Search

Abstract Background The pathophysiology of Alzheimer’s disease (AD) involves $$\beta$$ β -amyloid (A $$\beta$$ β ) accumulation. Early identification of individuals with abnormal $$\beta$$ β -amyloid levels is crucial, but A $$\beta$$ β quantification with positron emission tomography (PET) and cerebrospinal fluid (CSF) is invasive and expensive. Methods We propose a machine learning framework using standard non-invasive (MRI, demographics, APOE, neuropsychology) measures to predict future A $$\beta$$ β -positivity in A $$\beta$$ β -negative individuals. We separately study A $$\beta$$ β -positivity defined by PET and CSF. Results Cross-validated AUC for 4-year A $$\beta$$ β conversion prediction was 0.78 for the CSF-based and 0.68 for the PET-based A $$\beta$$ β definitions. Although not trained for the clinical status-change prediction, the CSF-based model excelled in predicting future mild cognitive impairment (MCI)/dementia conversion in cognitively normal/MCI individuals (AUCs, respectively, 0.76 and 0.89 with a separate dataset). Conclusion Standard measures have potential in detecting future A $$\beta$$ β -positivity and assessing conversion risk, even in cognitively normal individuals. The CSF-based definition led to better predictions than the PET-based definition.

Abstract A system of lymphatic vessels has been recently characterized in the meninges, with a postulated role in ‘cleaning’ the brain via cerebral fluid drainage. As meninges are the origin site of migraine pain, we hypothesized that malfunctioning of the lymphatic system should affect the local trigeminal nociception. To test this hypothesis, we studied nociceptive and inflammatory mechanisms in the hemiskull preparations (containing the meninges) of K14-VEGFR3-Ig (K14) mice lacking the meningeal lymphatic system. We recorded the spiking activity of meningeal afferents and estimated the local mast cells population, calcitonin gene-related peptide (CGRP) and cytokine levels as well as the dural trigeminal innervation in freshly-isolated hemiskull preparations from K14-VEGFR3-Ig (K14) or wild type C57BL/6 mice (WT). Spiking activity data have been confirmed in an acquired model of meningeal lymphatic dysfunction (AAV-mVEGFR3(1–4)Ig induced lymphatic ablation). We found that levels of the pro-inflammatory cytokine IL12-p70 and CGRP, implicated in migraine, were reduced in the meninges of K14 mice, while the levels of the mast cell activator MCP-1 were increased. The other migraine-related pro-inflammatory cytokines (basal and stimulated), did not differ between the two genotypes. The patterns of trigeminal innervation in meninges remained unchanged and we did not observe alterations in basal or ATP-induced nociceptive firing in the meningeal afferents associated with meningeal lymphatic dysfunction. In summary, the lack of meningeal lymphatic system is associated with a new balance between pro- and anti-migraine mediators but does not directly trigger meningeal nociceptive state.

Predictive modeling studies have started to reveal brain measures underlying cognition; however, most studies are based on cross-sectional data (brain measures acquired at one time point). Since brain development comprises of continuously ongoing events leading to cognitive development, predictive modeling studies need to consider ‘longitudinal brain change’ as opposed to ‘cross-sectional brain measures’. Using longitudinal neuroimaging and cognitive data (global executive composite score, an index of executive function) from 82 individuals (aged 5–14 years, scanned 3 times), we built highly accurate prediction models (r = 0.61, p = 1.6e-09) of future cognition (assessed at visit 3) based on developmental changes in cortical anatomy (from visit 1 to 2). More importantly, longitudinal brain change (i.e. change in cortical anatomy from visit 1 to 2) and cross-sectional brain measures (cortical anatomy at visit 1 and 2) were critical for predicting future cognition, suggesting the need for considering longitudinal brain change in predicting cognitive outcomes.

