In this paper, we propose a new automated procedure for lesion identification from single images based on the detection of outlier voxels. We demonstrate the utility of this procedure using artificial and real lesions. The scheme rests on two innovations: First, we augment the generative model used for combined segmentation and normalization of images, with an empirical prior for an atypical tissue class, which can be optimised iteratively. Second, we adopt a fuzzy clustering procedure to identify outlier voxels in normalised gray and white matter segments. These two advances suppress misclassification of voxels and restrict lesion identification to gray/white matter lesions respectively. Our analyses show a high sensitivity for detecting and delineating brain lesions with different sizes, locations, and textures. Our approach has important implications for the generation of lesion overlap maps of a given population and the assessment of lesion-deficit mappings. From a clinical perspective, our method should help to compute the total volume of lesion or to trace precisely lesion boundaries that might be pertinent for surgical or diagnostic purposes.

Competing theories of short-term memory function make specific predictions about the functional anatomy of auditory short-term memory and its role in language comprehension. We analysed high-resolution structural magnetic resonance images from 210 stroke patients and employed a novel voxel based analysis to test the relationship between auditory short-term memory and speech comprehension. Using digit span as an index of auditory short-term memory capacity we found that the structural integrity of a posterior region of the superior temporal gyrus and sulcus predicted auditory short-term memory capacity, even when performance on a range of other measures was factored out. We show that the integrity of this region also predicts the ability to comprehend spoken sentences. Our results therefore support cognitive models that posit a shared substrate between auditory short-term memory capacity and speech comprehension ability. The method applied here will be particularly useful for modelling structure–function relationships within other complex cognitive domains.

Functional imaging studies of language have shown bilateral superior temporal activations in response to ‘passive’ perception of speech when the baseline condition did not control for the acoustic complexity of speech. Controlling for this complexity demonstrates speech-specific processing lateralized to the left temporal lobe, and our recent positron emission tomography study has emphasized a role for left anterolateral temporal cortex in speech comprehension. This contrasts with the more usual view that relates speech comprehension to left temporal-parietal cortex, the ill-defined area of Wernicke. This study attempted to reconcile these differences, using a more sensitive 3 T functional magnetic resonance imaging system, and a sparse sampling paradigm. We found left lateralized activations for intelligible speech with two distinct foci, one in the anterior superior temporal sulcus and the other on the posterior temporal lobe. Therefore, the results demonstrate that there are neural responses to intelligible speech along the length of the left lateral temporal neocortex, although the precise processing roles of the anterior and posterior regions cannot be determined from this study.

Cerebral microbleeds are a neuroimaging biomarker of stroke risk. A crucial clinical question is whether cerebral microbleeds indicate patients with recent ischaemic stroke or transient ischaemic attack in whom the rate of future intracranial haemorrhage is likely to exceed that of recurrent ischaemic stroke when treated with antithrombotic drugs. We therefore aimed to establish whether a large burden of cerebral microbleeds or particular anatomical patterns of cerebral microbleeds can identify ischaemic stroke or transient ischaemic attack patients at higher absolute risk of intracranial haemorrhage than ischaemic stroke. We did a pooled analysis of individual patient data from cohort studies in adults with recent ischaemic stroke or transient ischaemic attack. Cohorts were eligible for inclusion if they prospectively recruited adult participants with ischaemic stroke or transient ischaemic attack; included at least 50 participants; collected data on stroke events over at least 3 months follow-up; used an appropriate MRI sequence that is sensitive to magnetic susceptibility; and documented the number and anatomical distribution of cerebral microbleeds reliably using consensus criteria and validated scales. Our prespecified primary outcomes were a composite of any symptomatic intracranial haemorrhage or ischaemic stroke, symptomatic intracranial haemorrhage, and symptomatic ischaemic stroke. We registered this study with the PROSPERO international prospective register of systematic reviews, number CRD42016036602. Between Jan 1, 1996, and Dec 1, 2018, we identified 344 studies. After exclusions for ineligibility or declined requests for inclusion, 20 322 patients from 38 cohorts (over 35 225 patient-years of follow-up; median 1·34 years [IQR 0·19-2·44]) were included in our analyses. The adjusted hazard ratio [aHR] comparing patients with cerebral microbleeds to those without was 1·35 (95% CI 1·20-1·50) for the composite outcome of intracranial haemorrhage and ischaemic stroke; 2·45 (1·82-3·29) for intracranial haemorrhage and 1·23 (1·08-1·40) for ischaemic stroke. The aHR increased with increasing cerebral microbleed burden for intracranial haemorrhage but this effect was less marked for ischaemic stroke (for five or more cerebral microbleeds, aHR 4·55 [95% CI 3·08-6·72] for intracranial haemorrhage vs 1·47 [1·19-1·80] for ischaemic stroke; for ten or more cerebral microbleeds, aHR 5·52 [3·36-9·05] vs 1·43 [1·07-1·91]; and for ≥20 cerebral microbleeds, aHR 8·61 [4·69-15·81] vs 1·86 [1·23-1·82]). However, irrespective of cerebral microbleed anatomical distribution or burden, the rate of ischaemic stroke exceeded that of intracranial haemorrhage (for ten or more cerebral microbleeds, 64 ischaemic strokes [95% CI 48-84] per 1000 patient-years vs 27 intracranial haemorrhages [17-41] per 1000 patient-years; and for ≥20 cerebral microbleeds, 73 ischaemic strokes [46-108] per 1000 patient-years vs 39 intracranial haemorrhages [21-67] per 1000 patient-years). In patients with recent ischaemic stroke or transient ischaemic attack, cerebral microbleeds are associated with a greater relative hazard (aHR) for subsequent intracranial haemorrhage than for ischaemic stroke, but the absolute risk of ischaemic stroke is higher than that of intracranial haemorrhage, regardless of cerebral microbleed presence, antomical distribution, or burden. British Heart Foundation and UK Stroke Association.

