Infarct volume and outcome of cerebral ischaemia, a systematic review and meta-analysis

Artificial Intelligence and Machine Learning Applications in Critically Ill Brain Injured Patients

The utilization of Artificial Intelligence (AI) and Machine Learning (ML) is paving the way for significant strides in patient diagnosis, treatment, and prognostication in neurocritical care. These technologies offer the potential to unravel complex patterns within vast datasets ranging from vast clinical data and EEG (electroencephalogram) readings to advanced cerebral imaging facilitating a more nuanced understanding of patient conditions. Despite their promise, the implementation of AI and ML faces substantial hurdles. Historical biases within training data, the challenge of interpreting multifaceted data streams, and the "black box" nature of ML algorithms present barriers to widespread clinical adoption. Moreover, ethical considerations around data privacy and the need for transparent, explainable models remain paramount to ensure trust and efficacy in clinical decision-making.This article reflects on the emergence of AI and ML as integral tools in neurocritical care, discussing their roles from the perspective of both their scientific promise and the associated challenges. We underscore the importance of extensive validation in diverse clinical settings to ensure the generalizability of ML models, particularly considering their potential to inform critical medical decisions such as withdrawal of life-sustaining therapies. Advancement in computational capabilities is essential for implementing ML in clinical settings, allowing for real-time analysis and decision support at the point of care. As AI and ML are poised to become commonplace in clinical practice, it is incumbent upon health care professionals to understand and oversee these technologies, ensuring they adhere to the highest safety standards and contribute to the realization of personalized medicine. This engagement will be pivotal in integrating AI and ML into patient care, optimizing outcomes in neurocritical care through informed and data-driven decision-making.

Neuroimaging markers of patient-reported outcome measures in acute ischemic stroke

Objectives

To determine the relationship between patient-reported outcome measures (PROMs) and volumetric imaging markers in acute ischemic stroke (AIS).

Patients and Methods

Patients presenting at Massachusetts General Hospital between February 14, 2017 and February 5, 2020 with a confirmed AIS by MRI were eligible and underwent a telephone interview including PROM-10 questionnaires 3-15 months after stroke. White matter hyperintensity (VWMH) and brain volumes (VBrain) were automatically determined using admission clinical MRI. Stroke lesions were manually segmented and volumes calculated (VLesion). Multivariable and ordinal regression analyses were performed to identify associations between global and PROM-10 subscores with brain volumetrics and clinical variables.

Results

Utilizing data from 167 patients (mean age: 64.7; 41.9% female), higher VWMH was associated with worse global physical (β=-0.6), global mental (β=-0.65), physical health (OR=0.68), social satisfaction (OR=0.66), fatigue (OR=0.69) and social activities (OR=0.59) scores. Higher VLesion was associated with poorer global mental (β=-0.79), mental health (OR=0.68), physical (OR=0.66) and social activities (OR=0.55), and emotional distress (OR=0.68) scores. Higher VBrain was linked to better global mental (β=0.93), global physical (β=0.79), mental health (OR=1.54) and physical activities (OR=1.72) scores.

Conclusions

Neuroimaging biomarkers were significantly associated with PROMs, where higher VWMH and VLesion led to worse outcome, while higher VBrain was protective. The inclusion of neuroimaging analyses and PROMs in routine assessment provides enhanced understanding of post-stroke outcomes.

Using ex vivo arterial injection and dissection to assess white matter vascularization

Advances in the techniques for assessing human cerebral white matter have recently contributed to greater attention to structural connectivity. Yet, little is known about the vascularization of most white matter fasciculi and the fascicular composition of the vascular territories. This paper presents an original method to label the arterial supply of macroscopic white matter fasciculi based on a standardized protocol for post-mortem injection of colored material into main cerebral arteries combined with a novel fiber dissection technique. Twelve whole human cerebral hemispheres obtained post-mortem were included. A detailed description of every step, from obtaining the specimen to image acquisition of its dissection, is provided. Injection and dissection were reproducible and manageable without any sophisticated equipment. They successfully showed the arterial supply of the dissected fasciculi. In addition, we discuss the challenges we faced and overcame during the development of the presented method, highlight its originality. Henceforth, this innovative method serves as a tool to provide a precise anatomical description of the vascularization of the main white matter tracts.

