar methods, with up to 15% of users not adequately describing their methods used. A key finding from the work was that hardware-based methods of recording artefact are preferable to data-driven approaches, yet data-driven approaches are most commonly used. The second part of this thesis looks at motion artefact in fMRI studies - specifically, the utility of novel hardware - carbon wire loops (CWL) - for detecting head motion. We aimed to determine whether CWL could detect sub-volume motion onset and if so, whether CWL would suggest further data censoring when using a commonly used data-driven method, Framewise Displacement (FD). We hypothesised that the volume prior to motion onset in FD may often be motion affected, due to motion occurring part-way through volume acquisition, and that this motion would be detected by CWL. The results showed that CWL successfu

Altered postural control in the trunk/hip musculature is a characteristic of multiple neurological and musculoskeletal conditions. Previously it was not possible to determine if altered cortical and subcortical sensorimotor brain activation underlies impairments in postural control. This study used a novel fMRI-compatible paradigm to identify the brain activation associated with postural control in the trunk and hip musculature. BOLD fMRI imaging was conducted as participants performed two versions of a lower limb task involving lifting the left leg to touch the foot to a target. For the supported leg raise (SLR) the leg is raised from the knee while the thigh remains supported. For the unsupported leg raise (ULR) the leg is raised from the hip, requiring postural muscle activation in the abdominal/hip extensor musculature. Significant brain activation during the SLR task occurred predominantly in the right primary and secondary sensorimotor cortical regions. Brain activation during the ULR task occurred bilaterally in the primary and secondary sensorimotor cortical regions, as well as cerebellum and putamen. In comparison with the SLR, the ULR was associated with significantly greater activation in the right premotor/SMA, left primary motor and cingulate cortices, primary somatosensory cortex, supramarginal gyrus/parietal operculum, superior parietal lobule, cerebellar vermis, and cerebellar hemispheres. Cortical and subcortical regions activated during the ULR, but not during the SLR, were consistent with the planning, and execution of a task involving multisegmental, bilateral postural control. Future studies using this paradigm will determine mechanisms underlying impaired postural control in patients with neurological and musculoskeletal dysfunction.

Magnetic resonance imaging (MRI) is one of the pillars of modern medical imaging, providing a non-invasive means to generate 3D images of the body with high soft-tissue contrast. Furthermore, the possibilities afforded by the design of MRI sequences enable the signal to be sensitized to a multitude of physiological tissue properties, resulting in a wide variety of distinct MRI modalities for clinical and research use. This thesis presents a number of advanced brain MRI applications, which fulfill, to differing extents, two complementary aims. On the one hand, they explore the benefits of a multimodal approach to MRI, combining structural, functional and diffusion MRI, in a variety of contexts. On the other, they emphasize the use of advanced mathematical and computational tools in the analysis of MRI data, such as deep learning, Bayesian statistics, and graph signal processing. Paper I introduces an anatomically-adapted extension to previous work in Bayesian spatial priors for functional MRI data, where anatomical information is introduced from a T1-weighted image to compensate for the low anatomical contrast of functional MRI data. It has been observed that the spatial correlation structure of the BOLD signal in brain white matter follows the orientation of the underlying axonal fibers. Paper II argues about the implications of this fact on the ideal shape of spatial filters for the analysis of white matter functional MRI data. By using axonal orientation information extracted from diffusion MRI, and leveraging the possibilities afforded by graph signal processing, a graph-based description of the white matter structure is introduced, which, in turn, enables the definition of spatial filters whose shape is adapted to the underlying axonal structure, and demonstrates the increased detection power resulting from their use. One of the main clinical applications of functional MRI is functional localization of the eloquent areas of the brain prior to brain surgery., Funding agencies: CENIIT (Center for industrial information technology) and LiU Cancer, as well as the ITEA/VINNOVA-funded projects IMPACT and ASSIST. Center for Medical Image Science and Visualization (CMIV) at Linköping University.

The dissertation consists of three peer-reviewed publications and is related to the basic research of autism spectrum disorder (ASD), especially the assessment of changes in brain function using functional magnetic resonance imaging (fMRI). The purpose was to discover possible differences in cued and spontaneous brain activity in autistic child and adolescent participants compared to typically developing controls. We used blood oxygen level dependent (BOLD) fMRI imaging of the brain, with which the participants were examined at rest and while looking at facial expressions. The resting state (RS) fMRI data artifacts were reduced, and brain networks were identified using independent component analysis. In addition, the RS was analyzed 1) over the entire measurement period using the regional homogeneity (ReHo) method, which measures local connectivity, and 2) based on the states of different brain networks grouped into shorter periods using the co-activation patterns (CAP) method. Statistically significant differences between groups were found in RS, more clearly with the CAP method. Also, significant differences in brain activity were found between the groups regarding the observation of facial expressions. The dissertation increases the understanding of changes in brain networks related to the autism spectrum, strengthening and supplementing previous research results. Based on our results, analyses of brain networks grouped into similar activation phases of shorter duration are worth further development. The new information can help develop earlier and more accurate imaging diagnostics, tentatively recognizing possible intervention target brain networks and evaluating therapeutic effects., Tiivistelmä Väitöskirja koostuu kolmesta vertaisarvioidusta julkaisusta ja liittyy autismikirjon kehityshäiriön perustutkimukseen, erityisesti aivotoiminnan muutosten arviointiin toiminnallisen magneettikuvauksen (functional MRI, fMRI) avulla. Tutkimuksen tarkoituksena oli selvittää stimuloidun ja spontaanin aivotoiminnan mahdollisia eroavaisuuksia lasten ja nuorten autismikirjossa neurotyypillisiin verrokkeihin nähden. Tutkimusmenetelmänä käytettiin veren happipitoisuudesta riippuvaista aivojen fMRI-kuvausta, jolla osallistujia tutkittiin levossa sekä heidän katsellessaan kasvojen ilmeitä. Itsenäisten komponenttien analyysilla (ICA) vähennettiin lepotilan fMRI-datan häiriöitä ja tunnistettiin aivoverkostoja. Lisäksi lepotilaa analysoitiin 1) koko mittausjakson ajalta signaalien alueellista homogeenisuutta ts. aivojen paikallista kytkennällisyyttä mittaavalla regional homogeneity (ReHo) -menetelmällä ja 2) eri aivoverkostojen tilojen perusteella lyhyemmiksi ajanjaksoiksi ns. yhtäaikaisten aktivaatioiden kuvioihin (co-activation patterns; CAP) ryhmiteltyinä. Näissä löydettiin tilastollisesti merkittäviä ryhmien välisiä eroja, selkeämmin CAP-menetelmällä. Myös kasvojen ilmeiden tarkkailuun liittyen havaittiin tilastollisesti merkittäviä aivotoiminnan eroja ryhmien välillä. Väitöskirja lisää ymmärrystä autismikirjoon liittyvistä aivoverkostojen muutoksista vahvistaen ja täydentäen aiempia tutkimustuloksia. Sen perusteella samankaltaisiin lyhempikestoisiin aktivaatiovaiheisiin ryhmiteltyjen aivoverkostojen analyyseja kannattaa kehittää. Uusi tieto voi auttaa varhaisemman ja tarkemman kuvantamisdiagnostiikan kehittämisessä, tarvittaessa oikeisiin aivoverkostoihin kohdennetuissa interventioissa ja niiden vaikutusten arvioinnissa ja seurannassa.

