The new insights into human brain imaging after stroke

Magnetic resonance imaging-based deep learning imaging biomarker for predicting functional outcomes after acute ischemic stroke

PURPOSE

Clinical risk scores are essential for predicting outcomes in stroke patients. The advancements in deep learning (DL) techniques provide opportunities to develop prediction applications using magnetic resonance (MR) images. We aimed to develop an MR-based DL imaging biomarker for predicting outcomes in acute ischemic stroke (AIS) and evaluate its additional benefit to current risk scores.

METHOD

This study included 3338 AIS patients. We trained a DL model using deep neural network architectures on MR images and radiomics to predict poor functional outcomes at three months post-stroke. The DL model generated a DL score, which served as the DL imaging biomarker. We compared the predictive performance of this biomarker to five risk scores on a holdout test set. Additionally, we assessed whether incorporating the imaging biomarker into the risk scores improved the predictive performance.

RESULTS

The DL imaging biomarker achieved an area under the receiver operating characteristic curve (AUC) of 0.788. The AUCs of the five studied risk scores were 0.789, 0.793, 0.804, 0.810, and 0.826, respectively. The imaging biomarker's predictive performance was comparable to four of the risk scores but inferior to one (p = 0.038). Adding the imaging biomarker to the risk scores improved the AUCs (p-values) to 0.831 (0.003), 0.825 (0.001), 0.834 (0.003), 0.836 (0.003), and 0.839 (0.177), respectively. The net reclassification improvement and integrated discrimination improvement indices also showed significant improvements (all p < 0.001).

CONCLUSIONS

Using DL techniques to create an MR-based imaging biomarker is feasible and enhances the predictive ability of current risk scores.

Adaptation of the Concept of Brain Reserve for the Prediction of Stroke Outcome: Proxies, Neural Mechanisms, and Significance for Research

The prediction of stroke outcome is challenging due to the high inter-individual variability in stroke patients. We recently suggested the adaptation of the concept of brain reserve (BR) to improve the prediction of stroke outcome. This concept was initially developed alongside the one for the cognitive reserve for neurodegeneration and forms a valuable theoretical framework to capture high inter-individual variability in stroke patients. In the present work, we suggest and discuss (i) BR-proxies-quantitative brain characteristics at the time stroke occurs (e.g., brain volume, hippocampus volume), and (ii) proxies of brain pathology reducing BR (e.g., brain atrophy, severity of white matter hyperintensities), parameters easily available from a routine MRI examination that might improve the prediction of stroke outcome. Though the influence of these parameters on stroke outcome has been partly reported individually, their independent and combined impact is yet to be determined. Conceptually, BR is a continuous measure determining the amount of brain structure available to mitigate and compensate for stroke damage, thus reflecting individual differences in neural resources and a capacity to maintain performance and recover after stroke. We suggest that stroke outcome might be defined as an interaction between BR at the time stroke occurs and lesion load. BR in stroke can potentially be influenced, e.g., by modifying cardiovascular risk factors. In addition to the potential power of the BR concept in a mechanistic understanding of inter-individual variability in stroke outcome and establishing individualized therapeutic approaches, it might help to strengthen the synergy of preventive measures in stroke, neurodegeneration, and healthy aging.

FIND Stroke Recovery Study (FIND): rationale and protocol for a longitudinal observational cohort study of trajectories of recovery and biomarkers poststroke

INTRODUCTION

Comprehensive studies mapping domain-specific trajectories of recovery after stroke and biomarkers reflecting these processes are scarce. We, therefore, initiated an exploratory prospective observational study of stroke cases with repeated evaluation, the FIND Stroke Recovery Study. We aim to capture trajectories of recovery from different impairments, including cognition, in combination with broad profiling of blood and imaging biomarkers of the recovery.

METHODS AND ANALYSIS

We recruit individuals with first-ever stroke at the stroke unit at the Sahlgrenska University Hospital, Sweden, to FIND. The inclusion started early 2018 and we aim to enrol minimum 500 patients. Neurological and cognitive impairments across multiple domains are assessed using validated clinical assessment methods, advanced neuroimaging is performed and blood samples for biomarker measuring (protein, RNA and DNA) at inclusion and follow-up visits at 3 months, 6 months, 1 year, 2 years and 5 years poststroke. At baseline and at each follow-up visit, we also register clinical variables known to influence outcomes such as prestroke functioning, stroke severity, acute interventions, rehabilitation, other treatments, socioeconomic status, infections (including COVID-19) and other comorbidities. Recurrent stroke and other major vascular events are identified continuously in national registers.

ETHICS AND DISSEMINATION

FIND composes a unique stroke cohort with detailed phenotyping, repetitive assessments of outcomes across multiple neurological and cognitive domains and patient-reported outcomes as well as blood and imaging biomarker profiling. Ethical approval for the FIND study has been obtained from the Regional Ethics Review Board in Gothenburg and the Swedish Ethics Review Board. The results of this exploratory study will provide novel data on the time course of recovery and biomarkers after stroke. The description of this protocol will inform the stroke research community of our ongoing study and facilitate comparisons with other data sets.

