Predicting functional outcome in patients with acute brainstem infarction using deep neuroimaging features.

BACKGROUND AND PURPOSE: Acute brainstem infarctions can lead to serious functional impairments. We aimed to predict functional outcomes in patients with acute brainstem infarction using deep neuroimaging features extracted by convolutional neural networks (CNNs). METHODS: This nationwide multicenter stroke registry study included 1482 patients with acute brainstem infarction. We applied CNNs to automatically extract deep neuroimaging features from diffusion-weighted imaging. Deep learning models based on clinical features, laboratory features, conventional imaging features (infarct volume, number of infarctions), and deep neuroimaging features were trained to predict functional outcomes at 3 months poststroke. Unfavorable outcome was defined as modified Rankin Scale score of 3 or higher at 3 months. The models were evaluated by comparing the area under the receiver operating characteristic curve (AUC). RESULTS: A model based solely on 14 deep neuroimaging features from CNNs achieved an extremely high AUC of 0.975 (95% confidence interval [CI] = 0.934-0.997) and significantly outperformed the model combining clinical, laboratory, and conventional imaging features (0.772, 95% CI = 0.691-0.847, p < 0.001) in prediction of functional outcomes. The deep neuroimaging model also demonstrated significant improvement over traditional prognostic scores. In an interpretability analysis, the deep neuroimaging features displayed a significant correlation with age, National Institutes of Health Stroke Scale score, infarct volume, and inflammation factors. CONCLUSIONS: Deep learning models can successfully extract objective neuroimaging features from the routine radiological data in an automatic manner and aid in predicting the functional outcomes in patients with brainstem infarction at 3 months with very high accuracy.

Temporal Evolution of Perihematomal Blood-Brain Barrier Compromise and Edema Growth After Intracerebral Hemorrhage.

PURPOSE: The aim of this study was to investigate the temporal evolution of perihematomal blood-brain barrier (BBB) compromise and edema growth and to determine the role of BBB compromise in edema growth. METHODS: Spontaneous intracerebral hemorrhage patients who underwent computed tomography perfusion (CTP) were divided into five groups according to the time interval from symptom onset to CTP examination. Permeability-surface area product (PS) maps were generated using CTP source images. Ipsilateral and contralateral mean PS values were computed in the perihematomal and contralateral mirror regions. The relative PS (rPS) value was calculated as a ratio of ipsilateral to contralateral PS value. Hematoma and perihematomal edema volume were determined on non-contrast CT images. RESULTS: In the total of 101 intracerebral hemorrhage patients, the ipsilateral mean PS value was significantly higher than that in contralateral region (z = -8.284, p < 0.001). The perihematomal BBB permeability showed a course of dynamic changes including an increase in the hyperacute and acute phases, a decrease in the early subacute phase and a second increase in the late subacute phase and chronic phase. Perihematomal edema increased gradually until the late subacute phase and then slightly increased. There was a relationship between rPS value and edema volume (ß = 0.254, p = 0.006). CONCLUSION: The perihematomal BBB permeability is dynamic changes, and edema growth is gradually increased in patients following intracerebral hemorrhage. BBB compromise plays an essential role in edema growth. The quantitative assessment of BBB compromise may provide valuable information in therapeutic interventions of intracerebral hemorrhage patients.

Poststroke motor, cognitive and speech rehabilitation with brain-computer interface: a perspective review.

Brain-computer interface (BCI) technology translates brain activity into meaningful commands to establish a direct connection between the brain and the external world. Neuroscientific research in the past two decades has indicated a tremendous potential of BCI systems for the rehabilitation of patients suffering from poststroke impairments. By promoting the neuronal recovery of the damaged brain networks, BCI systems have achieved promising results for the recovery of poststroke motor, cognitive, and language impairments. Also, several assistive BCI systems that provide alternative means of communication and control to severely paralysed patients have been proposed to enhance patients' quality of life. In this article, we present a perspective review of the recent advances and challenges in the BCI systems used in the poststroke rehabilitation of motor, cognitive, and communication impairments.

Data-driven clustering approach to identify novel phenotypes using multiple biomarkers in acute ischaemic stroke: A retrospective, multicentre cohort study.