Goal: Structural brain graphs are conventionally limited to defining nodes as gray matter regions from an atlas,with edges reflecting the density of axonal projections between pairs of nodes. Here we explicitly model the entire set of voxels within a brain mask as nodes of high-resolution, subject-specific graphs. Methods: We define the strength of local voxel-to-voxel connections using diffusion tensors and orientation distribution functions derived from diffusion MRI data. We study the graphs’ Laplacian spectral properties on data from the Human Connectome Project. We then assess the extent of inter-subject variability of the Laplacian eigenmodes via a procrustes validation scheme. Finally, we demonstrate the extent to which functional MRI data are shaped by the underlying anatomical structure via graph signal processing. Results: The graph Laplacian eigenmodes manifest highly resolved spatial profiles, reflecting distributed patterns that correspond to major white matter pathways. We show that the intrinsic dimensionality of the eigenspace of such high-resolution graphs is only a mere fraction of the graph dimensions. By projecting task and resting-state data on low frequency graph Laplacian eigenmodes, we show that brain activity can be well approximated by a small subset of low frequency components. Conclusions: The proposed graphs open new avenues in studying the brain, be it, by exploring their organisational properties via graph or spectral graph theory, or by treating them as the scaffold on which brain function is observed at the individual level.

Background Focal cortical dysplasias (FCD) are a frequent cause of drug-resistant epilepsy in children but are often undetected on structural magnetic resonance imaging (MRI). We aimed to measure and validate the variation of resting state functional MRI (rs-fMRI) blood oxygenation level dependent (BOLD) metrics in surgically proven FCDs in children, to assess the potential yield for detecting and understanding these lesions. Methods We prospectively included pediatric patients with surgically proven FCD with inconclusive structural MRI and healthy controls, who underwent a ten-minute rs-fMRI acquired at 3T. Rs-fMRI data was pre-processed and maps of values of regional homogeneity (ReHo), degree centrality (DC), amplitude of low frequency fluctuations (ALFF) and fractional ALFF (fALFF) were calculated. The variations of BOLD metrics within the to-be-resected areas were analyzed visually, and quantitatively using lateralization indices. BOLD metrics variations were also analyzed in fluorodeoxyglucose-positron emission tomography (FDG-PET) hypometabolic areas. Results We included 7 patients (range: 3-15 years) and 6 aged-matched controls (range: 6-17 years). ReHo lateralization indices were positive in the to-be-resected areas in 4/7 patients, and in 6/7 patients in the additional PET hypometabolic areas. These indices were significantly higher compared to controls in 3/7 and 4/7 patients, respectively. Visual analysis revealed a good spatial correlation between high ReHo areas and MRI structural abnormalities (when present) or PET hypometabolic areas. No consistent variation was seen using DC, ALFF, or fALFF. Conclusion Resting-state fMRI metrics, noticeably increase in ReHo, may have potential to help detect MRI-negative FCDs in combination with other morphological and functional techniques, used in clinical practice and epilepsy-surgery screening.

Goal: Structural brain graphs are conventionally limited to defining nodes as gray matter regions from an atlas,with edges reflecting the density of axonal projections between pairs of nodes. Here we explicitly model the entire set of voxels within a brain mask as nodes of high-resolution, subject-specific graphs. Methods: We define the strength of local voxel-to-voxel connections using diffusion tensors and orientation distribution functions derived from diffusion MRI data. We study the graphs’ Laplacian spectral properties on data from the Human Connectome Project. We then assess the extent of inter-subject variability of the Laplacian eigenmodes via a procrustes validation scheme. Finally, we demonstrate the extent to which functional MRI data are shaped by the underlying anatomical structure via graph signal processing. Results: The graph Laplacian eigenmodes manifest highly resolved spatial profiles, reflecting distributed patterns that correspond to major white matter pathways. We show that the intrinsic dimensionality of the eigenspace of such high-resolution graphs is only a mere fraction of the graph dimensions. By projecting task and resting-state data on low frequency graph Laplacian eigenmodes, we show that brain activity can be well approximated by a small subset of low frequency components. Conclusions: The proposed graphs open new avenues in studying the brain, be it, by exploring their organisational properties via graph or spectral graph theory, or by treating them as the scaffold on which brain function is observed at the individual level.

Goal: Structural brain graphs are conventionally limited to defining nodes as gray matter regions from an atlas,with edges reflecting the density of axonal projections between pairs of nodes. Here we explicitly model the entire set of voxels within a brain mask as nodes of high-resolution, subject-specific graphs. Methods: We define the strength of local voxel-to-voxel connections using diffusion tensors and orientation distribution functions derived from diffusion MRI data. We study the graphs’ Laplacian spectral properties on data from the Human Connectome Project. We then assess the extent of inter-subject variability of the Laplacian eigenmodes via a procrustes validation scheme. Finally, we demonstrate the extent to which functional MRI data are shaped by the underlying anatomical structure via graph signal processing. Results: The graph Laplacian eigenmodes manifest highly resolved spatial profiles, reflecting distributed patterns that correspond to major white matter pathways. We show that the intrinsic dimensionality of the eigenspace of such high-resolution graphs is only a mere fraction of the graph dimensions. By projecting task and resting-state data on low frequency graph Laplacian eigenmodes, we show that brain activity can be well approximated by a small subset of low frequency components. Conclusions: The proposed graphs open new avenues in studying the brain, be it, by exploring their organisational properties via graph or spectral graph theory, or by treating them as the scaffold on which brain function is observed at the individual level.

Functional MRI (fMRI) and EEG may reveal residual consciousness in patients with disorders of consciousness (DoC), as reflected by a rapidly expanding literature on chronic DoC. However, acute DoC is rarely investigated, although identifying residual consciousness is key to clinical decision-making in the intensive care unit (ICU). Therefore, the objective of the prospective, observational, tertiary centre cohort, diagnostic phase IIb study ‘Consciousness in neurocritical care cohort study using EEG and fMRI’ (CONNECT-ME, NCT02644265) was to assess the accuracy of fMRI and EEG to identify residual consciousness in acute DoC in the ICU. Between April 2016 and November 2020, 87 acute DoC patients with traumatic or non-traumatic brain injury were examined with repeated clinical assessments, fMRI and EEG. Resting-state EEG and EEG with external stimulations were evaluated by visual analysis, spectral band analysis and a Support Vector Machine (SVM) consciousness classifier. In addition, within- and between-network resting-state connectivity for canonical resting-state fMRI networks was assessed. Next, we used EEG and fMRI data at study enrolment in two different machine-learning algorithms (Random Forest and SVM with a linear kernel) to distinguish patients in a minimally conscious state or better (≥MCS) from those in coma or unresponsive wakefulness state (≤UWS) at time of study enrolment and at ICU discharge (or before death). Prediction performances were assessed with area under the curve (AUC). Of 87 DoC patients (mean age, 50.0 ± 18 years, 43% female), 51 (59%) were ≤UWS and 36 (41%) were ≥ MCS at study enrolment. Thirty-one (36%) patients died in the ICU, including 28 who had life-sustaining therapy withdrawn. EEG and fMRI predicted consciousness levels at study enrolment and ICU discharge, with maximum AUCs of 0.79 (95% CI 0.77–0.80) and 0.71 (95% CI 0.77–0.80), respectively. Models based on combined EEG and fMRI features predicted consciousness levels at stud

