Artificial intelligence is one of the most exciting methodological shifts in our era. It holds the potential to transform healthcare as we know it, to a system where humans and machines work together to provide better treatment for our patients. It is now clear that cutting edge artificial intelligence models in conjunction with high-quality clinical data will lead to improved prognostic and diagnostic models in neurological disease, facilitating expert-level clinical decision tools across healthcare settings. Despite the clinical promise of artificial intelligence, machine and deep-learning algorithms are not a one-size-fits-all solution for all types of clinical data and questions. In this article, we provide an overview of the core concepts of artificial intelligence, particularly contemporary deep-learning methods, to give clinician and neuroscience researchers an appreciation of how artificial intelligence can be harnessed to support clinical decisions. We clarify and emphasize the data quality and the human expertise needed to build robust clinical artificial intelligence models in neurology. As artificial intelligence is a rapidly evolving field, we take the opportunity to iterate important ethical principles to guide the field of medicine is it moves into an artificial intelligence enhanced future.

OBJECTIVE: The aim of this study was to test the hypothesis that white matter degeneration of the perforant path - as part of the Papez circuit - is a key feature of amyotrophic lateral sclerosis (ALS), even in the absence of frontotemporal dementia (FTD) or deposition of pTDP-43 inclusions in hippocampal granule cells. METHODS: We used diffusion Magnetic Resonance Imaging (dMRI), polarized light imaging (PLI) and immunohistochemical analysis of post mortem hippocampus specimens from controls (n = 5) and ALS patients (n = 14) to study white matter degeneration in the perforant path. RESULTS: diffusion Magnetic Resonance Imaging demonstrated a decrease in fractional anisotropy (P = 0.01) and an increase in mean diffusivity (P = 0.01) in the perforant path in ALS compared to controls. PLI-myelin density was lower in ALS (P = 0.05) and correlated with fractional anisotropy (r = 0.52, P = 0.03). These results were confirmed by immunohistochemistry; both myelin (proteolipid protein, P = 0.03) and neurofilaments (SMI-312, P = 0.02) were lower in ALS. Two out of the fourteen ALS cases showed pTDP-43 pathology in the dentate gyrus, but with comparable myelination levels in the perforant path to other ALS cases. CONCLUSION: We conclude that degeneration of the perforant path occurs in ALS patients and that this may occur before, or independent of, pTDP-43 aggregation in the dentate gyrus of the hippocampus. Future research should focus on correlating the degree of cognitive decline to the amount of white matter atrophy in the perforant path.

Cannabis use is highly prevalent and often considered to be relatively harmless. Nonetheless, a subset of regular cannabis users may develop dependence, experiencing poorer quality of life and greater mental health problems relative to non-dependent users. The neuroanatomy characterizing cannabis use versus dependence is poorly understood. We aimed to delineate the contributing role of cannabis use and dependence on morphology of the hippocampus, one of the most consistently altered brain regions in cannabis users, in a large multi-site dataset aggregated across four research sites. We compared hippocampal volume and vertex-level hippocampal shape differences (1) between 121 non-using controls and 140 cannabis users; (2) between 106 controls, 50 non-dependent users and 70 dependent users; and (3) between a subset of 41 controls, 41 non-dependent users and 41 dependent users, matched on sample characteristics and cannabis use pattern (onset age and dosage). Cannabis users did not differ from controls in hippocampal volume or shape. However, cannabis-dependent users had significantly smaller right and left hippocampi relative to controls and non-dependent users, irrespective of cannabis dosage. Shape analysis indicated localized deflations in the superior-medial body of the hippocampus. Our findings support neuroscientific theories postulating dependence-specific neuroadaptations in cannabis users. Future efforts should uncover the neurobiological risk and liabilities separating dependent and non-dependent use of cannabis.