TRIAL REGISTRATION NUMBER

The protocol is registered at http://www.

CLINICALTRIALS

gov, Study ID: NCT05708807.

Structural disconnectome mapping of cognitive function in poststroke patients

BACKGROUND AND PURPOSE

Sequalae following stroke represents a significant challenge in current rehabilitation. The location and size of focal lesions are only moderately predictive of the diverse cognitive outcome after stroke. One explanation building on recent work on brain networks proposes that the cognitive consequences of focal lesions are caused by damages to anatomically distributed brain networks supporting cognition rather than specific lesion locations.

METHODS

To investigate the association between poststroke structural disconnectivity and cognitive performance, we estimated individual level whole-brain disconnectivity probability maps based on lesion maps from 102 stroke patients using normative data from healthy controls. Cognitive performance was assessed in the whole sample using Montreal Cognitive Assessment, and a more comprehensive computerized test protocol was performed on a subset (n = 82).

RESULTS

Multivariate analysis using Partial Least Squares on the disconnectome maps revealed that higher disconnectivity in right insular and frontal operculum, superior temporal gyrus and putamen was associated with poorer MoCA performance, indicating that lesions in regions connected with these brain regions are more likely to cause cognitive impairment. Furthermore, our results indicated that disconnectivity within these clusters was associated with poorer performance across multiple cognitive domains.

CONCLUSIONS

These findings demonstrate that the extent and distribution of structural disconnectivity following stroke are sensitive to cognitive deficits and may provide important clinical information predicting poststroke cognitive sequalae.

Association of cerebral microvascular perfusion and diffusion dynamics detected by intravoxel incoherent motion-diffusion weighted imaging with initial neurological function and clinical outcome in acute ischemic stroke

Background

This work aimed to explore the association of cerebral microvascular perfusion and diffusion dynamics measured by intravoxel incoherent motion (IVIM) imaging with initial neurological function and clinical outcome in acute stroke.

Methods

In total, 39 patients were assessed with admission National Institutes of Health Stroke Scale (NIHSS) and day-90 modified Rankin Scale (mRS). The parametrical maps of IVIM were obtained, including apparent diffusion coefficient (ADC), pseudo-diffusion coefficient (D*), true diffusion coefficient (D) and perfusion fraction (f). The fD* was the product of f and D*. Moreover, the ratios of lesioned/contralateral parameters (rADC, rD, rD*, rf and rfD*) were also obtained. The differences of these parameters between the poor outcome group and good outcome group were evaluated. Partial correlation analysis was used to evaluate the correlations between the admission NIHSS/day-90 mRS and each parameter ratio, with lesion volumes controlled.

Results

The ADC, D, D*, f and fD* values of lesions were significantly reduced than those of the contralateral regions. The rADC and rD were significantly decreased in the poor outcome group than good outcome group (all p < 0.01). With lesion volume controlled, rADC showed a weak negative correlation (r = −0.340, p = 0.037) and a notable negative correlation (r = −0.688, p < 0.001) with admission NIHSS score and day-90 mRS score, respectively. In addition, rD showed a strong negative correlation (r = −0.731, p < 0.001) with day-90 mRS score.

Conclusion

Significant negative correlations were revealed between IVIM derived diffusion dynamics parameters and initial neurological function as well as clinical outcome for patients with acute ischemic stroke. IVIM can be therefore suggested as an effective non-invasive method for evaluating the acute ischemic stroke.

Dengzhanxixin Injection Ameliorates Cognitive Impairment Through a Neuroprotective Mechanism Based on Mitochondrial Preservation in Patients With Acute Ischemic Stroke

Acute ischemic stroke (AIS) is a global health burden and cognitive impairment is one of its most serious complication. Adequate interventions for AIS may have the potential to improve cognitive outcomes. In the present study, we selected Erigeron breviscapus (Vaniot) Hand.-Mazz. injection (Dengzhanxixin injection, DZXI), a widely used Chinese herbal injection, in contrast to edaravone as the positive control drug to test its potential to ameliorates neurological and cognitive impairments caused by AIS. We performed a 2-week randomized trial with these two drugs in AIS patients presenting mild to moderate cognitive impairments. Neuropsychological tests and MRI examinations showed that DZXI attenuated the neurological and cognitive impairments of patients and protected the grey matter in specific regions from ischemic damage. Notably, DZXI exerted better effects than edaravone in some neuropsychological tests, probably due to the protective effect of DZXI on grey matter. To explore the therapeutic mechanisms, we carried out an experiment with a middle cerebral artery occlusion rat model. We found that DZXI decreased the infarct volume and increased the survival of neuronal cells in the ischemic penumbra; furthermore, DZXI modulated the mitochondrial respiratory chain process and preserved the mitochondrial structure in the brain tissue. Overall, our data suggested that the administration of DZXI is effective at ameliorating neurological and cognitive impairments in AIS, and the underlying mechanisms are related to the protective effects of DZXI on cerebral neurons and neuronal mitochondria.