Magnetic resonance imaging (MRI) is one of the pillars of modern medical imaging, providing a non-invasive means to generate 3D images of the body with high soft-tissue contrast. Furthermore, the possibilities afforded by the design of MRI sequences enable the signal to be sensitized to a multitude of physiological tissue properties, resulting in a wide variety of distinct MRI modalities for clinical and research use. This thesis presents a number of advanced brain MRI applications, which fulfill, to differing extents, two complementary aims. On the one hand, they explore the benefits of a multimodal approach to MRI, combining structural, functional and diffusion MRI, in a variety of contexts. On the other, they emphasize the use of advanced mathematical and computational tools in the analysis of MRI data, such as deep learning, Bayesian statistics, and graph signal processing. Paper I introduces an anatomically-adapted extension to previous work in Bayesian spatial priors for functional MRI data, where anatomical information is introduced from a T1-weighted image to compensate for the low anatomical contrast of functional MRI data. It has been observed that the spatial correlation structure of the BOLD signal in brain white matter follows the orientation of the underlying axonal fibers. Paper II argues about the implications of this fact on the ideal shape of spatial filters for the analysis of white matter functional MRI data. By using axonal orientation information extracted from diffusion MRI, and leveraging the possibilities afforded by graph signal processing, a graph-based description of the white matter structure is introduced, which, in turn, enables the definition of spatial filters whose shape is adapted to the underlying axonal structure, and demonstrates the increased detection power resulting from their use. One of the main clinical applications of functional MRI is functional localization of the eloquent areas of the brain prior to brain surgery., Funding agencies: CENIIT (Center for industrial information technology) and LiU Cancer, as well as the ITEA/VINNOVA-funded projects IMPACT and ASSIST. Center for Medical Image Science and Visualization (CMIV) at Linköping University.

Magnetic resonance imaging (MRI) is one of the pillars of modern medical imaging, providing a non-invasive means to generate 3D images of the body with high soft-tissue contrast. Furthermore, the possibilities afforded by the design of MRI sequences enable the signal to be sensitized to a multitude of physiological tissue properties, resulting in a wide variety of distinct MRI modalities for clinical and research use. This thesis presents a number of advanced brain MRI applications, which fulfill, to differing extents, two complementary aims. On the one hand, they explore the benefits of a multimodal approach to MRI, combining structural, functional and diffusion MRI, in a variety of contexts. On the other, they emphasize the use of advanced mathematical and computational tools in the analysis of MRI data, such as deep learning, Bayesian statistics, and graph signal processing. Paper I introduces an anatomically-adapted extension to previous work in Bayesian spatial priors for functional MRI data, where anatomical information is introduced from a T1-weighted image to compensate for the low anatomical contrast of functional MRI data. It has been observed that the spatial correlation structure of the BOLD signal in brain white matter follows the orientation of the underlying axonal fibers. Paper II argues about the implications of this fact on the ideal shape of spatial filters for the analysis of white matter functional MRI data. By using axonal orientation information extracted from diffusion MRI, and leveraging the possibilities afforded by graph signal processing, a graph-based description of the white matter structure is introduced, which, in turn, enables the definition of spatial filters whose shape is adapted to the underlying axonal structure, and demonstrates the increased detection power resulting from their use. One of the main clinical applications of functional MRI is functional localization of the eloquent areas of the brain prior to brain surgery., Funding agencies: CENIIT (Center for industrial information technology) and LiU Cancer, as well as the ITEA/VINNOVA-funded projects IMPACT and ASSIST. Center for Medical Image Science and Visualization (CMIV) at Linköping University.

Magnetic resonance imaging (MRI) is one of the pillars of modern medical imaging, providing a non-invasive means to generate 3D images of the body with high soft-tissue contrast. Furthermore, the possibilities afforded by the design of MRI sequences enable the signal to be sensitized to a multitude of physiological tissue properties, resulting in a wide variety of distinct MRI modalities for clinical and research use. This thesis presents a number of advanced brain MRI applications, which fulfill, to differing extents, two complementary aims. On the one hand, they explore the benefits of a multimodal approach to MRI, combining structural, functional and diffusion MRI, in a variety of contexts. On the other, they emphasize the use of advanced mathematical and computational tools in the analysis of MRI data, such as deep learning, Bayesian statistics, and graph signal processing. Paper I introduces an anatomically-adapted extension to previous work in Bayesian spatial priors for functional MRI data, where anatomical information is introduced from a T1-weighted image to compensate for the low anatomical contrast of functional MRI data. It has been observed that the spatial correlation structure of the BOLD signal in brain white matter follows the orientation of the underlying axonal fibers. Paper II argues about the implications of this fact on the ideal shape of spatial filters for the analysis of white matter functional MRI data. By using axonal orientation information extracted from diffusion MRI, and leveraging the possibilities afforded by graph signal processing, a graph-based description of the white matter structure is introduced, which, in turn, enables the definition of spatial filters whose shape is adapted to the underlying axonal structure, and demonstrates the increased detection power resulting from their use. One of the main clinical applications of functional MRI is functional localization of the eloquent areas of the brain prior to brain surgery., Funding agencies: CENIIT (Center for industrial information technology) and LiU Cancer, as well as the ITEA/VINNOVA-funded projects IMPACT and ASSIST. Center for Medical Image Science and Visualization (CMIV) at Linköping University.

This dataset contains resting-state functional MRI data used in the study "Non-invasive modulation of human corticostriatal activity" (Caballero-Insaurriaga et al, PNAS, 2023). In this study two datasets were used: one from a transcranial static-magnetic-field stimulation (tSMS) experiment (tSMS20) and another one from the Human Connectome Project (HCP100). The tSMS20 dataset was originally acquired for a previous study tSMS over the Supplementary Motor Area (Pineda-Pardo et al, Commun Biol, 2019). The regions used in the study are also provided. As for the tSMS20 dataset, the stimulation protocol consisted of 30-minute tSMS using a single magnet placed over the supplementary motor area (SMA). Each subject underwent two stimulation sessions (real and sham) in two separate days, whose order was randomized. In each session, structural MRI was acquired before tSMS, and resting-state fMRI before and after. Structural images were T1-weighted (T1w), with 1 mm isotropic voxel. Functional data was acquired in 10 minutes-long sessions, TR/TE 2400/30 ms (250 volumes per session), with 3mm isotropic voxel. The preprocessed resting-state fMRI data are included in this repository (see dataset_description.txt file and Pineda-Pardo et al, Commun Biol, 2019 for more details) As for the HCP100 dataset, only the subject list is included, as data are already publicly available from the HCP initiative.

Published on 28 December 2022, The testing effect describes the finding that retrieval practice enhances memory performance compared to restudy practice. Prior evidence demonstrates that this effect can be boosted by providing feedback after retrieval attempts (i.e. test-potentiated encoding [TPE]). The present fMRI study investigated the neural processes during successful memory retrieval underlying this beneficial effect of correct answer feedback compared with restudy and whether additional performance feedback leads to further benefits. Twenty-seven participants learned cue-target pairs by (i) restudying, (ii) standard TPE including a restudy opportunity, or (iii) TPE including a restudy opportunity immediately after a positive or negative performance feedback. One day later, a cued retrieval recognition test was performed inside the MRI scanner. Behavioral results confirmed the testing effect and that adding explicit performance feedback-enhanced memory relative to restudy and standard TPE. Stronger functional engagement while retrieving items previously restudied was found in lateral prefrontal cortex and superior parietal lobe. By contrast, lateral temporo-parietal areas were more strongly recruited while retrieving items previously tested. Performance feedback increased the hippocampal activation and resulted in stronger functional coupling between hippocampus, supramarginal gyrus, and ventral striatum with lateral temporo-parietal cortex. Our results unveil the main functional dynamics and connectivity nodes underlying memory benefits from additional performance feedback.