The infant brain is unlike the adult brain, with considerable differences in morphological, neurodynamic, and haemodynamic features. As the majority of current MRI analysis tools were designed for use in adults, a primary objective of the Developing Human Connectome Project (dHCP) is to develop optimised methodological pipelines for the analysis of neonatal structural, resting state, and diffusion MRI data. Here, in an independent neonatal dataset we have extended and optimised the dHCP fMRI preprocessing pipeline for the analysis of stimulus-response fMRI data. We describe and validate this extended dHCP fMRI preprocessing pipeline to analyse changes in brain activity evoked following an acute noxious stimulus applied to the infant's foot. We compare the results obtained from this extended dHCP pipeline to results obtained from a typical FSL FEAT-based analysis pipeline, evaluating the pipelines' outputs using a wide range of tests. We demonstrate that a substantial increase in spatial specificity and sensitivity to signal can be attained with a bespoke neonatal preprocessing pipeline through optimised motion and distortion correction, ICA-based denoising, and haemodynamic modelling. The improved sensitivity and specificity, made possible with this extended dHCP pipeline, will be paramount in making further progress in our understanding of the development of sensory processing in the infant brain.

White matter hyperintensities (WMH), also known as white matter lesions, are localised white matter areas that appear hyperintense on MRI scans. WMH commonly occur in the ageing population, and are often associated with several factors such as cognitive disorders, cardiovascular risk factors, cerebrovascular and neurodegenerative diseases. Despite the fact that some links between lesion location and parametric factors such as age have already been established, the relationship between voxel-wise spatial distribution of lesions and these factors is not yet well understood. Hence, it would be of clinical importance to model the distribution of lesions at the population-level and quantitatively analyse the effect of various factors on the lesion distribution model. In this work we compare various methods, including our proposed method, to generate voxel-wise distributions of WMH within a population with respect to various factors. Our proposed Bayesian spline method models the spatio-temporal distribution of WMH with respect to a parametric factor of interest, in this case age, within a population. Our probabilistic model takes as input the lesion segmentation binary maps of subjects belonging to various age groups and provides a population-level parametric lesion probability map as output. We used a spline representation to ensure a degree of smoothness in space and the dimension associated with the parameter, and formulated our model using a Bayesian framework. We tested our algorithm output on simulated data and compared our results with those obtained using various existing methods with different levels of algorithmic and computational complexity. We then compared the better performing methods on a real dataset, consisting of 1000 subjects of the UK Biobank, divided in two groups based on hypertension diagnosis. Finally, we applied our method on a clinical dataset of patients with vascular disease. On simulated dataset, the results from our algorithm showed a mean square error (MSE) value of 7.27×10-5, which was lower than the MSE value reported in the literature, with the advantage of being robust and computationally efficient. In the UK Biobank data, we found that the lesion probabilities are higher for the hypertension group compared to the non-hypertension group and further verified this finding using a statistical t-test. Finally, when applying our method on patients with vascular disease, we observed that the overall probability of lesions is significantly higher in later age groups, which is in line with the current literature.

OBJECTIVES: The aim of this study was to describe the development of morphologic and diffusion tensor imaging sequences of peripheral nerves at 7 T, using carpal tunnel syndrome (CTS) as a model system of focal nerve injury. MATERIALS AND METHODS: Morphologic images were acquired at 7 T using a balanced steady-state free precession sequence. Diffusion tensor imaging was performed using single-shot echo-planar imaging and readout-segmented echo-planar imaging sequences. Different acquisition and postprocessing methods were compared to describe the optimal analysis pipeline. Magnetic resonance imaging parameters including cross-sectional areas, signal intensity, fractional anisotropy (FA), as well as mean, axial, and radial diffusivity were compared between patients with CTS (n = 8) and healthy controls (n = 6) using analyses of covariance corrected for age (significance set at P < 0.05). Pearson correlations with Bonferroni correction were used to determine association of magnetic resonance imaging parameters with clinical measures (significance set at P < 0.01). RESULTS: The 7 T acquisitions with high in-plane resolution (0.2 × 0.2mm) afforded detailed morphologic resolution of peripheral nerve fascicles. For diffusion tensor imaging, single-shot echo-planar imaging was more efficient than readout-segmented echo-planar imaging in terms of signal-to-noise ratio per unit scan time. Distortion artifacts were pronounced, but could be corrected during postprocessing. Registration of FA maps to the morphologic images was successful. The developed imaging and analysis pipeline identified lower median nerve FA (pisiform bone, 0.37 [SD 0.10]) and higher radial diffusivity (1.08 [0.20]) in patients with CTS compared with healthy controls (0.53 [0.06] and 0.78 [0.11], respectively, P < 0.047). Fractional anisotropy and radial diffusivity strongly correlated with patients' symptoms (r = -0.866 and 0.866, respectively, P = 0.005). CONCLUSIONS: Our data demonstrate the feasibility of morphologic and diffusion peripheral nerve imaging at 7 T. Fractional anisotropy and radial diffusivity were found to be correlates of symptom severity.