Non-invasive magnetic resonance imaging (MRI) methods have proved useful in the diagnosis and prognosis of neurodegenerative diseases. However, the interpretation of imaging outcomes in terms of tissue pathology is still challenging. This study goes beyond the current interpretation of in vivo diffusion tensor imaging (DTI) by constructing multivariate models of quantitative tissue microstructure in status epilepticus (SE)-induced brain damage. We performed in vivo DTI and histology in rats at 79 days after SE and control animals. The analyses focused on the corpus callosum, hippocampal subfield CA3b, and layers V and VI of the parietal cortex. Comparison between control and SE rats indicated that a combination of microstructural tissue changes occurring after SE, such as cellularity, organization of myelinated axons, and/or morphology of astrocytes, affect DTI parameters. Subsequently, we constructed a multivariate regression model for explaining and predicting histological parameters based on DTI. The model revealed that DTI predicted well the organization of myelinated axons (cross-validated R = 0.876) and astrocyte processes (cross-validated R = 0.909) and possessed a predictive value for cell density (CD) (cross-validated R = 0.489). However, the morphology of astrocytes (cross-validated R > 0.05) was not well predicted. The inclusion of parameters from CA3b was necessary for modeling histopathology. Moreover, the multivariate DTI model explained better histological parameters than any univariate model. In conclusion, we demonstrate that combining several analytical and statistical tools can help interpret imaging outcomes to microstructural tissue changes, opening new avenues to improve the non-invasive diagnosis and prognosis of brain tissue damage.

With DeepACSON, Abdollahzadeh et al. combines existing deep learning-based methods for semantic segmentation and a novel shape decomposition technique for the instance segmentation. The pipeline is used to segment low-resolution 3D-EM datasets allowing quantification of white matter morphology in large fields-of-view.

We develop a cylindrical shape decomposition (CSD) algorithm to decompose an object, a union of several tubular structures, into its semantic components. We decompose the object using its curve skeleton and restricted translational sweeps. For that, CSD partitions the curve skeleton into maximal-length sub-skeletons over an orientation cost, each sub-skeleton corresponds to a semantic component. To find the intersection of the tubular components, CSD translationally sweeps the object in decomposition intervals to identify critical points at which the shape of the object changes substantially. CSD cuts the object at critical points and assigns the same label to parts along the same sub-skeleton, thereby constructing a semantic component. The proposed method further reconstructs the acquired semantic components at the intersection of object parts using generalized cylinders. We apply CSD for segmenting axons in large 3D electron microscopy images and decomposing vascular networks and synthetic objects. We show that our proposal is robust to severe surface noise and outperforms state-of-the-art decomposition techniques in its applications.

The Alzheimer’s Disease Assessment Scale-Cognitive subscale (ADAS-Cog) is a neuropsychological tool that has been designed to assess the severity of cognitive symptoms of dementia. Personalized prediction of the changes in ADAS-Cog scores could help in timing therapeutic interventions in dementia and at-risk populations. In the present work, we compared single and multitask learning approaches to predict the changes in ADAS-Cog scores based on T1-weighted anatomical magnetic resonance imaging (MRI). In contrast to most machine learning-based prediction methods ADAS-Cog changes, we stratified the subjects based on their baseline diagnoses and evaluated the prediction performances in each group. Our experiments indicated a positive relationship between the predicted and observed ADAS-Cog score changes in each diagnostic group, suggesting that T1-weighted MRI has a predictive value for evaluating cognitive decline in the entire AD continuum. We further studied whether correction of the differences in the magnetic field strength of MRI would improve the ADAS-Cog score prediction. The partial least square-based domain adaptation slightly improved the prediction performance, but the improvement was marginal. In summary, this study demonstrated that ADAS-Cog change could be, to some extent, predicted based on anatomical MRI. Based on this study, the recommended method for learning the predictive models is a single-task regularized linear regression due to its simplicity and good performance. It appears important to combine the training data across all subject groups for the most effective predictive models.