Functional MRI (fMRI) and EEG may reveal residual consciousness in patients with disorders of consciousness (DoC), as reflected by a rapidly expanding literature on chronic DoC. However, acute DoC is rarely investigated, although identifying residual consciousness is key to clinical decision-making in the intensive care unit (ICU). Therefore, the objective of the prospective, observational, tertiary centre cohort, diagnostic phase IIb study ‘Consciousness in neurocritical care cohort study using EEG and fMRI’ (CONNECT-ME, NCT02644265) was to assess the accuracy of fMRI and EEG to identify residual consciousness in acute DoC in the ICU. Between April 2016 and November 2020, 87 acute DoC patients with traumatic or non-traumatic brain injury were examined with repeated clinical assessments, fMRI and EEG. Resting-state EEG and EEG with external stimulations were evaluated by visual analysis, spectral band analysis and a Support Vector Machine (SVM) consciousness classifier. In addition, within- and between-network resting-state connectivity for canonical resting-state fMRI networks was assessed. Next, we used EEG and fMRI data at study enrolment in two different machine-learning algorithms (Random Forest and SVM with a linear kernel) to distinguish patients in a minimally conscious state or better (≥MCS) from those in coma or unresponsive wakefulness state (≤UWS) at time of study enrolment and at ICU discharge (or before death). Prediction performances were assessed with area under the curve (AUC). Of 87 DoC patients (mean age, 50.0 ± 18 years, 43% female), 51 (59%) were ≤UWS and 36 (41%) were ≥ MCS at study enrolment. Thirty-one (36%) patients died in the ICU, including 28 who had life-sustaining therapy withdrawn. EEG and fMRI predicted consciousness levels at study enrolment and ICU discharge, with maximum AUCs of 0.79 (95% CI 0.77–0.80) and 0.71 (95% CI 0.77–0.80), respectively. Models based on combined EEG and fMRI features predicted consciousness levels at stud

BackgroundFronto-striatal disconnection is thought to be at the basis of dysexecutive symptoms in patients with Parkinson's disease (PD). Multiple reserve-related processes may offer resilience against functional decline. Among these, cognitive reserve (CR) refers to the adaptability of cognitive processes.ObjectiveTo test the hypothesis that functional connectivity of pathways associated with executive dysfunction in PD is modulated by CR.MethodsTwenty-six PD patients and 24 controls underwent resting-state functional magnetic resonance imaging. Functional connectivity was explored with independent component analysis and seed-based approaches. The following networks were selected from the outcome of the independent component analysis: default-mode (DMN), left and right fronto-parietal (l/rFPN), salience (SalN), sensorimotor (SMN), and occipital visual (OVN). Seed regions were selected in the substantia nigra and in the dorsolateral and ventromedial prefrontal cortex for the assessment of seed-based functional connectivity maps. Educational and occupational attainments were used as CR proxies.ResultsCompared with their counterparts with high CR, PD individuals with low CR had reduced posterior DMN functional connectivity in the anterior cingulate and basal ganglia, and bilaterally reduced connectivity in fronto-parietal regions within the networks defined by the dorsolateral and ventrolateral prefrontal seeds. Hyper-connectivity was detected within medial prefrontal regions when comparing low-CR PD with low-CR controls.ConclusionCR may exert a modulatory effect on functional connectivity in basal ganglia and executive-attentional fronto-parietal networks. In PD patients with low CR, attentional control networks seem to be downregulated, whereas higher recruitment of medial frontal regions suggests compensation via an upregulation mechanism. This upregulation might contribute to maintaining efficient cognitive functioning when posterior cortical function is progressi

MRI is undoubtedly the cornerstone of brain tumor imaging, playing a key role in all phases of patient management, starting from diagnosis, through therapy planning, to treatment response and/or recurrence assessment. Currently, neuroimaging can describe morphologic and non-morphologic (functional, hemodynamic, metabolic, cellular, microstructural, and sometimes even genetic) characteristics of brain tumors, greatly contributing to diagnosis and follow-up. Knowing the technical aspects, strength and limits of each MR technique is crucial to correctly interpret MR brain studies and to address clinicians to the best treatment strategy. This article aimed to provide an overview of neuroimaging in the assessment of adult primary brain tumors. We started from the basilar role of conventional/morphological MR sequences, then analyzed, one by one, the non-morphological techniques, and finally highlighted future perspectives, such as radiomics and artificial intelligence.

Following traumatic brain injury (TBI), cognitive impairments manifest through interactions between microscopic and macroscopic changes. On the microscale, a neurometabolic cascade alters neurotransmission, while on the macroscale diffuse axonal injury impacts the integrity of long-range connections. Large-scale brain network modeling allows us to make predictions across these spatial scales by integrating neuroimaging data with biophysically based models to investigate how microscale changes invisible to conventional neuroimaging influence large-scale brain dynamics. To this end, we analyzed structural and functional neuroimaging data from a well characterized sample of 44 adult TBI patients recruited from a regional trauma center, scanned at 1-2 weeks postinjury, and with follow-up behavioral outcome assessed 6 months later. Thirty-six age-matched healthy adults served as comparison participants. Using The Virtual Brain, we fit simulations of whole-brain resting-state functional MRI to the empirical static and dynamic functional connectivity of each participant. Multivariate partial least squares (PLS) analysis showed that patients with acute traumatic intracranial lesions had lower cortical regional inhibitory connection strengths than comparison participants, while patients without acute lesions did not differ from the comparison group. Further multivariate PLS analyses found correlations between lower semiacute regional inhibitory connection strengths and more symptoms and lower cognitive performance at a 6 month follow-up. Critically, patients without acute lesions drove this relationship, suggesting clinical relevance of regional inhibitory connection strengths even when traumatic intracranial lesions were not present. Our results suggest that large-scale connectome-based models may be sensitive to pathophysiological changes in semi-acute phase TBI patients and predictive of their chronic outcomes.

Recent neuroimaging studies have reported alterations in brain activation during cognitive tasks in cancer patients who have undergone chemotherapy treatment. However, the location of these altered brain activation patterns after chemotherapy varies considerably across studies. The aim of the present meta-analysis was to quantitatively synthesise this body of evidence using Activation Likelihood Estimation to identify reliable regions of altered brain activation in cancer patients treated with chemotherapy, compared to healthy controls and no chemotherapy controls. Our systematic search identified 12 studies that adopted task-related fMRI on non-central nervous system cancer patients who received chemotherapy relative to controls. All studies were included in the analyses and were grouped into four contrasts. Cancer patients treated with chemotherapy showed reduced activation in the left superior parietal lobe/precuneus (family-wise error corrected p < .05) compared to no chemotherapy controls. No significant clusters were found in three of our contrasts. The majority of studies did not support an association between altered brain activation and cognitive performance after chemotherapy. Findings point towards a possible chemotherapy-induced alteration, which could inform targeted treatment strategies. With continued work in this field using homogenous task-related protocols and cancer populations, fMRI may be used as a biomarker of cognitive deficits in the future.

