The ENIGMA Stroke Recovery Working Group: Big data neuroimaging to study brain-behavior relationships after stroke.

The goal of the Enhancing Neuroimaging Genetics through Meta-Analysis (ENIGMA) Stroke Recovery working group is to understand brain and behavior relationships using well-powered meta- and mega-analytic approaches. ENIGMA Stroke Recovery has data from over 2,100 stroke patients collected across 39 research studies and 10 countries around the world, comprising the largest multisite retrospective stroke data collaboration to date. This article outlines the efforts taken by the ENIGMA Stroke Recovery working group to develop neuroinformatics protocols and methods to manage multisite stroke brain magnetic resonance imaging, behavioral and demographics data. Specifically, the processes for scalable data intake and preprocessing, multisite data harmonization, and large-scale stroke lesion analysis are described, and challenges unique to this type of big data collaboration in stroke research are discussed. Finally, future directions and limitations, as well as recommendations for improved data harmonization through prospective data collection and data management, are provided.

Sensorimotor cortex beta oscillations reflect motor skill learning ability after stroke.

Recovery of skilled movement after stroke is assumed to depend on motor learning. However, the capacity for motor learning and factors that influence motor learning after stroke have received little attention. In this study, we first compared motor skill acquisition and retention between well-recovered stroke patients and age- and performance-matched healthy controls. We then tested whether beta oscillations (15-30 Hz) from sensorimotor cortices contribute to predicting training-related motor performance. Eighteen well-recovered chronic stroke survivors (mean age 64 ± 8 years, range: 50-74 years) and 20 age- and sex-matched healthy controls were trained on a continuous tracking task and subsequently retested after initial training (45-60 min and 24 h later). Scalp electroencephalography was recorded during the performance of a simple motor task before each training and retest session. Stroke patients demonstrated capacity for motor skill learning, but it was diminished compared to age- and performance-matched healthy controls. Furthermore, although the properties of beta oscillations prior to training were comparable between stroke patients and healthy controls, stroke patients did show less change in beta measures with motor learning. Lastly, although beta oscillations did not help to predict motor performance immediately after training, contralateral (ipsilesional) sensorimotor cortex post-movement beta rebound measured after training helped predict future motor performance, 24 h after training. This finding suggests that neurophysiological measures such as beta oscillations can help predict response to motor training in chronic stroke patients and may offer novel targets for therapeutic interventions.

Support vector machine-based classification of neuroimages in Alzheimer's disease: direct comparison of FDG-PET, rCBF-SPECT and MRI data acquired from the same individuals

Objective: To conduct the first support vector machine (SVM)-based study comparing the diagnostic accuracy of T1-weighted magnetic resonance imaging (T1-MRI), F-fluorodeoxyglucose-positron emission tomography (FDG-PET) and regional cerebral blood flow single-photon emission computed tomography (rCBF-SPECT) in Alzheimer's disease (AD). Method: Brain T1-MRI, FDG-PET and rCBF-SPECT scans were acquired from a sample of mild AD patients (n=20) and healthy elderly controls (n=18). SVM-based diagnostic accuracy indices were calculated using whole-brain information and leave-one-out cross-validation. Results: The accuracy obtained using PET and SPECT data were similar. PET accuracy was 68∼71% and area under curve (AUC) 0.77∼0.81; SPECT accuracy was 68∼74% and AUC 0.75∼0.79, and both had better performance than analysis with T1-MRI data (accuracy of 58%, AUC 0.67). The addition of PET or SPECT to MRI produced higher accuracy indices (68∼74%; AUC: 0.74∼0.82) than T1-MRI alone, but these were not clearly superior to the isolated neurofunctional modalities. Conclusion: In line with previous evidence, FDG-PET and rCBF-SPECT more accurately identified patients with AD than T1-MRI, and the addition of either PET or SPECT to T1-MRI data yielded increased accuracy. The comparable SPECT and PET performances, directly demonstrated for the first time in the present study, support the view that rCBF-SPECT still has a role to play in AD diagnosis.

Support vector machine-based classification of neuroimages in Alzheimer's disease: direct comparison of FDG-PET, rCBF-SPECT and MRI data acquired from the same individuals.

