Automatic segmentation of hemispheric CSF on MRI using deep learning: Quantifying cerebral edema following large hemispheric infarction.

Background and objective: Cerebral edema (CED) is a serious complication of acute ischemic stroke (AIS), especially in patients with large hemispheric infarction (LHI). Herein, a deep learning-based approach is implemented to extract CSF from T2-Weighted Imaging (T2WI) and evaluate the relationship between quantified cerebrospinal fluid and outcomes. Methods: Patients with acute LHI who underwent magnetic resonance imaging (MRI) were included. We used a deep learning algorithm to segment the CSF from T2WI. The hemispheric CSF ratio was calculated to evaluate its relationship with the degree of brain edema and prognosis in patients with LHI. Results: For the 93 included patients, the left and right cerebrospinal fluid regions were automatically extracted with a mean Dice similarity coefficient of 0.830. Receiver operating characteristic analysis indicated that hemispheric CSF ratio was an accurate marker for qualitative severe cerebral edema (area under receiver-operating-characteristic curve 0.867 [95% CI, 0.781-0.929]). Multivariate logistic regression analysis of functional prognosis showed that previous stroke (OR = 5.229, 95% CI 1.013-26.984), ASPECT≤6 (OR = 13.208, 95% CI 1.136-153.540) and low hemispheric CSF ratio (OR = 0.966, 95% CI 0.937-0.997) were significantly associated with higher chances for unfavorable functional outcome in patients with LHI. Conclusions: Automated assessment of CSF volume provides an objective biomarker of cerebral edema that can be leveraged to quantify the degree of cerebral edema and confirm its predictive effect on outcomes after LHI.

Brain death-like changes: A case report of anti-GQ1b antibody syndrome.

RATIONALE: Anti-GQ1b antibody syndrome originates in the peripheral nervous system or the central nervous system. Various overlapping syndromes and atypical manifestations are also seen in clinic. It is rare to see multiple sites (almost all targets) of central and peripheral damage at the same time, resulting in brain death-like changes in patient. PATIENT CONCERNS: This is a 55-year-old female patient, with a history of prodromal upper respiratory tract infection, began with headache, quickly developed body paralysis, eye paralysis, disturbance of consciousness, apnea, and tested positive for anti-GQ1b antibody. The patient is diagnosed clearly, the disease progresses rapidly, and almost all GQ1b sites in the central nervous system and peripheral nervous system are involved, which is rare. DIAGNOSES: Anti-GQ1b antibody syndrome. INTERVENTIONS AND OUTCOMES: The patient was treated with tracheal intubation, ventilator assisted breathing, and immunoglobulin. The patient recovered quickly and was discharged after about 30 days in hospital. LESSONS: The concept of anti-GQ1b antibody syndrome is not only beneficial for clinical diagnosis, but also beneficial for understanding the continuous disease spectrum with the same etiology and different clinical manifestations. The pathogenesis of each subtype has not been fully defined. There are mild patients with isolated syndromes and severe patients with multiple subtypes overlapping. Encounter severe patients but also active response, the general prognosis is good.

Brain symmetry index predicts 3-month mortality in patients with acute large hemispheric infarction.

Quantitative electroencephalography data are helpful to predict outcomes of cerebral infarction patients. The study was performed to evaluate the value of brain symmetry index by quantitative electroencephalography in predicting 3-month mortality of large hemispheric infarction. We studied a prospective, consecutive series of patients with large supratentorial cerebral infarction confirmed within 3 days from the onset in 2 intensive care units from August 2017 to February 2020. The electroencephalography was recorded once admission. The brain symmetry index (BSI) which is divided into BSIfast and BSIslow were calculated for each electrodes pair. The outcome was mortality at 3 months after the onset. A total of 38 patients were included. The subjects were divided into the mortality group (15 patients) and survival group (23 patients). Of the BSIfast and BSIslow at each electrodes pair, higher BSIfastC3-C4, higher BSIslowC3-C4, and higher BSIslowO1-O2 were noticed in the mortality group than that in the survival group at 3 months (P = .001; P = .010; P = .009). Multivariable analysis indicated that BSIfastC3-C4 was an independent predictor of 3-month mortality (odds ratio = 1.059, 95%CI 1.003, 1.119, P = .039). BSIfastC3-C4 could significant predict 3-month mortality (area under curve = 0.805, P = .005). And when we combined BSIfastC3-C4, Glasgow Coma Scale and infarct volume together to predict the 3-month mortality, the predicted value increased (area under curve = 0.840, P = .002). BSIfastC3-C4 could independently predict the 3-month mortality of large hemispheric infarction. The combination marker which includes Glasgow Coma Scale, infarct volume, and BSIfastC3-C4 has a better diagnostic value. Further clinical trials with a large sample size are still needed.

Quantitative EEG parameters can improve the predictive value of the non-traumatic neurological ICU patient prognosis through the machine learning method.