Simultaneous mapping of multiple behavioral domains into brain networks remains a major challenge. Here, we shed some light on this problem by employing a combination of machine learning, structural and functional brain networks at different spatial resolutions (also known as scales), together with performance scores across multiple neurobehavioral domains, including sensation, motor skills, and cognition. Provided by the Human Connectome Project, we make use of three cohorts: 640 participants for model training, 160 subjects for validation, and 200 subjects for model performance testing thus enhancing prediction generalization. Our modeling consists of two main stages, namely dimensionality reduction in brain network features at multiple scales, followed by canonical correlation analysis, which determines an optimal linear combination of connectivity features to predict multiple behavioral performance scores. To assess the differences in the predictive power of each modality, we separately applied three different strategies: structural unimodal, functional unimodal, and multimodal, that is, structural in combination with functional features of the brain network. Our results show that the multimodal association outperforms any of the unimodal analyses. Then, to answer which human brain structures were most involved in predicting multiple behavioral scores, we simulated different synthetic scenarios in which in each case we completely deleted a brain structure or a complete resting state network, and recalculated performance in its absence. In deletions, we found critical structures to affect performance when predicting single behavioral domains, but this occurred in a lesser manner for prediction of multi-domain behavior. Overall, our results confirm that although there are synergistic contributions between brain structure and function that enhance behavioral prediction, brain networks may also be mutually redundant in predicting multidomain behavior, such that even

Background and aims: Compulsive sexual behavior disorder (CSBD) is characterized by persistent patterns of failure to control sexual impulses resulting in repetitive sexual behavior, pursued despite adverse consequences. Despite previous indications of addiction-like mechanisms and the recent impulse-control disorder classification in the International Classification of Diseases (ICD-11), the neurobiological processes underlying CSBD are unknown. Methods: We designed and applied a behavioral paradigm aimed at disentangling processes related to anticipation and viewing of erotic stimuli. In 22 male CSBD patients (age: M = 38.7, SD = 11.7) and 20 healthy male controls (HC, age: M = 37.6, SD = 8.5), we measured behavioral responses and neural activity during functional magnetic resonance imaging (fMRI). The main outcomes were response time differences between erotic and non-erotic trials and ventral striatum (VS) activity during anticipation of visual stimuli. We related these outcomes with each other, to CSBD diagnosis, and symptom severity. Results: We found robust case-control differences on behavioral level, where CSBD patients showed larger response time differences between erotic and non-erotic trials than HC. The task induced reliable main activations within each group. While we did not observe significant group differences in VS activity, VS activity during anticipation correlated with response time differences and self-ratings for anticipation of erotic stimuli. Discussion and Conclusions: Our results support the validity and applicability of the developed task and suggest that CSBD is associated with altered behavioral correlates of anticipation, which were associated with ventral striatum activity during anticipation of erotic stimuli. This supports the idea that addiction-like mechanisms play a role in CSBD.

Background and aims: Compulsive sexual behavior disorder (CSBD) is characterized by persistent patterns of failure to control sexual impulses resulting in repetitive sexual behavior, pursued despite adverse consequences. Despite previous indications of addiction-like mechanisms and the recent impulse-control disorder classification in the International Classification of Diseases (ICD-11), the neurobiological processes underlying CSBD are unknown. Methods: We designed and applied a behavioral paradigm aimed at disentangling processes related to anticipation and viewing of erotic stimuli. In 22 male CSBD patients (age: M = 38.7, SD = 11.7) and 20 healthy male controls (HC, age: M = 37.6, SD = 8.5), we measured behavioral responses and neural activity during functional magnetic resonance imaging (fMRI). The main outcomes were response time differences between erotic and non-erotic trials and ventral striatum (VS) activity during anticipation of visual stimuli. We related these outcomes with each other, to CSBD diagnosis, and symptom severity. Results: We found robust case-control differences on behavioral level, where CSBD patients showed larger response time differences between erotic and non-erotic trials than HC. The task induced reliable main activations within each group. While we did not observe significant group differences in VS activity, VS activity during anticipation correlated with response time differences and self-ratings for anticipation of erotic stimuli. Discussion and Conclusions: Our results support the validity and applicability of the developed task and suggest that CSBD is associated with altered behavioral correlates of anticipation, which were associated with ventral striatum activity during anticipation of erotic stimuli. This supports the idea that addiction-like mechanisms play a role in CSBD.

Background and aims: Compulsive sexual behavior disorder (CSBD) is characterized by persistent patterns of failure to control sexual impulses resulting in repetitive sexual behavior, pursued despite adverse consequences. Despite previous indications of addiction-like mechanisms and the recent impulse-control disorder classification in the International Classification of Diseases (ICD-11), the neurobiological processes underlying CSBD are unknown. Methods: We designed and applied a behavioral paradigm aimed at disentangling processes related to anticipation and viewing of erotic stimuli. In 22 male CSBD patients (age: M = 38.7, SD = 11.7) and 20 healthy male controls (HC, age: M = 37.6, SD = 8.5), we measured behavioral responses and neural activity during functional magnetic resonance imaging (fMRI). The main outcomes were response time differences between erotic and non-erotic trials and ventral striatum (VS) activity during anticipation of visual stimuli. We related these outcomes with each other, to CSBD diagnosis, and symptom severity. Results: We found robust case-control differences on behavioral level, where CSBD patients showed larger response time differences between erotic and non-erotic trials than HC. The task induced reliable main activations within each group. While we did not observe significant group differences in VS activity, VS activity during anticipation correlated with response time differences and self-ratings for anticipation of erotic stimuli. Discussion and Conclusions: Our results support the validity and applicability of the developed task and suggest that CSBD is associated with altered behavioral correlates of anticipation, which were associated with ventral striatum activity during anticipation of erotic stimuli. This supports the idea that addiction-like mechanisms play a role in CSBD.

Much of our understanding of the neural basis of behavioral differences are attributable to studies conducted with magnetic resonance imaging (MRI). Despite its critical role in cognitive neuroscience, MRI findings lack generalizability for translational uses due to the small and homogeneous samples of traditional MRI studies. To improve on the generalizability issue, large-scale, population-based neuroimaging studies are conducted where MRI data are collected from thousands of participants that are systematically sampled to represent the general population. Now, we have a unique opportunity to harness the statistical power afforded by population-based neuroimaging studies to characterize and quantify the behavioral relevance of MRI measures at the population level.The goal of this work is to take advance of a population-based neurodevelopmental study, the Adolescent Brain Cognitive Development (ABCD) Study, to shed light on the optimal fMRI design and analysis pipelines for the detection of behaviorally relevant brain signals. In Chapter 2, I challenged the traditional statistical mapping approach of MRI analysis which assumed that the behavioral differences are explained by sparse, localized brain regions. I demonstrated that at a large, population level, the effects of the association between brain activations and behavioral differences are not sparse but distributed across the cortex (Chapter 2). Aggregating the small effect sizes across the whole cortex can greatly increase the magnitude of the behavioral associations detected by fMRI tasks. This finding is consistent with the observation that behavioral differences are associated with individual differences in distributed, functional brain networks whose activities are measured by functional connectivity (FC), the pairwise correlation of activation across brain regions. In Chapter 3, I carried out a systematic investigation of the optimal fMRI paradigms for the detection of behaviorally relevant FC patterns by

Following traumatic brain injury (TBI), cognitive impairments manifest through interactions between microscopic and macroscopic changes. On the microscale, a neurometabolic cascade alters neurotransmission, while on the macroscale diffuse axonal injury impacts the integrity of long-range connections. Large-scale brain network modeling allows us to make predictions across these spatial scales by integrating neuroimaging data with biophysically based models to investigate how microscale changes invisible to conventional neuroimaging influence large-scale brain dynamics. To this end, we analyzed structural and functional neuroimaging data from a well characterized sample of 44 adult TBI patients recruited from a regional trauma center, scanned at 1-2 weeks postinjury, and with follow-up behavioral outcome assessed 6 months later. Thirty-six age-matched healthy adults served as comparison participants. Using The Virtual Brain, we fit simulations of whole-brain resting-state functional MRI to the empirical static and dynamic functional connectivity of each participant. Multivariate partial least squares (PLS) analysis showed that patients with acute traumatic intracranial lesions had lower cortical regional inhibitory connection strengths than comparison participants, while patients without acute lesions did not differ from the comparison group. Further multivariate PLS analyses found correlations between lower semiacute regional inhibitory connection strengths and more symptoms and lower cognitive performance at a 6 month follow-up. Critically, patients without acute lesions drove this relationship, suggesting clinical relevance of regional inhibitory connection strengths even when traumatic intracranial lesions were not present. Our results suggest that large-scale connectome-based models may be sensitive to pathophysiological changes in semi-acute phase TBI patients and predictive of their chronic outcomes.