Outcome Prediction for Patients With Ischemic Stroke in Acute Care: New Three-Level Model by Eating and Bladder Functions

OBJECTIVE

To develop a new prediction model by combining independence in eating and bladder management functions, and to assess its utility in an acute care setting.

METHODS

Patients with ischemic stroke who were admitted in our acute stroke care unit (n=250) were enrolled in this study. Functional Independence Measure (FIM) scores for eating and bladder management on the initial day of rehabilitative treatment (median, 3 days) were collected as predictive variables. These scores were divided into low (<5) and high (≥5) and categorized as values 0 and 1, respectively. From the simple summation of these two-level model values, we derived a three-level model that categorized the scores as values 0, 1, and 2. The FIM-motor scores at discharge (median, 14 days) were collected as outcome measurements. The three-level model was assessed by observing the distribution patterns of the outcome FIM-motor scores and logistic regression analyses.

RESULTS

The median outcome FIM-motor score was 19 (interquartile range [IQR],13.8-45.3) for the value 0 category (n=14), 66.5 (IQR, 59.5-81.8) for the value 1 category (n=16), and 84 (IQR, 77-89) for the value 2 category (n=95) in the three-level model. Data fitting by logistic regression for FIM-motor scores of 41.3 and 61.4 reached 50% probability of values 1 and 2, respectively.

CONCLUSION

Despite the simplicity of the three-level model, it may be useful for predicting outcomes of patients with ischemic stroke in acute care.

A Voxel-Based Radiographic Analysis Reveals the Biological Character of Proneural-Mesenchymal Transition in Glioblastoma

Introduction: Proneural and mesenchymal subtypes are the most distinct demarcated categories in classification scheme, and there is often a shift from proneural type to mesenchymal subtype in the progression of glioblastoma (GBM). The molecular characters are determined by specific genomic methods, however, the application of radiography in clinical practice remains to be further studied. Here, we studied the topography features of GBM in proneural subtype, and further demonstrated the survival characteristics and proneural-mesenchymal transition (PMT) progression of samples by combining with the imaging variables.

Methods: Data were acquired from The Cancer Imaging Archive (TCIA, http://cancerimagingarchive.net). The radiography image, clinical variables and transcriptome subtype from 223 samples were used in this study. Proneural and mesenchymal subtype on GBM topography based on overlay and Voxel-based lesion-symptom mapping (VLSM) analysis were revealed. Besides, we carried out the comparison of survival analysis and PMT progression in and outside the VLSM-determined area.

Results: The overlay of total GBM and separated image of proneural and mesenchymal subtype revealed a correlation of the two subtypes. By VLSM analysis, proneural subtype was confirmed to be related to left inferior temporal medulla, and no significant voxel was found for mesenchymal subtype. The subsequent comparison between samples in and outside the VLSM-determined area showed difference in overall survival (OS) time, tumor purity, epithelial-mesenchymal transition (EMT) score and clinical variables.

Conclusions: PMT progression was determined by radiography approach. GBM samples in the VLSM-determined area tended to harbor the signature of proneural subtype. This study provides a valuable VLSM-determined area related to the predilection site, prognosis and PMT progression by the association between GBM topography and molecular characters.

Relevance of Porcine Stroke Models to Bridge the Gap from Pre-Clinical Findings to Clinical Implementation

In the search of animal stroke models providing translational advantages for biomedical research, pigs are large mammals with interesting brain characteristics and wide social acceptance. Compared to rodents, pigs have human-like highly gyrencephalic brains. In addition, increasingly through phylogeny, animals have more sophisticated white matter connectivity; thus, ratios of white-to-gray matter in humans and pigs are higher than in rodents. Swine models provide the opportunity to study the effect of stroke with emphasis on white matter damage and neuroanatomical changes in connectivity, and their pathophysiological correlate. In addition, the subarachnoid space surrounding the swine brain resembles that of humans. This allows the accumulation of blood and clots in subarachnoid hemorrhage models mimicking the clinical condition. The clot accumulation has been reported to mediate pathological mechanisms known to contribute to infarct progression and final damage in stroke patients. Importantly, swine allows trustworthy tracking of brain damage evolution using the same non-invasive multimodal imaging sequences used in the clinical practice. Moreover, several models of comorbidities and pathologies usually found in stroke patients have recently been established in swine. We review here ischemic and hemorrhagic stroke models reported so far in pigs. The advantages and limitations of each model are also discussed.