Decoding models, such as those underlying multivariate classification algorithms, have been increasingly used to infer cognitive or clinical brain states from measures of brain activity obtained by functional magnetic resonance imaging (fMRI). The practicality of current classifiers, however, is restricted by two major challenges. First, due to the high data dimensionality and low sample size, algorithms struggle to separate informative from uninformative features, resulting in poor generalization performance. Second, popular discriminative methods such as support vector machines (SVMs) rarely afford mechanistic interpretability. In this paper, we address these issues by proposing a novel generative-embedding approach that incorporates neurobiologically interpretable generative models into discriminative classifiers. Our approach extends previous work on trial-by-trial classification for electrophysiological recordings to subject-by-subject classification for fMRI and offers two key advantages over conventional methods: it may provide more accurate predictions by exploiting discriminative information encoded in ‘hidden’ physiological quantities such as synaptic connection strengths; and it affords mechanistic interpretability of clinical classifications. Here, we introduce generative embedding for fMRI using a combination of dynamic causal models (DCMs) and SVMs. We propose a general procedure of DCM-based generative embedding for subject-wise classification, provide a concrete implementation, and suggest good-practice guidelines for unbiased application of generative embedding in the context of fMRI. We illustrate the utility of our approach by a clinical example in which we classify moderately aphasic patients and healthy controls using a DCM of thalamo-temporal regions during speech processing. Generative embedding achieves a near-perfect balanced classification accuracy of 98% and significantly outperforms conventional activation-based and correlation-based methods. This example demonstrates how disease states can be detected with very high accuracy and, at the same time, be interpreted mechanistically in terms of abnormalities in connectivity. We envisage that future applications of generative embedding may provide crucial advances in dissecting spectrum disorders into physiologically more well-defined subgroups. Neurological and psychiatric spectrum disorders are typically defined in terms of particular symptom sets, despite increasing evidence that the same symptom may be caused by very different pathologies. Pathophysiological classification and effective treatment of such disorders will increasingly require a mechanistic understanding of inter-individual differences and clinical tools for making accurate diagnostic inference in individual patients. Previous classification studies have shown that functional magnetic resonance imaging (fMRI) can be used to differentiate between healthy controls and neurological or psychiatric patients. However, these studies are typically based on descriptive patterns and indirect measures of neural activity, and they rarely afford mechanistic insights into the underlying condition. In this paper, we address this challenge by proposing a classification approach that rests on a model of brain function and exploits the rich discriminative information encoded in directed interregional connection strengths. Based on an fMRI dataset acquired from moderately aphasic patients and healthy controls, we illustrate that our approach enables more accurate classification and deeper mechanistic insights about disease processes than conventional classification methods.