Background Differentiating between bipolar disorder (BD) and major depressive disorder (MDD) during the depressive episode is an important clinical challenge. Reward system abnormalities have received much attention as one of the biological underpinnings of BD and MDD, but few studies have directly compared these abnormalities in remitted and depressed states. Methods This was a functional MRI study using the Monetary Incentive Delay task in 65 patients (BD [n = 33], MDD [n = 32]) and 33 healthy controls (HC). Regions of interest (ROI) analysis with 21 ROIs related to reward anticipation and 17 ROIs related to gain outcome were implemented, as well as whole-brain analysis. The difference in the dimensional effect of depression on brain activation was also examined. Results Relative to the HC group, BD patients showed significantly decreased activation during reward anticipation in the anterior cingulate cortex, anterior insula (AI), and putamen, and MDD patients showed significantly decreased activation in the AI and brainstem. The dimensional effect of depression severity showed a trend-level difference between BD and MDD in the right brainstem and left AI. Conclusions The current study showed a possible differential effect of depression on the reward system between MDD and BD. Further studies on reward systems might offer reliable markers to distinguish between MDD and BD patients in the depressive phase.

After falling asleep, the brain needs to detach from waking activity and reorganize into a functionally distinct state. A functional MRI (fMRI) study has recently revealed that the transition to unconsciousness induced by propofol involves a global decline of brain activity followed by a transient reduction in cortico-subcortical coupling. We have analyzed the relationships between transitional brain activity and breathing changes as one example of a vital function that needs the brain to readapt. Thirty healthy participants were originally examined. The analysis involved the correlation between breathing and fMRI signal upon loss of consciousness. We proposed that a decrease in ventilation would be coupled to the initial decline in fMRI signal in brain areas relevant for modulating breathing in the awake state, and that the subsequent recovery would be coupled to fMRI signal in structures relevant for controlling breathing during the unconscious state. Results showed that a slight reduction in breathing from wakefulness to unconsciousness was distinctively associated with decreased activity in brain systems underlying different aspects of consciousness including the prefrontal cortex, the default mode network and somatosensory areas. Breathing recovery was distinctively coupled to activity in deep brain structures controlling basic behaviors such as the hypothalamus and amygdala. Activity in the brainstem, cerebellum and hippocampus was associated with breathing variations in both states. Therefore, our brain maps illustrate potential drives to breathe, unique to wakefulness, in the form of brain systems underlying cognitive awareness, self-awareness and sensory awareness, and to unconsciousness involving structures controlling instinctive and homeostatic behaviors., Peer Reviewed, Postprint (published version)

Background Differentiating between bipolar disorder (BD) and major depressive disorder (MDD) during the depressive episode is an important clinical challenge. Reward system abnormalities have received much attention as one of the biological underpinnings of BD and MDD, but few studies have directly compared these abnormalities in remitted and depressed states. Methods This was a functional MRI study using the Monetary Incentive Delay task in 65 patients (BD [n = 33], MDD [n = 32]) and 33 healthy controls (HC). Regions of interest (ROI) analysis with 21 ROIs related to reward anticipation and 17 ROIs related to gain outcome were implemented, as well as whole-brain analysis. The difference in the dimensional effect of depression on brain activation was also examined. Results Relative to the HC group, BD patients showed significantly decreased activation during reward anticipation in the anterior cingulate cortex, anterior insula (AI), and putamen, and MDD patients showed significantly decreased activation in the AI and brainstem. The dimensional effect of depression severity showed a trend-level difference between BD and MDD in the right brainstem and left AI. Conclusions The current study showed a possible differential effect of depression on the reward system between MDD and BD. Further studies on reward systems might offer reliable markers to distinguish between MDD and BD patients in the depressive phase.

Background: Exposure to 40 Hz stroboscopic light, for one hour a day, has previously been published as a potential treatment option for Alzheimer's disease in animal models. However, exposure for an hour a day to 40 Hz stroboscopic light can be strenuous and examining other types of 40 Hz inducing stimuli is paramount if chronic treatment is wanted. Objective: A core assumption behind ensuring a therapeutic outcome is that the visual stimuli can induce 40 Hz gamma entrainment. Here, we examine whether a specific visual stimulus, 40 Hz invisible spectral flicker (ISF), can induce gamma entrainment and how it differs from both continuous light (CON) and 40 Hz stroboscopic light (STROBE). Methods: The study included non-simultaneous EEG-fMRI neuroimaging of 13 young healthy volunteers during light exposure. Each light condition (i.e., CON, ISF, or STROBE) was active for 30 seconds followed immediately by the next. Results: Entrainment of 40 Hz neural activity were significantly higher signal-to-noise ratio during exposure to ISF (mean: 3.03, 95% CI 2.07 to 3.99) and STROBE (mean: 12.04, 95% CI 10.18 to 13.87) compared to CON. Additionally STROBE had a higher entrainment than ISF (mean: 9.01, 95% CI 7.16 to 12.14). Conclusion: This study presents a novel method of 40 Hz entrainment using ISF. This enables the possibility of future randomized placebo-controlled clinical trials with acceptable double blinding due to the essentially imperceivable flicker, which is expected to substantially reduce discomfort compared to interventions with stroboscopic flicker.

Background Differentiating between bipolar disorder (BD) and major depressive disorder (MDD) during the depressive episode is an important clinical challenge. Reward system abnormalities have received much attention as one of the biological underpinnings of BD and MDD, but few studies have directly compared these abnormalities in remitted and depressed states. Methods This was a functional MRI study using the Monetary Incentive Delay task in 65 patients (BD [n = 33], MDD [n = 32]) and 33 healthy controls (HC). Regions of interest (ROI) analysis with 21 ROIs related to reward anticipation and 17 ROIs related to gain outcome were implemented, as well as whole-brain analysis. The difference in the dimensional effect of depression on brain activation was also examined. Results Relative to the HC group, BD patients showed significantly decreased activation during reward anticipation in the anterior cingulate cortex, anterior insula (AI), and putamen, and MDD patients showed significantly decreased activation in the AI and brainstem. The dimensional effect of depression severity showed a trend-level difference between BD and MDD in the right brainstem and left AI. Conclusions The current study showed a possible differential effect of depression on the reward system between MDD and BD. Further studies on reward systems might offer reliable markers to distinguish between MDD and BD patients in the depressive phase.