Diagnostic value and mechanism of plasma S100A1 protein in acute ischemic stroke: a prospective and observational study

Background

Plasma S100A1 protein is a novel inflammatory biomarker associated with acute myocardial infarction and neurodegenerative disease's pathophysiological mechanisms. This study aimed to determine the levels of this protein in patients with acute ischemic stroke early in the disease progression and to investigate its role in the pathogenesis of acute ischemic stroke.

Methods

A total of 192 participants from hospital stroke centers were collected for the study. Clinically pertinent data were recorded. The volume of the cerebral infarction was calculated according to the Pullicino formula. Multivariate logistic regression analysis was used to select independent influences. ROC curve was used to analyze the diagnostic value of AIS and TIA. The correlation between S100A1, NF-κB p65, and IL-6 levels and cerebral infarction volume was detected by Pearson correlation analysis.

Results

There were statistically significant differences in S100A1, NF-κB p65, and IL-6 among the AIS,TIA, and PE groups (S100A1, [230.96 ± 39.37] vs [185.85 ± 43.24] vs [181.47 ± 27.39], P < 0.001; NF-κB p65, [3.99 ± 0.65] vs [3.58 ± 0.74] vs [3.51 ± 0.99], P = 0.001; IL-6, [13.32 ± 1.57] vs [11.61 ± 1.67] vs [11.42 ± 2.34], P < 0.001). Multivariate logistic regression analysis showed that S100A1 might be an independent predictive factor for the diagnosis of disease (P < 0.001). The AUC of S100A1 for diagnosis of AIS was 0.818 (P < 0.001, 95% CI [0.749-0.887], cut off 181.03, Jmax 0.578, Se 95.0%, Sp 62.7%). The AUC of S100A1 for diagnosis of TIA was 0.720 (P = 0.001, 95% CI [0.592-0.848], cut off 150.14, Jmax 0.442, Se 50.0%, Sp 94.2%). There were statistically significant differences in S100A1, NF-κB p65, and IL-6 among the SCI,MCI, and LCI groups (S100A1, [223.98 ± 40.21] vs [225.42 ± 30.92] vs [254.25 ± 37.07], P = 0.001; NF-κB p65, [3.88 ± 0.66] vs [3.85 ± 0.64] vs [4.41 ± 0.45], P < 0.001; IL-6, [13.27 ± 1.65] vs [12.77 ± 1.31] vs [14.00 ± 1.40], P = 0.007). Plasma S100A1, NF-κB p65, and IL-6 were significantly different from cerebral infarction volume (S100A1, r = 0.259, P = 0.002; NF-κB p65, r = 0.316, P < 0.001; IL-6, r = 0.177, P = 0.036). There was a positive correlation between plasma S100A1 and IL-6 with statistical significance (R = 0.353, P < 0.001). There was no significant positive correlation between plasma S100A1 and NF-κB p65 (R < 0.3), but there was statistical significance (R = 0.290, P < 0.001). There was a positive correlation between IL-6 and NF-κB p65 with statistical significance (R = 0.313, P < 0.001).

Conclusion

S100A1 might have a better diagnostic efficacy for AIS and TIA. S100A1 was associated with infarct volume in AIS, and its level reflected the severity of acute cerebral infarction to a certain extent. There was a correlation between S100A1 and IL-6 and NF-κB p65, and it was reasonable to speculate that this protein might mediate the inflammatory response through the NF-κB pathway during the pathophysiology of AIS.