Registration-based methods are commonly used in the automatic segmentation of magnetic resonance (MR) brain images. However, these methods are not robust to the presence of gross pathologies that can alter the brain anatomy and affect the alignment of the atlas image with the target image. In this work, we develop a robust algorithm, MU-Net-R, for automatic segmentation of the normal and injured rat hippocampus based on an ensemble of U-net-like Convolutional Neural Networks (CNNs). MU-Net-R was trained on manually segmented MR images of sham-operated rats and rats with traumatic brain injury (TBI) by lateral fluid percussion. The performance of MU-Net-R was quantitatively compared with methods based on single and multi-atlas registration using MR images from two large preclinical cohorts. Automatic segmentations using MU-Net-R and multi-atlas registration were of excellent quality, achieving cross-validated Dice scores above 0.90 despite the presence of brain lesions, atrophy, and ventricular enlargement. In contrast, the performance of single-atlas segmentation was unsatisfactory (cross-validated Dice scores below 0.85). Interestingly, the registration-based methods were better at segmenting the contralateral than the ipsilateral hippocampus, whereas MU-Net-R segmented the contralateral and ipsilateral hippocampus equally well. We assessed the progression of hippocampal damage after TBI by using our automatic segmentation tool. Our data show that the presence of TBI, time after TBI, and whether the hippocampus was ipsilateral or contralateral to the injury were the parameters that explained hippocampal volume.

Our study investigates the potential of diffusion MRI (dMRI), including diffusion tensor imaging (DTI), fixel-based analysis (FBA) and neurite orientation dispersion and density imaging (NODDI), to detect microstructural tissue abnormalities in rats after mild traumatic brain injury (mTBI). The brains of sham-operated and mTBI rats 35 days after lateral fluid percussion injury were imaged ex vivo in a 11.7-T scanner. Voxel-based analyses of DTI-, fixel- and NODDI-based metrics detected extensive tissue changes in directly affected brain areas close to the primary injury, and more importantly, also in distal areas connected to primary injury and indirectly affected by the secondary injury mechanisms. Histology revealed ongoing axonal abnormalities and inflammation, 35 days after the injury, in the brain areas highlighted in the group analyses. Fractional anisotropy (FA), fiber density (FD) and fiber density and fiber bundle cross-section (FDC) showed similar pattern of significant areas throughout the brain; however, FA showed more significant voxels in gray matter areas, while FD and FDC in white matter areas, and orientation dispersion index (ODI) in areas most damage based on histology. Region-of-interest (ROI)-based analyses on dMRI maps and histology in selected brain regions revealed that the changes in MRI parameters could be attributed to both alterations in myelinated fiber bundles and increased cellularity. This study demonstrates that the combination of dMRI methods can provide a more complete insight into the microstructural alterations in white and gray matter after mTBI, which may aid diagnosis and prognosis following a mild brain injury.

Accurate detection and quantification of unruptured intracranial aneurysms (UIAs) is important for rupture risk assessment and to allow an informed treatment decision to be made. Currently, 2D manual measures used to assess UIAs on Time-of-Flight magnetic resonance angiographies (TOF-MRAs) lack 3D information and there is substantial inter-observer variability for both aneurysm detection and assessment of aneurysm size and growth. 3D measures could be helpful to improve aneurysm detection and quantification but are time-consuming and would therefore benefit from a reliable automatic UIA detection and segmentation method. The Aneurysm Detection and segMentation (ADAM) challenge was organised in which methods for automatic UIA detection and segmentation were developed and submitted to be evaluated on a diverse clinical TOF-MRA dataset.A training set (113 cases with a total of 129 UIAs) was released, each case including a TOF-MRA, a structural MR image (T1, T2 or FLAIR), annotation of any present UIA(s) and the centre voxel of the UIA(s). A test set of 141 cases (with 153 UIAs) was used for evaluation. Two tasks were proposed: (1) detection and (2) segmentation of UIAs on TOF-MRAs. Teams developed and submitted containerised methods to be evaluated on the test set. Task 1 was evaluated using metrics of sensitivity and false positive count. Task 2 was evaluated using dice similarity coefficient, modified hausdorff distance (95th percentile) and volumetric similarity. For each task, a ranking was made based on the average of the metrics.In total, eleven teams participated in task 1 and nine of those teams participated in task 2. Task 1 was won by a method specifically designed for the detection task (i.e. not participating in task 2). Based on segmentation metrics, the top two methods for task 2 performed statistically significantly better than all other methods. The detection performance of the top-ranking methods was comparable to visual inspection for larger aneurysms. Segmentation performance of the top ranking method, after selection of true UIAs, was similar to interobserver performance. The ADAM challenge remains open for future submissions and improved submissions, with a live leaderboard to provide benchmarking for method developments at https://adam.isi.uu.nl/.