Background: The mechanisms underlying the antidepressant effect and cognitive side effects of Electroconvulsive Therapy (ECT) remain elusive. The measurement of cerebral perfusion provides an insight into brain physiology. Objective: We investigated ECT-related perfusion changes in depressed patients and tested whether these changes correlate with clinical effects. Methods: A sample of 22 in-patients was examined at three time points: 1) within two days before, 2) within one week after, and 3) six months after an ECT series. Cerebral perfusion was quantified using arterial spin labeling magnetic resonance imaging. The primary regions of interest were the bilateral dorsolateral prefrontal cortices (DL-PFC) and hippocampi. The depression severity was assessed by the six-item Hamilton Depression Rating Scale, and cognitive performance by the Screen for Cognitive Impairment in Psychiatry. A linear mixed model and partial correlation were used for statistical analyses. Results: Following an ECT series, perfusion decreased in the right (−6.0%, p =.01) and left DL-PFC (−5.6%, p =.001). Perfusion increased in the left hippocampus (4.8%, p =.03), while on the right side the increase was insignificant (2.3%, p =.23). A larger perfusion reduction in the right DL-PFC correlated with a better antidepressant effect, and a larger perfusion increase in the right hippocampus with worse cognitive impairment. Conclusion: ECT-induced attenuation of prefrontal activity may be related to clinical improvement, whereas a hippocampal process triggered by the treatment is likely associated with cognitive side effects.

Background: Exposure to 40 Hz stroboscopic light, for one hour a day, has previously been published as a potential treatment option for Alzheimer's disease in animal models. However, exposure for an hour a day to 40 Hz stroboscopic light can be strenuous and examining other types of 40 Hz inducing stimuli is paramount if chronic treatment is wanted. Objective: A core assumption behind ensuring a therapeutic outcome is that the visual stimuli can induce 40 Hz gamma entrainment. Here, we examine whether a specific visual stimulus, 40 Hz invisible spectral flicker (ISF), can induce gamma entrainment and how it differs from both continuous light (CON) and 40 Hz stroboscopic light (STROBE). Methods: The study included non-simultaneous EEG-fMRI neuroimaging of 13 young healthy volunteers during light exposure. Each light condition (i.e., CON, ISF, or STROBE) was active for 30 seconds followed immediately by the next. Results: Entrainment of 40 Hz neural activity were significantly higher signal-to-noise ratio during exposure to ISF (mean: 3.03, 95% CI 2.07 to 3.99) and STROBE (mean: 12.04, 95% CI 10.18 to 13.87) compared to CON. Additionally STROBE had a higher entrainment than ISF (mean: 9.01, 95% CI 7.16 to 12.14). Conclusion: This study presents a novel method of 40 Hz entrainment using ISF. This enables the possibility of future randomized placebo-controlled clinical trials with acceptable double blinding due to the essentially imperceivable flicker, which is expected to substantially reduce discomfort compared to interventions with stroboscopic flicker.

Purpose of Review The aim was to synthesize key findings regarding the use of functional MRI (fMRI) to assess olfactory dysfunction (OD), and thus, to evaluate whether fMRI could be a reliable clinical diagnostic tool. Recent Findings In response to olfactory stimulation, patients with quantitative OD display reduced activation in olfactory-related brain regions but also stronger activation in non-olfactory brain areas. Parosmic patients also seem to show both weaker and higher brain signals. As to trigeminal chemosensory system, fMRI suggests that central processing may be declined in patients with OD. Functional connectivity studies report a possible correlation between altered neuronal connections within brain networks and olfactory performances. fMRI emerges as a valuable and promising objective method in OD evaluation. Yet, its high inter-individual variability still precludes its routine clinical use for diagnostic purpose. Future research should focus on optimizing stimulation paradigms and analysis methods.

Background: The mechanisms underlying the antidepressant effect and cognitive side effects of Electroconvulsive Therapy (ECT) remain elusive. The measurement of cerebral perfusion provides an insight into brain physiology. Objective: We investigated ECT-related perfusion changes in depressed patients and tested whether these changes correlate with clinical effects. Methods: A sample of 22 in-patients was examined at three time points: 1) within two days before, 2) within one week after, and 3) six months after an ECT series. Cerebral perfusion was quantified using arterial spin labeling magnetic resonance imaging. The primary regions of interest were the bilateral dorsolateral prefrontal cortices (DL-PFC) and hippocampi. The depression severity was assessed by the six-item Hamilton Depression Rating Scale, and cognitive performance by the Screen for Cognitive Impairment in Psychiatry. A linear mixed model and partial correlation were used for statistical analyses. Results: Following an ECT series, perfusion decreased in the right (−6.0%, p =.01) and left DL-PFC (−5.6%, p =.001). Perfusion increased in the left hippocampus (4.8%, p =.03), while on the right side the increase was insignificant (2.3%, p =.23). A larger perfusion reduction in the right DL-PFC correlated with a better antidepressant effect, and a larger perfusion increase in the right hippocampus with worse cognitive impairment. Conclusion: ECT-induced attenuation of prefrontal activity may be related to clinical improvement, whereas a hippocampal process triggered by the treatment is likely associated with cognitive side effects.

Objectives: Understanding the dynamics of reorganized network-level brain functions after hemispherectomy is important for treatment, prognostication, and rehabilitation of brain injury, but also for investigating questions of fundamental neurobehavioral interest: How does the brain promote consciousness despite loss of one hemisphere? Methods: We studied resting-state functional connectivity (RSFC) in a high-functioning middle-aged man 6 years after functional hemispherectomy following malignant middle cerebral artery infarction, and we compared results to RSFC in 20 healthy controls. Results: Our analysis indicates increased between-network connectivity for all seven networks examined in the patient's preserved hemisphere, compared to healthy controls, suggesting a shift toward increased between-network connectivity following near-complete loss of one hemisphere during adulthood. Conclusions: These data corroborate and extend recent findings of increased between-network connectivity in the remaining hemisphere after surgical hemispherectomy for intractable epilepsy during childhood. Our results support a neuroplasticity model with reorganization of distributed brain connectivity within the preserved hemisphere as part of the road to recovery after brain injury, as well as recovery of consciousness and cognitive functions, after hemispherectomy.

Background: Exposure to 40 Hz stroboscopic light, for one hour a day, has previously been published as a potential treatment option for Alzheimer's disease in animal models. However, exposure for an hour a day to 40 Hz stroboscopic light can be strenuous and examining other types of 40 Hz inducing stimuli is paramount if chronic treatment is wanted. Objective: A core assumption behind ensuring a therapeutic outcome is that the visual stimuli can induce 40 Hz gamma entrainment. Here, we examine whether a specific visual stimulus, 40 Hz invisible spectral flicker (ISF), can induce gamma entrainment and how it differs from both continuous light (CON) and 40 Hz stroboscopic light (STROBE). Methods: The study included non-simultaneous EEG-fMRI neuroimaging of 13 young healthy volunteers during light exposure. Each light condition (i.e., CON, ISF, or STROBE) was active for 30 seconds followed immediately by the next. Results: Entrainment of 40 Hz neural activity were significantly higher signal-to-noise ratio during exposure to ISF (mean: 3.03, 95% CI 2.07 to 3.99) and STROBE (mean: 12.04, 95% CI 10.18 to 13.87) compared to CON. Additionally STROBE had a higher entrainment than ISF (mean: 9.01, 95% CI 7.16 to 12.14). Conclusion: This study presents a novel method of 40 Hz entrainment using ISF. This enables the possibility of future randomized placebo-controlled clinical trials with acceptable double blinding due to the essentially imperceivable flicker, which is expected to substantially reduce discomfort compared to interventions with stroboscopic flicker.