OBJECTIVE: To conduct the first support vector machine (SVM)-based study comparing the diagnostic accuracy of T1-weighted magnetic resonance imaging (T1-MRI), F-fluorodeoxyglucose-positron emission tomography (FDG-PET) and regional cerebral blood flow single-photon emission computed tomography (rCBF-SPECT) in Alzheimer's disease (AD). METHOD: Brain T1-MRI, FDG-PET and rCBF-SPECT scans were acquired from a sample of mild AD patients (n=20) and healthy elderly controls (n=18). SVM-based diagnostic accuracy indices were calculated using whole-brain information and leave-one-out cross-validation. RESULTS: The accuracy obtained using PET and SPECT data were similar. PET accuracy was 68∼71% and area under curve (AUC) 0.77∼0.81; SPECT accuracy was 68∼74% and AUC 0.75∼0.79, and both had better performance than analysis with T1-MRI data (accuracy of 58%, AUC 0.67). The addition of PET or SPECT to MRI produced higher accuracy indices (68∼74%; AUC: 0.74∼0.82) than T1-MRI alone, but these were not clearly superior to the isolated neurofunctional modalities. CONCLUSION: In line with previous evidence, FDG-PET and rCBF-SPECT more accurately identified patients with AD than T1-MRI, and the addition of either PET or SPECT to T1-MRI data yielded increased accuracy. The comparable SPECT and PET performances, directly demonstrated for the first time in the present study, support the view that rCBF-SPECT still has a role to play in AD diagnosis.

Brain regions important for recovery after severe post-stroke upper limb paresis.

Background The ability to predict outcome after stroke is clinically important for planning treatment and for stratification in restorative clinical trials. In relation to the upper limbs, the main predictor of outcome is initial severity, with patients who present with mild to moderate impairment regaining about 70% of their initial impairment by 3 months post-stroke. However, in those with severe presentations, this proportional recovery applies in only about half, with the other half experiencing poor recovery. The reasons for this failure to recover are not established although the extent of corticospinal tract damage is suggested to be a contributory factor. In this study, we investigated 30 patients with chronic stroke who had presented with severe upper limb impairment and asked whether it was possible to differentiate those with a subsequent good or poor recovery of the upper limb based solely on a T1-weighted structural brain scan. Methods A support vector machine approach using voxel-wise lesion likelihood values was used to show that it was possible to classify patients as good or poor recoverers with variable accuracy depending on which brain regions were used to perform the classification. Results While considering damage within a corticospinal tract mask resulted in 73% classification accuracy, using other (non-corticospinal tract) motor areas provided 87% accuracy, and combining both resulted in 90% accuracy. Conclusion This proof of concept approach highlights the relative importance of different anatomical structures in supporting post-stroke upper limb motor recovery and points towards methodologies that might be used to stratify patients in future restorative clinical trials.

Decoding post-stroke motor function from structural brain imaging.

Clinical research based on neuroimaging data has benefited from machine learning methods, which have the ability to provide individualized predictions and to account for the interaction among units of information in the brain. Application of machine learning in structural imaging to investigate diseases that involve brain injury presents an additional challenge, especially in conditions like stroke, due to the high variability across patients regarding characteristics of the lesions. Extracting data from anatomical images in a way that translates brain damage information into features to be used as input to learning algorithms is still an open question. One of the most common approaches to capture regional information from brain injury is to obtain the lesion load per region (i.e. the proportion of voxels in anatomical structures that are considered to be damaged). However, no systematic evaluation has yet been performed to compare this approach with using patterns of voxels (i.e. considering each voxel as a single feature). In this paper we compared both approaches applying Gaussian Process Regression to decode motor scores in 50 chronic stroke patients based solely on data derived from structural MRI. For both approaches we compared different ways to delimit anatomical areas: regions of interest from an anatomical atlas, the corticospinal tract, a mask obtained from fMRI analysis with a motor task in healthy controls and regions selected using lesion-symptom mapping. Our analysis showed that extracting features through patterns of voxels that represent lesion probability produced better results than quantifying the lesion load per region. In particular, from the different ways to delimit anatomical areas compared, the best performance was obtained with a combination of a range of cortical and subcortical motor areas as well as the corticospinal tract. These results will inform the appropriate methodology for predicting long term motor outcomes from early post-stroke structural brain imaging.

Pattern changes of EEG oscillations and BOLD signals associated with temporal lobe epilepsy as revealed by a working memory task.

BACKGROUND: It is known that the abnormal neural activity in epilepsy may be associated to the reorganization of neural circuits and brain plasticity in various ways. On that basis, we hypothesized that changes in neuronal circuitry due to epilepsy could lead to measurable variations in patterns of both EEG and BOLD signals in patients performing some cognitive task as compared to what would be obtained in normal condition. Thus, the aim of this study was to compare the cerebral areas involved in EEG oscillations versus fMRI signal patterns during a working memory (WM) task in normal controls and patients with refractory mesial temporal lobe epilepsy (MTLE) associated with hippocampal sclerosis (HS). The study included six patients with left MTLE-HS (left-HS group) and seven normal controls (control group) matched to the patients by age and educational level, both groups undergoing a blocked design paradigm based on Sternberg test during separated EEG and fMRI sessions. This test consisted of encoding and maintenance of a variable number of consonant letters on WM. RESULTS: EEG analysis for the encoding period revealed the presence of theta and alpha oscillations in the frontal and parietal areas, respectively. Likewise, fMRI showed the co-occurrence of positive and negative BOLD signals in both brain regions. As for the maintenance period, whereas EEG analysis revealed disappearance of theta oscillation, fMRI showed decrease of positive BOLD in frontal area and increase of negative BOLD in the posterior part of the brain. CONCLUSIONS: Generally speaking, these patterns of electrophysiological and hemodynamic signals were observed for both control and left-HS groups. However, the data also revealed remarkable differences between these groups that are consistent with the hypothesis of reorganization of brain circuitry associated with epilepsy.