It is necessary to develop reliable biomarkers for epileptogenesis and cognitive impairment after traumatic brain injury when searching for novel antiepileptogenic and cognition-enhancing treatments. We hypothesized that a multiparametric magnetic resonance imaging (MRI) analysis along the septotemporal hippocampal axis could predict the development of post-traumatic epilepsy and cognitive impairment. We performed quantitative T2 and T2* MRIs at 2, 7 and 21 days, and diffusion tensor imaging at 7 and 21 days after lateral fluid-percussion injury in male rats. Morris water maze tests conducted between 35–39 days post-injury were used to diagnose cognitive impairment. One-month-long continuous video-electroencephalography monitoring during the 6th post-injury month was used to diagnose epilepsy. Single-parameter and regularized multiple linear regression models were able to differentiate between sham-operated and brain-injured rats. In the ipsilateral hippocampus, differentiation between the groups was achieved at most septotemporal locations (cross-validated area under the receiver operating characteristic curve (AUC) 1.0, 95% confidence interval 1.0–1.0). In the contralateral hippocampus, the highest differentiation was evident in the septal pole (AUC 0.92, 95% confidence interval 0.82–0.97). Logistic regression analysis of parameters imaged at 3.4 mm from the contralateral hippocampus’s temporal end differentiated between the cognitively impaired rats and normal rats (AUC 0.72, 95% confidence interval 0.55–0.84). Neither single nor multiparametric approaches could identify the rats that would develop post-traumatic epilepsy. Multiparametric MRI analysis of the hippocampus can be used to identify cognitive impairment after an experimental traumatic brain injury. This information can be used to select subjects for preclinical trials of cognition-improving interventions.

Noise due to genetic heterogeneity potentially impacts the the discovery of genes that contribute to diseases such as schizophrenia (SCZ). In this study, authors minimize the disease-irrelevant noise between SCZ and healthy individuals by profiling transcriptional signatures among discordant monozygotic twin pairs, and demonstrate that although sexes share many of the final common pathways, the underlying primary pathophysiology of SCZ differs between males and females.

Skull-stripping and region segmentation are fundamental steps in preclinical magnetic resonance imaging (MRI) studies, and these common procedures are usually performed manually. We present Multi-task U-Net (MU-Net), a convolutional neural network designed to accomplish both tasks simultaneously. MU-Net achieved higher segmentation accuracy than state-of-the-art multi-atlas segmentation methods with an inference time of 0.35 s and no pre-processing requirements.We trained and validated MU-Net on 128 T2-weighted mouse MRI volumes as well as on the publicly available MRM NeAT dataset of 10 MRI volumes. We tested MU-Net with an unusually large dataset combining several independent studies consisting of 1782 mouse brain MRI volumes of both healthy and Huntington animals, and measured average Dice scores of 0.906 (striati), 0.937 (cortex), and 0.978 (brain mask). Further, we explored the effectiveness of our network in the presence of different architectural features, including skip connections and recently proposed framing connections, and the effects of the age range of the training set animals.These high evaluation scores demonstrate that MU-Net is a powerful tool for segmentation and skull-stripping, decreasing inter and intra-rater variability of manual segmentation. The MU-Net code and the trained model are publicly available at https://github.com/Hierakonpolis/MU-Net.

This video-based study examines the pragmatic non-verbal comprehension skills and corresponding neural-level findings in young Finnish autistic adults, and controls. Items from the Assessment Battery of Communication (ABaCo) were chosen to evaluate the comprehension of non-verbal communication. Inter-subject correlation (ISC) analysis of the functional magnetic resonance imaging data was used to reveal the synchrony of brain activation across participants during the viewing of pragmatically complex scenes of ABaCo videos. The results showed a significant difference between the ISC maps of the autistic and control groups in tasks involving the comprehension of non-verbal communication, thereby revealing several brain regions where correlation of brain activity was greater within the control group. The results suggest a possible weaker modulation of brain states in response to the pragmatic non-verbal communicative situations in autistic participants. Although there was no difference between the groups in behavioural responses to ABaCo items, there was more variability in the accuracy of the responses in the autistic group. Furthermore, mean answering and reaction times correlated with the severity of autistic traits. The results indicate that even if young autistic adults may have learned to use compensatory resources in their communicative-pragmatic comprehension, pragmatic processing in naturalistic situations still requires additional effort.