Background: Better outcome prediction could assist in reliable classification of the illnesses in neurological intensive care unit (ICU) severity to support clinical decision-making. We developed a multifactorial model including quantitative electroencephalography (QEEG) parameters for outcome prediction of patients in neurological ICU. Methods: We retrospectively analyzed neurological ICU patients from November 2018 to November 2021. We used 3-month mortality as the outcome. Prediction models were created using a linear discriminant analysis (LDA) based on QEEG parameters, APACHEII score, and clinically relevant features. Additionally, we compared our best models with APACHEII score and Glasgow Coma Scale (GCS). The DeLong test was carried out to compare the ROC curves in different models. Results: A total of 110 patients were included and divided into a training set (n=80) and a validation set (n = 30). The best performing model had an AUC of 0.85 in the training set and an AUC of 0.82 in the validation set, which were better than that of GCS (training set 0.64, validation set 0.61). Models in which we selected only the 4 best QEEG parameters had an AUC of 0.77 in the training set and an AUC of 0.71 in the validation set, which were similar to that of APACHEII (training set 0.75, validation set 0.73). The models also identified the relative importance of each feature. Conclusion: Multifactorial machine learning models using QEEG parameters, clinical data, and APACHEII score have a better potential to predict 3-month mortality in non-traumatic patients in neurological ICU.

Continuous Quantitative Electroencephalogram (EEG) Monitoring for Early Detection of Brain Herniation in Large Hemispheric Infarction (LHI): A Case Report.

BACKGROUND: Computer-assisted electroencephalography (EEG) systems may improve the likelihood of detecting abnormal EEGs in adult patients with severe disease. CASE PRESENTATION: We implemented long-range EEG monitoring in a patient with large hemispheric infarction (LHI) and explored its real-time changes in reflecting the patient's brain function. The bands of Alpha, Beta, Delta, Theta, DAR (Delta/Alpha), DTABR (Delta+Theta/Alpha+Beta), and brain symmetry index (BSI) were calculated as a ratio of total power. The test results showed that this patient presents a progressive worsening trend and developed brain herniation. The sigh at the electrophysiological level of brain herniation could be seen 6 h in advance based on the quantitative EEG (QEEG) parameters test. We calculated QEEG at both C3 and C4, electrode locations simultaneously, and the results showed that the trend of QEEG at both electrodes was consistent with the global, affected, and unaffected side. CONCLUSIONS: QEEG parameters can reflect the trend of LHI patients in real-time and may predict the occurrence of LHI brain herniation. For LHI patients, monitoring with fewer EEG electrodes can be tried to predict the changes in conditions.

The change index of quantitative electroencephalography for evaluating the prognosis of large hemispheric infarction.

A growing number of studies have demonstrated the role of quantitative electroencephalography in assessing brain function in neuro-intensive care units. Still, few studies have examined patients with large hemisphere infarction. Thirty patients with large hemisphere infarction were included in this preliminary study, and the patients were divided into the death group (twelve patients) and survival group (eighteen patients). Electroencephalography monitored the patients, and a computerized tomography inspection was performed. The quantitative electroencephalography of the alpha-beta/delta-theta ratio change index was calculated and used to predict the prognosis of early large hemisphere infarction patients. The relationship between three months modified Rankin Scale, and alpha-beta/delta-theta ratio change index was analyzed. The death group had negative changes for alpha-beta/delta-theta ratio change index (-0.0140 ± 0.0193), while there was an opposite trend in the survival group, the median is 0.004 (-0.0067, 0.0137). The death group's brain function decreased more severely and rapidly than the survival group (P = 0.004). The highest diagnostic value (AUC value 0.815, P < 0.001) was observed when the alpha-beta/delta-theta ratio change index dropped and exceeded -0.008. The area under the GCS curve was 0.674, but its predictive ability was low (P = 0.094). The correlation analysis result showed that the 3-month modified Rankin Scale was negatively correlated with the alpha-beta/delta-theta ratio change index (r = -0.489, P = 0.006). The alpha-beta/delta-theta ratio change index is considered an indicator for predicting the prognosis of large hemisphere infarction. Therefore, the alpha-beta/delta-theta ratio change index may be a reliable quantitative EEG parameter that predicts the early prognosis of patients with acute large hemispheric infarction.

[Changes of p38 MAPK and nuclear factor-kappa B in lung tissue of acute paraquat poisoned rats].

OBJECTIVE: To investigate NF-kappaB activity and the expression of phosphorylated p38 MAPK protein in lung tissue of acute paraquat poisoned rats and the effect of MT. METHODS: One hundred and twenty-eight Sprague-Dawley (SD) rats were randomly divided into three experimental groups: poisoned group, MT group and control group. On the 1st, the 3rd, the 7th and the 14th day after exposure, levels of malondialdehyde (MDA) in serum were detected, NF-kappaB activity in the lung tissues was assessed by electrophoresis mobility shift assay (EMSA), the expression of the phosphorylated p38 MAPK was evaluated by Western blot method, the lung pathological changes of rats were observed. RESULTS: The level of malondialdehyde (MDA) in serum increased significantly in poisoned group on the 1st day (4.45 +/- 1.23), the 3rd day (3.77 +/- 1.12) and the 7th day (2.84 +/- 0.96) nmol/ml compared with that in control group (1.36 +/- 0.52) nmol/ml (P < 0.01). There was a significant decrease in MT group on the 1st day (2.68 +/- 0.85), the 3rd day (1.97 +/- 0.74) and the 7th day (1.53 +/- 0.62) nmol/ml compared with poisoned group (P < 0.05). The expression of the phosphorylated p38 MAPK and NF-kappaB activity in lung tissue of poisoned group significantly increased compared with control group (P < 0.01). There was a significant decrease in NF-kappaB activity and expression of the phosphorylated p38 MAPK in the lung tissues in MT group compared with poisoned group (P < 0.05). CONCLUSION: NF-kappaB and p38 MAPK could play an important role in lung injury of poisoned rats. MT may inhibit the expression of NF-kappaB and phosphorylated p38 MAPK, and therefore might have the therapeutical effect on acute paraquat poisoning.