Follow-Up Infarct Volume Prediction by CTP-Based Hypoperfusion Index, and the Discrepancy between Small Follow-Up Infarct Volume and Poor Functional Outcome-A Multicenter Study

(1) Background: Follow-up infarct volume (FIV) may have implications for prognostication in acute ischemic stroke patients. Factors predicting the discrepancy between FIV and 90-day outcomes are poorly understood. We aimed to develop a comprehensive predictive model of FIV and explore factors associated with the discrepancy. (2) Methods: Patients with acute anterior circulation large vessel occlusion were included. Baseline clinical and CT features were extracted and analyzed, including the CTP-based hypoperfusion index (HI) and the NCCT-based e-ASPECT, measured by automated software. FIV was assessed on follow-up NCCT at 3−7 days. Multiple linear regression was used to construct the predictive model. Subgroup analysis was performed to explore factors associated with poor outcomes (90-mRS scores 3−6) in small FIV (<70 mL). (3) Results: There were 170 patients included. Baseline e-ASPECT, infarct core volume, hypoperfusion volume, HI, baseline international normalized ratio, and successful recanalization were associated with FIV and included in constructing the predictive model. Baseline NIHSS, baseline hypertension, stroke history, and current tobacco use were associated with poor outcomes in small FIV. (4) Conclusions: A comprehensive predictive model (including HI) of FIV was constructed. We also emphasized the importance of hypertension and smoking status at baseline for the functional outcomes in patients with a small FIV.

ISLES 2022: A multi-center magnetic resonance imaging stroke lesion segmentation dataset

Magnetic resonance imaging (MRI) is an important imaging modality in stroke. Computer based automated medical image processing is increasingly finding its way into clinical routine. The Ischemic Stroke Lesion Segmentation (ISLES) challenge is a continuous effort to develop and identify benchmark methods for acute and sub-acute ischemic stroke lesion segmentation. Here we introduce an expert-annotated, multicenter MRI dataset for segmentation of acute to subacute stroke lesions ( https://doi.org/10.5281/zenodo.7153326 ). This dataset comprises 400 multi-vendor MRI cases with high variability in stroke lesion size, quantity and location. It is split into a training dataset of n = 250 and a test dataset of n = 150. All training data is publicly available. The test dataset will be used for model validation only and will not be released to the public. This dataset serves as the foundation of the ISLES 2022 challenge ( https://www.isles-challenge.org/ ) with the goal of finding algorithmic methods to enable the development and benchmarking of automatic, robust and accurate segmentation methods for ischemic stroke.

Glymphatic Dysfunction in Patients With Ischemic Stroke

Objectives: Rodent experiments have provided some insight into the changes of glymphatic function in ischemic stroke. The diffusion tensor image analysis along the perivascular space (DTI-ALPS) method offers an opportunity for the noninvasive investigation of the glymphatic system in patients with ischemic stroke. We aimed to investigate the changes of glymphatic function in ischemic stroke and the factors associated with the changes. Materials and Methods: A total of 50 patients (mean age 56.7 years; 30 men) and 44 normal subjects (mean age 53.3 years; 23 men) who had preoperative diffusion-tensor imaging for calculation of the analysis along the perivascular space (ALPS) index were retrospectively included. Information collected from each patient included sex, age, time since stroke onset, infarct location, hemorrhagic change, infarct volume, infarct apparent diffusion coefficient (ADC), infarct fractional anisotropy (FA), and ALPS index of both hemispheres. Interhemispheric differences in ALPS index (infarct side vs. contralateral normal side) were assessed with a paired t-test in all patients. ALPS index was normalized by calculating ALPS ratios (right-to-left and left-to-right) for comparisons between patients and normal subjects. Comparisons of ALPS ratios between patients and normal subjects were performed using analysis of covariance with adjustments for age and sex. Linear regression analyses were performed to identify factors associated with the ALPS index. Results: In patients, the mean ALPS index ipsilateral to infarct was 1.162 ± 0.126, significantly lower (P < 0.001) than that of the contralateral side (1.335 ± 0.160). The right-to-left ALPS index ratio of patients with right cerebral infarct was 0.84 ± 0.08, significantly lower (P < 0.001) than that of normal subjects (0.95 ± 0.07). The left-to-right ALPS ratio of patients with left cerebral infarct was 0.92 ± 0.09, significantly (P < 0.001) lower than that of normal subjects (1.05 ± 0.08). On multiple linear regression analysis, time since stroke onset (β = 0.794, P < 0.001) was the only factor associated with the ALPS index. Conclusion: The ALPS index showed lower values in ischemic stroke suggesting impaired glymphatic function. Following initial impairment, the ALPS index increased with the time since stroke onset, which is suggestive of glymphatic function recovery.