We present a fully convolutional neural network (ConvNet), named RatLesNetv2, for segmenting lesions in rodent magnetic resonance (MR) brain images. RatLesNetv2 architecture resembles an autoencoder and it incorporates residual blocks that facilitate its optimization. RatLesNetv2 is trained end to end on three-dimensional images and it requires no preprocessing. We evaluated RatLesNetv2 on an exceptionally large dataset composed of 916 T2-weighted rat brain MRI scans of 671 rats at nine different lesion stages that were used to study focal cerebral ischemia for drug development. In addition, we compared its performance with three other ConvNets specifically designed for medical image segmentation. RatLesNetv2 obtained similar to higher Dice coefficient values than the other ConvNets and it produced much more realistic and compact segmentations with notably fewer holes and lower Hausdorff distance. The Dice scores of RatLesNetv2 segmentations also exceeded inter-rater agreement of manual segmentations. In conclusion, RatLesNetv2 could be used for automated lesion segmentation, reducing human workload and improving reproducibility. RatLesNetv2 is publicly available at https://github.com/jmlipman/RatLesNetv2.

Hexanucleotide repeat expansion (HRE) in the chromosome 9 open-reading frame 72 (C9orf72) gene is the most common genetic cause underpinning frontotemporal lobar degeneration (FTLD) and amyotrophic lateral sclerosis (ALS). It leads to the accumulation of toxic RNA foci and various dipeptide repeat (DPR) proteins into cells. These C9orf72 HRE-specific hallmarks are abundant in neurons. So far, the role of microglia, the immune cells of the brain, in C9orf72 HRE-associated FTLD/ALS is unclear. In this study, we overexpressed C9orf72 HRE of a pathological length in the BV-2 microglial cell line and used biochemical methods and fluorescence imaging to investigate its effects on their phenotype, viability, and functionality. We found that BV-2 cells expressing the C9orf72 HRE presented strong expression of specific DPR proteins but no sense RNA foci. Transiently increased levels of cytoplasmic TAR DNA-binding protein 43 (TDP-43), slightly altered levels of p62 and lysosome-associated membrane protein (LAMP) 2A, and reduced levels of polyubiquitinylated proteins, but no signs of cell death were detected in HRE overexpressing cells. Overexpression of the C9orf72 HRE did not affect BV-2 cell phagocytic activity or response to an inflammatory stimulus, nor did it shift their RNA profile toward disease-associated microglia. These findings suggest that DPR proteins do not affect microglial cell viability or functionality in BV-2 cells. However, additional studies in other models are required to further elucidate the role of C9orf72 HRE in microglia.

Individuals with attention-deficit/hyperactivity disorder (ADHD) have difficulties navigating dynamic everyday situations that contain multiple sensory inputs that need to either be attended to or ignored. As conventional experimental tasks lack this type of everyday complexity, we administered a film-based multi-talker condition with auditory distractors in the background. ADHD-related aberrant brain responses to this naturalistic stimulus were identified using intersubject correlations (ISCs) in functional magnetic resonance imaging (fMRI) data collected from 51 adults with ADHD and 29 healthy controls. A novel permutation-based approach introducing studentized statistics and subject-wise voxel-level null-distributions revealed that several areas in cerebral attention networks and sensory cortices were desynchronized in participants with ADHD (n = 20) relative to healthy controls (n = 20). Specifically, desynchronization of the posterior parietal cortex occurred when irrelevant speech or music was presented in the background, but not when irrelevant white noise was presented, or when there were no distractors. We also show regionally distinct ISC signatures for inattention and impulsivity. Finally, post-scan recall of the film contents was associated with stronger ISCs in the default-mode network for the ADHD and in the dorsal attention network for healthy controls. The present study shows that ISCs can further our understanding of how a complex environment influences brain states in ADHD.