Stuttering is a disorder in which the smooth flow of speech is interrupted. People who stutter show structural and functional abnormalities in the speech and motor system. It is unclear whether functional differences reflect general traits of the disorder or are specifically related to the dysfluent speech state. We used a hierarchical approach to separate state and trait effects within stuttering. We collected sparse-sampled functional MRI during two overt speech tasks (sentence reading and picture description) in 17 people who stutter and 16 fluent controls. Separate analyses identified indicators of: (1) general traits of people who stutter; (2) frequency of dysfluent speech states in subgroups of people who stutter; and (3) the differences between fluent and dysfluent states in people who stutter. We found that reduced activation of left auditory cortex, inferior frontal cortex bilaterally, and medial cerebellum were general traits that distinguished fluent speech in people who stutter from that of controls. The stuttering subgroup with higher frequency of dysfluent states during scanning (n = 9) had reduced activation in the right subcortical grey matter, left temporo-occipital cortex, the cingulate cortex, and medial parieto-occipital cortex relative to the subgroup who were more fluent (n = 8). Finally, during dysfluent states relative to fluent ones, there was greater activation of inferior frontal and premotor cortex extending into the frontal operculum, bilaterally. The above differences were seen across both tasks. Subcortical state effects differed according to the task. Overall, our data emphasise the independence of trait and state effects in stuttering.

The allostatic load index quantifies the cumulative multisystem physiological response to chronic everyday stress, and includes cardiovascular, metabolic and inflammatory measures. Despite its central role in the stress response, research of the effect of allostatic load on the ageing brain has been limited. We investigated the relation of mid-life allostatic load index and multifactorial predictors of stroke (Framingham stroke risk) and diabetes (metabolic syndrome) with voxelwise structural grey and white matter brain integrity measures in the ageing Whitehall II cohort (N = 349, mean age = 69.6 (SD 5.2) years, N (male) = 281 (80.5%), mean follow-up before scan = 21.4 (SD 0.82) years). Higher levels of all three markers were significantly associated with lower grey matter density. Only higher Framingham stroke risk was significantly associated with lower white matter integrity (low fractional anisotropy and high mean diffusivity). Our findings provide some empirical support for the concept of allostatic load, linking the effect of everyday stress on the body with features of the ageing human brain.

BACKGROUND: Amyotrophic lateral sclerosis (ALS) is a clinically and histopathologically heterogeneous neurodegenerative disorder, in which therapy is hindered by the rapid progression of disease and lack of biomarkers. Magnetic resonance imaging (MRI) has demonstrated its potential for detecting the pathological signature and tracking disease progression in ALS. However, the microstructural and molecular pathological substrate is poorly understood and generally defined histologically. One route to understanding and validating the pathophysiological correlates of MRI signal changes in ALS is to directly compare MRI to histology in post mortem human brains. RESULTS: The article delineates a universal whole brain sampling strategy of pathologically relevant grey matter (cortical and subcortical) and white matter tracts of interest suitable for histological evaluation and direct correlation with MRI. A standardised systematic sampling strategy that was compatible with co-registration of images across modalities was established for regions representing phosphorylated 43-kDa TAR DNA-binding protein (pTDP-43) patterns that were topographically recognisable with defined neuroanatomical landmarks. Moreover, tractography-guided sampling facilitated accurate delineation of white matter tracts of interest. A digital photography pipeline at various stages of sampling and histological processing was established to account for structural deformations that might impact alignment and registration of histological images to MRI volumes. Combined with quantitative digital histology image analysis, the proposed sampling strategy is suitable for routine implementation in a high-throughput manner for acquisition of large-scale histology datasets. Proof of concept was determined in the spinal cord of an ALS patient where multiple MRI modalities (T1, T2, FA and MD) demonstrated sensitivity to axonal degeneration and associated heightened inflammatory changes in the lateral corticospinal tract. Furthermore, qualitative comparison of R2* and susceptibility maps in the motor cortex of 2 ALS patients demonstrated varying degrees of hyperintense signal changes compared to a control. Upon histological evaluation of the same region, intensity of signal changes in both modalities appeared to correspond primarily to the degree of microglial activation. CONCLUSION: The proposed post mortem whole brain sampling methodology enables the accurate intraindividual study of pathological propagation and comparison with quantitative MRI data, to more fully understand the relationship of imaging signal changes with underlying pathophysiology in ALS.