Background: Acute ischaemic stroke (AIS) is a highly heterogeneous disorder and warrants further investigation to stratify patients with different outcomes and treatment responses. Using a large-scale stroke registry cohort, we applied data-driven approach to identify novel phenotypes based on multiple biomarkers. Methods: In a nationwide, prospective, 201-hospital registry study taking place in China between August 01, 2015 and March 31, 2018, the patients with AIS who were over 18 years of age and admitted to the hospital within 7 days from symptom onset were included. 92 biomarkers were included in the analysis. In the derivation cohort (n=9539), an unsupervised Gaussian mixture model was applied to categorize patients into distinct phenotypes. A classifier was developed using the most important biomarkers and was applied to categorize patients into their corresponding phenotypes in an validation cohort (n=2496). The differences in biological features, clinical outcomes, and treatment response were compared across the phenotypes. Findings: We identified four phenotypes with distinct characteristics in 9288 patients with non-cardioembolic ischaemic stroke. Phenotype 1 was associated with abnormal glucose and lipid metabolism. Phenotype 2 was characterized by inflammation and abnormal renal function. Phenotype 3 had the least laboratory abnormalities and small infarct lesions. Phenotype 4 was characterized by disturbance in homocysteine metabolism. Findings were replicated in the validation cohort. In comparison with phenotype 3, the risk of stroke recurrence (adjusted hazard ratio [aHR] 2.02, 95% confidence intervals [CI] 1.04-3.94), and mortality (aHR 18.14, 95%CI 6.62-49.71) at 3-month post-stroke were highest in phenotype 2, followed by phenotype 4 and phenotype 1, after adjustment for age, gender, smoking, drinking, history of stroke, hypertension, diabetes mellitus, dyslipidemia, and coronary heart disease. The Monte Carlo simulation showed that the patients with phenotype 2 could benefit from high-intensity statin therapy. Interpretation: A data-driven approach could aid in the identification of patients at a higher risk of adverse clinical outcomes following non-cardioembolic ischaemic stroke. These phenotypes, based on different pathophysiology, can suggest individualized treatment plans. Funding: Beijing Natural Science Foundation (grant number Z200016), Beijing Municipal Committee of Science and Technology (grant number Z201100005620010), National Natural Science Foundation of China (grant number 82101360, 92046016, 82171270), Chinese Academy of Medical Sciences Innovation Fund for Medical Sciences (grant number 2019-I2M-5-029).

Deep Learning for Neuroimaging Segmentation with a Novel Data Augmentation Strategy.

Brain insults such as cerebral ischemia and intracranial hemorrhage are critical stroke conditions with high mortality rates. Currently, medical image analysis for critical stroke conditions is still largely done manually, which is time-consuming and labor-intensive. While deep learning algorithms are increasingly being applied in medical image analysis, the performance of these methods still needs substantial improvement before they can be widely used in the clinical setting. Among other challenges, the lack of sufficient labelled data is one of the key problems that has limited the progress of deep learning methods in this domain. To mitigate this bottleneck, we propose an integrated method that includes a data augmentation framework using a conditional Generative Adversarial Network (cGAN) which is followed by a supervised segmentation with a Convolutional Neural Network (CNN). The adopted cGAN generates meaningful brain images from specially altered lesion masks as a form of data augmentation to supplement the training dataset, while the CNN incorporates depth-wise-convolution based X-blocks as well as Feature Similarity Module (FSM) to ease and aid the training process, resulting in better lesion segmentation. We evaluate the proposed deep learning strategy on the Anatomical Tracings of Lesions After Stroke (ATLAS) dataset and show that this approach outperforms the current state-of-art methods in task of stroke lesion segmentation.

BCI for stroke rehabilitation: motor and beyond.

Stroke is one of the leading causes of long-term disability among adults and contributes to major socio-economic burden globally. Stroke frequently results in multifaceted impairments including motor, cognitive and emotion deficits. In recent years, brain-computer interface (BCI)-based therapy has shown promising results for post-stroke motor rehabilitation. In spite of the success received by BCI-based interventions in the motor domain, non-motor impairments are yet to receive similar attention in research and clinical settings. Some preliminary encouraging results in post-stroke cognitive rehabilitation using BCI seem to suggest that it may also hold potential for treating non-motor deficits such as cognitive and emotion impairments. Moreover, past studies have shown an intricate relationship between motor, cognitive and emotion functions which might influence the overall post-stroke rehabilitation outcome. A number of studies highlight the inability of current treatment protocols to account for the implicit interplay between motor, cognitive and emotion functions. This indicates the necessity to explore an all-inclusive treatment plan targeting the synergistic influence of these standalone interventions. This approach may lead to better overall recovery than treating the individual deficits in isolation. In this paper, we review the recent advances in BCI-based post-stroke motor rehabilitation and highlight the potential for the use of BCI systems beyond the motor domain, in particular, in improving cognition and emotion of stroke patients. Building on the current results and findings of studies in individual domains, we next discuss the possibility of a holistic BCI system for motor, cognitive and affect rehabilitation which may synergistically promote restorative neuroplasticity. Such a system would provide an all-encompassing rehabilitation platform, leading to overarching clinical outcomes and transfer of these outcomes to a better quality of living. This is one of the first works to analyse the possibility of targeting cross-domain influence of post-stroke functional recovery enabled by BCI-based rehabilitation.

A Multi-view CNN with Novel Variance Layer for Motor Imagery Brain Computer Interface.