Non-invasive imaging methods have become essential tools for understanding the central nervous system (CNS) in health and disease. In particular, magnetic resonance imaging (MRI) techniques provide information about the anatomy, microstructure, and function of the brain and spinal cord in vivo non-invasively. However, MRI is limited by its spatial resolution and signal specificity. In order to mitigate these shortcomings, it is crucial to validate MRI with an array of ancillary ex vivo imaging techniques. These techniques include histological methods, such as light and electron microscopy (EM), which can provide specific information on the tissue structure in healthy and diseased brain and spinal cord, at cellular and subcellular level. However, these conventional histological techniques are intrinsically two-dimensional (2D) and, as a result of sectioning, lack volumetric information of the tissue. This limitation can be overcome with genuine three-dimensional (3D) imaging approaches of the tissue. 3D highly resolved information of the CNS achievable by means of other imaging techniques can complement and improve the interpretation of MRI measurements. In this article, we provide an overview of different 3D imaging techniques that can be used to validate MRI. As an example, we introduce an approach of how to combine diffusion MRI and synchrotron X-ray phase contrast tomography (SXRPCT) data. Our approach paves the way for a new multiscale assessment of the CNS allowing to validate and to improve our understanding of in vivo imaging (such as MRI).

A hierarchical clustering algorithm was applied to magnetic resonance images (MRI) of a cohort of 751 subjects having a mild cognitive impairment (MCI), 282 subjects having received Alzheimer’s disease (AD) diagnosis, and 428 normal controls (NC). MRIs were preprocessed to gray matter density maps and registered to a stereotactic space. By first rendering the gray matter density maps comparable by regressing out age, gender, and years of education, and then performing the hierarchical clustering, we found clusters displaying structural features of typical AD, cortically-driven atypical AD, limbic-predominant AD, and early-onset AD (EOAD). Among these clusters, EOAD subjects displayed marked cortical gray matter atrophy and atrophy of the precuneus. Furthermore, EOAD subjects had the highest progression rates as measured with ADAS slopes during the longitudinal follow-up of 36 months. Striking heterogeneities in brain atrophy patterns were observed with MCI subjects. We found clusters of stable MCI, clusters of diffuse brain atrophy with fast progression, and MCI subjects displaying similar atrophy patterns as the typical or atypical AD subjects. Bidirectional differences in structural phenotypes were found with MCI subjects involving the anterior cerebellum and the frontal cortex. The diversity of the MCI subjects suggests that the structural phenotypes of MCI subjects would deserve a more detailed investigation with a significantly larger cohort. Our results demonstrate that the hierarchical agglomerative clustering method is an efficient tool in dividing a cohort of subjects with gray matter atrophy into coherent clusters manifesting different structural phenotypes.

Rey's Auditory Verbal Learning Test (RAVLT) is a powerful neuropsychological tool for testing episodic memory, which is widely used for the cognitive assessment in dementia and pre-dementia conditions. Several studies have shown that an impairment in RAVLT scores reflect well the underlying pathology caused by Alzheimer's disease (AD), thus making RAVLT an effective early marker to detect AD in persons with memory complaints. We investigated the association between RAVLT scores (RAVLT Immediate and RAVLT Percent Forgetting) and the structural brain atrophy caused by AD. The aim was to comprehensively study to what extent the RAVLT scores are predictable based on structural magnetic resonance imaging (MRI) data using machine learning approaches as well as to find the most important brain regions for the estimation of RAVLT scores. For this, we built a predictive model to estimate RAVLT scores from gray matter density via elastic net penalized linear regression model. The proposed approach provided highly significant cross-validated correlation between the estimated and observed RAVLT Immediate (R = 0.50) and RAVLT Percent Forgetting (R = 0.43) in a dataset consisting of 806 AD, mild cognitive impairment (MCI) or healthy subjects. In addition, the selected machine learning method provided more accurate estimates of RAVLT scores than the relevance vector regression used earlier for the estimation of RAVLT based on MRI data. The top predictors were medial temporal lobe structures and amygdala for the estimation of RAVLT Immediate and angular gyrus, hippocampus and amygdala for the estimation of RAVLT Percent Forgetting. Further, the conversion of MCI subjects to AD in 3-years could be predicted based on either observed or estimated RAVLT scores with an accuracy comparable to MRI-based biomarkers.