Purpose The objective of this systematic review was to synthesize evidence on the effectiveness of workplace-based return-to-work (RTW) interventions and work disability management (DM) interventions that assist workers with musculoskeletal (MSK) and pain-related conditions and mental health (MH) conditions with RTW. Methods We followed a systematic review process developed by the Institute for Work & Health and an adapted best evidence synthesis that ranked evidence as strong, moderate, limited, or insufficient. Results Seven electronic databases were searched from January 1990 until April 2015, yielding 8898 non-duplicate references. Evidence from 36 medium and high quality studies were synthesized on 12 different intervention categories across three broad domains: health-focused, service coordination, and work modification interventions. There was strong evidence that duration away from work from both MSK or pain-related conditions and MH conditions were significantly reduced by multi-domain interventions encompassing at least two of the three domains. There was moderate evidence that these multi-domain interventions had a positive impact on cost outcomes. There was strong evidence that cognitive behavioural therapy interventions that do not also include workplace modifications or service coordination components are not effective in helping workers with MH conditions in RTW. Evidence for the effectiveness of other single-domain interventions was mixed, with some studies reporting positive effects and others reporting no effects on lost time and work functioning. Conclusions While there is substantial research literature focused on RTW, there are only a small number of quality workplace-based RTW intervention studies that involve workers with MSK or pain-related conditions and MH conditions. We recommend implementing multi-domain interventions (i.e. with healthcare provision, service coordination, and work accommodation components) to help reduce lost time for MSK or pain-related conditions and MH conditions. Practitioners should also consider implementing these programs to help improve work functioning and reduce costs associated with work disability.

UK Biobank is a large-scale prospective epidemiological study with all data accessible to researchers worldwide. It is currently in the process of bringing back 100,000 of the original participants for brain, heart and body MRI, carotid ultrasound and low-dose bone/fat x-ray. The brain imaging component covers 6 modalities (T1, T2 FLAIR, susceptibility weighted MRI, Resting fMRI, Task fMRI and Diffusion MRI). Raw and processed data from the first 10,000 imaged subjects has recently been released for general research access. To help convert this data into useful summary information we have developed an automated processing and QC (Quality Control) pipeline that is available for use by other researchers. In this paper we describe the pipeline in detail, following a brief overview of UK Biobank brain imaging and the acquisition protocol. We also describe several quantitative investigations carried out as part of the development of both the imaging protocol and the processing pipeline.

Diffusion MRI is an exquisitely sensitive probe of tissue microstructure, and is currently the only non-invasive measure of the brain's fibre architecture. As this technique becomes more sophisticated and microstructurally informative, there is increasing value in comparing diffusion MRI with microscopic imaging in the same tissue samples. This study compared estimates of fibre orientation dispersion in white matter derived from diffusion MRI to reference measures of dispersion obtained from polarized light imaging and histology. Three post-mortem brain specimens were scanned with diffusion MRI and analyzed with a two-compartment dispersion model. The specimens were then sectioned for microscopy, including polarized light imaging estimates of fibre orientation and histological quantitative estimates of myelin and astrocytes. Dispersion estimates were correlated on region - and voxel-wise levels in the corpus callosum, the centrum semiovale and the corticospinal tract. The region-wise analysis yielded correlation coefficients of r = 0.79 for the diffusion MRI and histology comparison, while r = 0.60 was reported for the comparison with polarized light imaging. In the corpus callosum, we observed a pattern of higher dispersion at the midline compared to its lateral aspects. This pattern was present in all modalities and the dispersion profiles from microscopy and diffusion MRI were highly correlated. The astrocytes appeared to have minor contribution to dispersion observed with diffusion MRI. These results demonstrate that fibre orientation dispersion estimates from diffusion MRI represents the tissue architecture well. Dispersion models might be improved by more faithfully incorporating an informed mapping based on microscopy data.