SCoRS--A Method Based on Stability for Feature Selection and Mapping inNeuroimaging [corrected].

Feature selection (FS) methods play two important roles in the context of neuroimaging based classification: potentially increase classification accuracy by eliminating irrelevant features from the model and facilitate interpretation by identifying sets of meaningful features that best discriminate the classes. Although the development of FS techniques specifically tuned for neuroimaging data is an active area of research, up to date most of the studies have focused on finding a subset of features that maximizes accuracy. However, maximizing accuracy does not guarantee reliable interpretation as similar accuracies can be obtained from distinct sets of features. In the current paper we propose a new approach for selecting features: SCoRS (survival count on random subsamples) based on a recently proposed Stability Selection theory. SCoRS relies on the idea of choosing relevant features that are stable under data perturbation. Data are perturbed by iteratively sub-sampling both features (subspaces) and examples. We demonstrate the potential of the proposed method in a clinical application to classify depressed patients versus healthy individuals based on functional magnetic resonance imaging data acquired during visualization of happy faces.

Brain plasticity for verbal and visual memories in patients with mesial temporal lobe epilepsy and hippocampal sclerosis: an fMRI study.

We aimed to identify the brain areas involved in verbal and visual memory processing in normal controls and patients with unilateral mesial temporal lobe epilepsy (MTLE) associated with unilateral hippocampal sclerosis (HS) by means of functional magnetic resonance imaging (fMRI). The sample comprised nine normal controls, eight patients with right MTLE, and nine patients with left MTLE. All subjects underwent fMRI with verbal and visual memory paradigms, consisting of encoding and immediate recall of 17 abstract words and 17 abstract drawings. A complex network including parietal, temporal, and frontal cortices seems to be involved in verbal memory encoding and retrieval in normal controls. Although similar areas of activation were identified in both patient groups, the extension of such activations was larger in the left-HS group. Patients with left HS also tended to exhibit more bilateral or right lateralized encoding related activations. This finding suggests a functional reorganization of verbal memory processing areas in these patients due to the failure of left MTL system. As regards visual memory encoding and retrieval, our findings support the hypothesis of a more diffuse and bilateral representation of this cognitive function in the brain. Compared to normal controls, encoding in the left-HS group recruited more widespread cortical areas, which were even more widespread in the right-HS group probably to compensate for their right mesial temporal dysfunction. In contrast, the right-HS group exhibited fewer activated areas during immediate recall than the other two groups, probably related to their greater difficulty in dealing with visual memory content.

Asymmetrical hippocampal connectivity in mesial temporal lobe epilepsy: evidence from resting state fMRI.

BACKGROUND: Mesial temporal lobe epilepsy (MTLE), the most common type of focal epilepsy in adults, is often caused by hippocampal sclerosis (HS). Patients with HS usually present memory dysfunction, which is material-specific according to the hemisphere involved and has been correlated to the degree of HS as measured by postoperative histopathology as well as by the degree of hippocampal atrophy on magnetic resonance imaging (MRI). Verbal memory is mostly affected by left-sided HS, whereas visuo-spatial memory is more affected by right HS. Some of these impairments may be related to abnormalities of the network in which individual hippocampus takes part. Functional connectivity can play an important role to understand how the hippocampi interact with other brain areas. It can be estimated via functional Magnetic Resonance Imaging (fMRI) resting state experiments by evaluating patterns of functional networks. In this study, we investigated the functional connectivity patterns of 9 control subjects, 9 patients with right MTLE and 9 patients with left MTLE. RESULTS: We detected differences in functional connectivity within and between hippocampi in patients with unilateral MTLE associated with ipsilateral HS by resting state fMRI. Functional connectivity resulted to be more impaired ipsilateral to the seizure focus in both patient groups when compared to control subjects. This effect was even more pronounced for the left MTLE group. CONCLUSIONS: The findings presented here suggest that left HS causes more reduction of functional connectivity than right HS in subjects with left hemisphere dominance for language.