(1) Background: Transfer learning refers to machine learning techniques that focus on acquiring knowledge from related tasks to improve generalization in the tasks of interest. In magnetic resonance imaging (MRI), transfer learning is important for developing strategies that address the variation in MR images from different imaging protocols or scanners. Additionally, transfer learning is beneficial for reutilizing machine learning models that were trained to solve different (but related) tasks to the task of interest. The aim of this review is to identify research directions, gaps in knowledge, applications, and widely used strategies among the transfer learning approaches applied in MR brain imaging; (2) Methods: We performed a systematic literature search for articles that applied transfer learning to MR brain imaging tasks. We screened 433 studies for their relevance, and we categorized and extracted relevant information, including task type, application, availability of labels, and machine learning methods. Furthermore, we closely examined brain MRI-specific transfer learning approaches and other methods that tackled issues relevant to medical imaging, including privacy, unseen target domains, and unlabeled data; (3) Results: We found 129 articles that applied transfer learning to MR brain imaging tasks. The most frequent applications were dementia-related classification tasks and brain tumor segmentation. The majority of articles utilized transfer learning techniques based on convolutional neural networks (CNNs). Only a few approaches utilized clearly brain MRI-specific methodology, and considered privacy issues, unseen target domains, or unlabeled data. We proposed a new categorization to group specific, widely-used approaches such as pretraining and fine-tuning CNNs; (4) Discussion: There is increasing interest in transfer learning for brain MRI. Well-known public datasets have clearly contributed to the popularity of Alzheimer’s diagnostics/prognostics and tumor segmentation as applications. Likewise, the availability of pretrained CNNs has promoted their utilization. Finally, the majority of the surveyed studies did not examine in detail the interpretation of their strategies after applying transfer learning, and did not compare their approach with other transfer learning approaches.

We developed a pipeline for the discovery of transcriptomics-derived disease-modifying therapies and used it to validate treatments in vitro and in vivo that could be repurposed for TBI treatment. Desmethylclomipramine, ionomycin, sirolimus and trimipramine, identified by in silico LINCS analysis as candidate treatments modulating the TBI-induced transcriptomics networks, were tested in neuron-BV2 microglial co-cultures, using tumour necrosis factor α as a monitoring biomarker for neuroinflammation, nitrite for nitric oxide-mediated neurotoxicity and microtubule associated protein 2-based immunostaining for neuronal survival. Based on (a) therapeutic time window in silico, (b) blood-brain barrier penetration and water solubility, (c) anti-inflammatory and neuroprotective effects in vitro (p < 0.05) and (d) target engagement of Nrf2 target genes (p < 0.05), desmethylclomipramine was validated in a lateral fluid-percussion model of TBI in rats. Despite the favourable in silico and in vitro outcomes, in vivo assessment of clomipramine, which metabolizes to desmethylclomipramine, failed to demonstrate favourable effects on motor and memory tests. In fact, clomipramine treatment worsened the composite neuroscore (p < 0.05). Weight loss (p < 0.05) and prolonged upregulation of plasma cytokines (p < 0.05) may have contributed to the worsened somatomotor outcome. Our pipeline provides a rational stepwise procedure for evaluating favourable and unfavourable effects of systems-biology discovered compounds that modulate post-TBI transcriptomics.