Accurate and robust classification of Motor Imagery (MI) from Electroencephalography (EEG) signals is among the most challenging tasks in Brain-Computer Interface (BCI) field. To address this challenge, this paper proposes a novel, neuro-physiologically inspired convolutional neural network (CNN) named Filter-Bank Convolutional Network (FBCNet) for MI classification. Capturing neurophysiological signatures of MI, FBCNet first creates a multi-view representation of the data by bandpass-filtering the EEG into multiple frequency bands. Next, spatially discriminative patterns for each view are learned using a CNN layer. Finally, the temporal information is aggregated using a new variance layer and a fully connected layer classifies the resultant features into MI classes. We evaluate the performance of FBCNet on a publicly available dataset from Korea University for classification of left vs right hand MI in a subject-specific 10-fold cross-validation setting. Results show that FBCNet achieves more than 6.7% higher accuracy compared to other state-of-the-art deep learning architectures while requiring less than 1% of the learning parameters. We explain the higher classification accuracy achieved by FBCNet using feature visualization where we show the superiority of FBCNet in learning interpretable and highly generalizable discriminative features. We provide the source code of FBCNet for reproducibility of results.

Federated Transfer Learning for EEG Signal Classification.

The success of deep learning (DL) methods in the Brain-Computer Interfaces (BCI) field for classification of electroencephalographic (EEG) recordings has been restricted by the lack of large datasets. Privacy concerns associated with EEG signals limit the possibility of constructing a large EEG-BCI dataset by the conglomeration of multiple small ones for jointly training machine learning models. Hence, in this paper, we propose a novel privacy-preserving DL architecture named federated transfer learning (FTL) for EEG classification that is based on the federated learning framework. Working with the single-trial covariance matrix, the proposed architecture extracts common discriminative information from multi-subject EEG data with the help of domain adaptation techniques. We evaluate the performance of the proposed architecture on the PhysioNet dataset for 2-class motor imagery classification. While avoiding the actual data sharing, our FTL approach achieves 2% higher classification accuracy in a subject-adaptive analysis. Also, in the absence of multi-subject data, our architecture provides 6% better accuracy compared to other state-of-the-art DL architectures.

Prognostic and Monitory EEG-Biomarkers for BCI Upper-Limb Stroke Rehabilitation.

With the availability of multiple rehabilitative interventions, identifying the one that elicits the best motor outcome based on the unique neuro-clinical profile of the stroke survivor is a challenging task. Predicting the potential of recovery using biomarkers specific to an intervention hence becomes important. To address this, we investigate intervention-specific prognostic and monitory biomarkers of motor function improvements using quantitative electroencephalography (QEEG) features in 19 chronic stroke patients following two different upper extremity rehabilitative interventions viz. Brain-computer interface (BCI) and transcranial direct current stimulation coupled BCI (tDCS-BCI). Brain symmetry index was found to be the best prognostic QEEG for clinical gains following BCI intervention ( r = -0.80 , p = 0.02 ), whereas power ratio index (PRI) was observed to be the best predictor for tDCS-BCI ( r = -0.96 , p = 0.004 ) intervention. Importantly, statistically significant between-intervention differences observed in the predictive capabilities of these features suggest that intervention-specific biomarkers can be identified. This approach can be further pursued to distinctly predict the expected response of a patient to available interventions. The intervention with the highest predicted gains may then be recommended to the patient, thereby enabling a personalized rehabilitation regime.

Quantitative EEG as Biomarkers for the Monitoring of Post-Stroke Motor Recovery in BCI and tDCS Rehabilitation.

This study investigates the neurological changes in the brain activity of chronic stroke patients undergoing different types of motor rehabilitative interventions and their relationship with the clinical recovery using the Quantitative Electroencephalography (QEEG) features. Over a period of two weeks, 19 hemiplegic chronic stroke patients underwent 10 sessions of upper extremity motor rehabilitation using a brain-computer interface paradigm (BCI group, n=9) and transcranial direct current stimulation coupled BCI paradigm (tDCS group, n=10). The pre- and post-treatment brain activations, as well as the intervention-induced changes in the neuronal activity, were quantified using 11 QEEG features and their relationship with clinical motor improvement was investigated. Significant treatment-induced change in the relative theta power was observed in the BCI group and the change was significantly correlated with the clinical improvements. Also, in the BCI group, the relative theta power and interactions between the theta, alpha, and beta power were identified as monitory biomarkers of motor recovery. On the contrary, the tDCS group was characterized by the significant change in brain asymmetry. Furthermore, we observed significant intergroup differences in the predictive capabilities of post-intervention QEEG features between the BCI and tDCS group. Based on the intergroup differences observed in this study and convergent results from the other neuroimaging analysis performed on the same cohort, we suggest that distinctly different mechanisms of neuronal recovery were facilitated by tDCS and BCI interventions and these treatment specific mechanisms can be encapsulated using QEEG.