Simultaneous mapping of multiple behavioral domains into brain networks remains a major challenge. Here, we shed some light on this problem by employing a combination of machine learning, structural and functional brain networks at different spatial resolutions (also known as scales), together with performance scores across multiple neurobehavioral domains, including sensation, motor skills, and cognition. Provided by the Human Connectome Project, we make use of three cohorts: 640 participants for model training, 160 subjects for validation, and 200 subjects for model performance testing thus enhancing prediction generalization. Our modeling consists of two main stages, namely dimensionality reduction in brain network features at multiple scales, followed by canonical correlation analysis, which determines an optimal linear combination of connectivity features to predict multiple behavioral performance scores. To assess the differences in the predictive power of each modality, we separately applied three different strategies: structural unimodal, functional unimodal, and multimodal, that is, structural in combination with functional features of the brain network. Our results show that the multimodal association outperforms any of the unimodal analyses. Then, to answer which human brain structures were most involved in predicting multiple behavioral scores, we simulated different synthetic scenarios in which in each case we completely deleted a brain structure or a complete resting state network, and recalculated performance in its absence. In deletions, we found critical structures to affect performance when predicting single behavioral domains, but this occurred in a lesser manner for prediction of multi-domain behavior. Overall, our results confirm that although there are synergistic contributions between brain structure and function that enhance behavioral prediction, brain networks may also be mutually redundant in predicting multidomain behavior, such that even

With the advent of advanced analytical methods applied to functional neuroimaging and neurophysiological data, cerebral conditions have been defined that challenge the established classification of disorders of consciousness. A subset of brain-damaged patients has been identified who cannot carry out motor commands, but who exhibit patterns of cerebral activation during mental imagery tasks that are indistinguishable from those in healthy controls. This condition, termed "cognitive motor dissociation", has overturned many assumptions regarding the detection, diagnosis, prognosis, and care of patients with brain injury. Three factors are likely to influence efforts to improve the classification of disorders of consciousness in the near future: the types of data that will become available to characterize brain states, the modeling paradigms utilized for data analysis, and the ability to implement classification schemes in the clinical setting. Here we review past achievements, present states, and future projections for the classification of impaired consciousness and responsiveness.

Objectives: Understanding the dynamics of reorganized network-level brain functions after hemispherectomy is important for treatment, prognostication, and rehabilitation of brain injury, but also for investigating questions of fundamental neurobehavioral interest: How does the brain promote consciousness despite loss of one hemisphere? Methods: We studied resting-state functional connectivity (RSFC) in a high-functioning middle-aged man 6 years after functional hemispherectomy following malignant middle cerebral artery infarction, and we compared results to RSFC in 20 healthy controls. Results: Our analysis indicates increased between-network connectivity for all seven networks examined in the patient's preserved hemisphere, compared to healthy controls, suggesting a shift toward increased between-network connectivity following near-complete loss of one hemisphere during adulthood. Conclusions: These data corroborate and extend recent findings of increased between-network connectivity in the remaining hemisphere after surgical hemispherectomy for intractable epilepsy during childhood. Our results support a neuroplasticity model with reorganization of distributed brain connectivity within the preserved hemisphere as part of the road to recovery after brain injury, as well as recovery of consciousness and cognitive functions, after hemispherectomy.

Background and aims: Compulsive sexual behavior disorder (CSBD) is characterized by persistent patterns of failure to control sexual impulses resulting in repetitive sexual behavior, pursued despite adverse consequences. Despite previous indications of addiction-like mechanisms and the recent impulse-control disorder classification in the International Classification of Diseases (ICD-11), the neurobiological processes underlying CSBD are unknown. Methods: We designed and applied a behavioral paradigm aimed at disentangling processes related to anticipation and viewing of erotic stimuli. In 22 male CSBD patients (age: M = 38.7, SD = 11.7) and 20 healthy male controls (HC, age: M = 37.6, SD = 8.5), we measured behavioral responses and neural activity during functional magnetic resonance imaging (fMRI). The main outcomes were response time differences between erotic and non-erotic trials and ventral striatum (VS) activity during anticipation of visual stimuli. We related these outcomes with each other, to CSBD diagnosis, and symptom severity. Results: We found robust case-control differences on behavioral level, where CSBD patients showed larger response time differences between erotic and non-erotic trials than HC. The task induced reliable main activations within each group. While we did not observe significant group differences in VS activity, VS activity during anticipation correlated with response time differences and self-ratings for anticipation of erotic stimuli. Discussion and Conclusions: Our results support the validity and applicability of the developed task and suggest that CSBD is associated with altered behavioral correlates of anticipation, which were associated with ventral striatum activity during anticipation of erotic stimuli. This supports the idea that addiction-like mechanisms play a role in CSBD.

Deep brain stimulation (DBS) has been used to modulate aberrant circuits associated with Parkinson's disease (PD) for decades and has shown robust therapeutic benefits. However, the mechanism of action of DBS remains incompletely understood. With technological advances, there is an emerging use of functional magnetic resonance imaging (fMRI) after DBS implantation to explore the effects of stimulation on brain networks in PD. This systematic review was designed following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines to summarize peer-reviewed articles published within the past 10 years in which fMRI was employed on patients with PD-DBS. Search in PubMed database provided 353 references, and screenings resulted in a total of 19 studies for qualitative synthesis regarding study designs (fMRI scan timepoints and paradigm), methodology, and PD subtypes. This review concluded that fMRI may be used in patients with PD-DBS after proper safety test; resting-state and block-based fMRI designs have been employed to explore the effects of DBS on brain networks and the mechanism of action of the DBS, respectively. With further validation of safety use of fMRI and advances in imaging techniques, fMRI may play an increasingly important role in better understanding of the mechanism of stimulation as well as in improving clinical care to provide subject-specific neuromodulation treatments.

In chapter 2, we assessed the acute effects of a single methylphenidate administration on amygdala reactivity during fear learning and fear reversal learning. As hypothesized, DBD-patients in the placebo condition (DBD-PCB) showed lower amygdala reactivity during the acquisition of fear compared to healthy controls (HCs) and to DBD patients in the MPH condition (DBD-MPH). These findings suggest a normalizing effect of MPH on amygdala functioning during the acquisition of fear in DBD. Contrary our hypothesis, at the group level there was no difference in reactivity between the groups in the vmPFC during fear reversal learning. In exploratory subgroup analyses it was found that amygdala hypoactivation and the subsequent normalizing effect of MPH seemed to be driven by a subgroup of DBD-patients without comorbid ADHD. In chapter 3, we assessed whether brain activity during the extinction of previously acquired fear was associated with the persistence or desistence of DBD and psychopathic traits. As hypothesized, DBD-P and DBD-D were both associated with increased reactivity of ACC and insula during fear extinction, compared to HCs. Furthermore, impulsive-irresponsible psychopathic traits were positive associated with hyperreactivity in the amygdala, insula and ACC. Moreover, these impulsive-irresponsible traits mediated the association of neural reactivity in the ACC and persisting and desisting of DBD. The acute effect of methylphenidate administration on reward and punishment sensitivity was investigated in chapter 4. We hypothesized that - compared to HCs – DBD-patients have deviancies in both reward and punishment processing and that a single dose of methylphenidate could (temporary) normalize these deviancies. DBD-PCB showed - compared to HCs - significantly lower amygdala responsiveness during reward receipt or omission. DBD-MPH showed, compared to HCs, similar amygdala responsiveness during omission of a reward suggesting normalization of amygdala reactivity dur