Deformable models are able to solve surface extraction problems challenged by image noise because imageindependent constraints are used to regularize the shape of the extracted surface. However, this ability of deformable models is shadowed by their application specificity, initialization sensitivity and the difficulty of the selection of proper values for user definable parameters. To overcome these problems restricting the automation of surface extraction, we present a new algorithm, named AdaCoP, for the global minimization of the energy of deformable surfaces. It iteratively performs constrained local minimizations of the energy. It avoids the detection of the same local minimum multiple times by constraining the local optimizations in an adaptive manner. AdaCoP escapes from local minima by imposing an adaptive penalty energy to it. These constraints and penalties prevent the convergence to the local minima already found. The performance of the AdaCoP algorithm is relatively independent on the nature of the underlying image as well as the shape of the surface to be extracted. The performance of the algorithm is evaluated by extracting surfaces from synthetic images. Moreover, the good properties of the algorithm are demonstrated by considering applications within the automated analysis of positron emission tomography images. Although AdaCoP cannot be proven to converge to the global minimum, it is insensitive to its initialization and it therefore provides a way to automate surface extraction problems within medical image analysis.

Within functional magnetic resonance imaging (fMRI), the use of the traditional general linear model (GLM) based analysis methods is often restricted to strictly controlled research setups requiring a parametric activation model. Instead, Inter-Subject Correlation (ISC) method is based on voxel-wise correlation between the time series of the subjects, which makes it completely non-parametric and thus suitable for naturalistic stimulus paradigms such as movie watching. In this study, we compared an ISC based analysis results with those of a GLM based in five distinct controlled research setups. We used International Consortium for Brain Mapping functional reference battery (FRB) fMRI data available from the Laboratory of Neuro Imaging image data archive. The selected data included measurements from 37 right-handed subjects, who all had performed the same five tasks from FRB. The GLM was expected to locate activations accurately in FRB data and thus provide good grounds for investigating relationship between ISC and stimulus induced fMRI activation. The statistical maps of ISC and GLM were compared with two measures. The first measure was the Pearson's correlation between the non-thresholded ISC test-statistics and absolute values of the GLM Z-statistics. The average correlation value over five tasks was 0.74. The second was the Dice index between the activation regions of the methods. The average Dice value over the tasks and three threshold levels was 0.73. The results of this study indicated how the data driven ISC analysis found the same foci as the model-based GLM analysis. The agreement of the results is highly interesting, because ISC is applicable in situations where GLM is not suitable, for example, when analyzing data from a naturalistic stimuli experiment.

Cinema is a promising naturalistic stimulus that enables, for instance, elicitation of robust emotions during functional magnetic resonance imaging (fMRI). Inter-subject correlation (ISC) has been used as a model-free analysis method to map the highly complex hemodynamic responses that are evoked during watching a movie. Here, we extended the ISC analysis to frequency domain using wavelet analysis combined with non-parametric permutation methods for making voxel-wise statistical inferences about frequency-band specific ISC. We applied these novel analysis methods to a dataset collected in our previous study where 12 subjects watched an emotionally engaging movie “Crash” during fMRI scanning. Our results suggest that several regions within the frontal and temporal lobes show ISC predominantly at low frequency bands, whereas visual cortical areas exhibit ISC also at higher frequencies. It is possible that these findings relate to recent observations of a cortical hierarchy of temporal receptive windows, or that the types of events processed in temporal and prefrontal cortical areas (e.g., social interactions) occur over longer time periods than the stimulus features processed in the visual areas. Software tools to perform frequency-specific ISC analysis, together with a visualization application, are available as open source Matlab code.

We propose and evaluate an automatic segmentation method for extracting striatal brain structures (caudate, putamen, and ventral striatum) from parametric C11-raclopride positron emission tomography (PET) brain images. We focus on the images acquired using a novel brain dedicated high-resolution (HRRT) PET scanner. The segmentation method first extracts the striatum using a deformable surface model and then divides the striatum into its substructures based on a graph partitioning algorithm. The weighted kernel k-means algorithm is used to partition the graph describing the voxel affinities within the striatum into the desired number of clusters. The method was experimentally validated with synthetic and real image data. The experiments showed that our method was able to automatically extract caudate, ventral striatum, and putamen from the images. Moreover, the putamen could be subdivided into anterior and posterior parts. An automatic method for the extraction of striatal structures from high-resolution PET images allows for inexpensive and reproducible extraction of the quantitative information from these images necessary in brain research and drug development.