An artificial intelligence-based prognostic prediction model for hemorrhagic stroke.

PURPOSE: The prognosis following a hemorrhagic stroke is usually extremely poor. Rating scales have been developed to predict the outcomes of patients with intracerebral hemorrhage (ICH). To date, however, the prognostic prediction models have not included the full range of relevant imaging features. We constructed a clinic-imaging fusion model based on convolutional neural networks (CNN) to predict the short-term prognosis of ICH patients. MATERIALS AND METHODS: This was a multi-center retrospective study, which included 1990 patients with ICH. Two CNN-based deep learning models were constructed to predict the neurofunctional outcomes at discharge; these were validated using a nested 5-fold cross-validation approach. The models' predictive efficiency was compared with the original ICH scale and the ICH grading scale. Poor neurological outcome was defined as a Glasgow Outcome Scale (GOS) score of 1-3. RESULTS: The training and test sets included 1599 and 391 patients, respectively. For the test set, the clinic-imaging fusion model had the highest area under the curve (AUC = 0.903), followed by the imaging-based model (AUC = 0.886), the ICH scale (AUC = 0.777), and finally the ICH grading scale (AUC = 0.747). CONCLUSION: The CNN prognostic prediction model based on neuroimaging features was more effective than the ICH scales in predicting the neurological outcomes of ICH patients at discharge. The CNN model's predictive efficiency slightly improved when clinical data were included.

A machine learning approach for predicting perihematomal edema expansion in patients with intracerebral hemorrhage.

OBJECTIVES: Preventing the expansion of perihematomal edema (PHE) represents a novel strategy for the improvement of neurological outcomes in intracerebral hemorrhage (ICH) patients. Our goal was to predict early and delayed PHE expansion using a machine learning approach. METHODS: We enrolled 550 patients with spontaneous ICH to study early PHE expansion, and 389 patients to study delayed expansion. Two imaging researchers rated the shape and density of hematoma in non-contrast computed tomography (NCCT). We trained a radiological machine learning (ML) model, a radiomics ML model, and a combined ML model, using data from radiomics, traditional imaging, and clinical indicators. We then validated these models on an independent dataset by using a nested 4-fold cross-validation approach. We compared models with respect to their predictive performance, which was assessed using the receiver operating characteristic curve. RESULTS: For both early and delayed PHE expansion, the combined ML model was most predictive (early/delayed AUC values were 0.840/0.705), followed by the radiomics ML model (0.799/0.663), the radiological ML model (0.779/0.631), and the imaging readers (reader 1: 0.668/0.565, reader 2: 0.700/0.617). CONCLUSION: We validated a machine learning approach with high interpretability for the prediction of early and delayed PHE expansion. This new technique may assist clinical practice for the management of neurocritical patients with ICH. KEY POINTS: • This is the first study to use artificial intelligence technology for the prediction of perihematomal edema expansion. • A combined machine learning model, trained on data from radiomics, clinical indicators, and imaging features associated with hematoma expansion, outperformed all other methods.

Validation of perihematomal edema expansion as a new imaging biomarker to predict clinical outcome in patients with intracerebral hemorrhage.

OBJECTIVES: The use of hematoma expansion (HE) in intracerebral hemorrhage (ICH) patients is limited due to its low sensitivity. Perihematomal edema (PHE) has been considered an important marker of secondary brain injury after ICH. Enrolling PHE expansion to redefine traditional ICH expansion merits exploration. MATERIALS AND METHODS: This study analyzed a cohort of patients with spontaneous ICH. The hematoma and PHE were manually segmented. Logistic regression analysis was utilized to identify risk factors for poor outcomes. Receiver operating characteristic curve analysis was performed to calculate the predictive values of PHE expansion and HE. Poor neurological outcome was defined as a modified Rankin Scale score of 4-6 at 90 days. RESULTS: Overall, 223 target patients were enrolled in the study. Multivariable analysis showed the larger PHE expansion is the independent risk factors for poor prognosis. The predictive value of absolute PHE expansion (AUC=0.776, sensitivity=67.9%, specificity=77.0%) was higher than that of absolute HE (AUC=0.573, sensitivity=41.7%, specificity=87.1%) and HE (>6 ml) (AUC=0.594, sensitivity=23.8%, specificity=95.0%). The best cutoff for early absolute/relative PHE expansion resulting in a poor outcome was 5.96 ml and 31%. CONCLUSIONS: Early PHE expansion was associated with a poor outcome, characterized by a better predictive value than HE.

Defining Delayed Perihematomal Edema Expansion in Intracerebral Hemorrhage: Segmentation, Time Course, Risk Factors and Clinical Outcome.

We attempt to generate a definition of delayed perihematomal edema expansion (DPE) and analyze its time course, risk factors, and clinical outcomes. A multi-cohort data was derived from the Chinese Intracranial Hemorrhage Image Database (CICHID). A non-contrast computed tomography (NCCT) -based deep learning model was constructed for fully automated segmentation hematoma and perihematomal edema (PHE). Time course of hematoma and PHE evolution correlated to initial hematoma volume was volumetrically assessed. Predictive values for DPE were calculated through receiver operating characteristic curve analysis and were tested in an independent cohort. Logistic regression analysis was utilized to identify risk factors for DPE formation and poor outcomes. The test cohort's Dice scores of lesion segmentation were 0.877 and 0.642 for hematoma and PHE, respectively. Overall, 1201 patients were enrolled for time-course analysis of ICH evolution. A total of 312 patients were further selected for DPE analysis. Time course analysis showed the growth peak of PHE approximately concentrates in 14 days after onset. The best cutoff for DPE to predict poor outcome was 3.34 mL of absolute PHE expansion from 4-7 days to 8-14 days (AUC=0.784, sensitivity=72.2%, specificity=81.2%), and 3.78 mL of absolute PHE expansion from 8-14 days to 15-21 days (AUC=0.682, sensitivity=59.3%, specificity=92.1%) in the derivation sample. Patients with DPE was associated with worse outcome (OR: 12.340, 95%CI: 6.378-23.873, P<0.01), and the larger initial hematoma volume (OR: 1.021, 95%CI: 1.000-1.043, P=0.049) was the significant risk factor for DPE formation. This study constructed a well-performance deep learning model for automatic segmentations of hematoma and PHE. A new definition of DPE was generated and is confirmed to be related to poor outcomes in ICH. Patients with larger initial hematoma volume have a higher risk of developing DPE formation.

Clinical Grading System, Surgical Outcomes and Prognostic Analysis of Cranial Base Chordomas.

OBJECTIVE: Cranial base chordomas are rare, but their treatment is challenging. Tumor recurrence is still common despite improvements in microsurgical techniques and postoperative radiotherapy. We retrospectively analyzed the course of treatment, overall survival, and recurrence/progression of chordomas over the past 10 years. METHODS: We retrospectively reviewed 50 patients who underwent surgery at Tianjin Huanhu Hospital between 2010 and 2020 and were pathologically diagnosed with chordomas. Tumor resection was performed within the maximum safe range in all patients; the extent of resection was evaluated by imaging; and the incidence of complications, recurrence or progression, and overall survival were assessed. RESULTS: Fifty patients were divided into the low-risk group (LRG) and high-risk group (HRG) based on the cranial chordoma grading system (CCGS). The Karnofsky Performance Scale scores and gross total resection rate of the LRG were significantly higher than those of the HRG (p<0.05). The incidence of complications and mortality in the LRG were lower than those of HRG. The analysis of cumulative survival and cumulative recurrence free survival/progression free survival (RFS/PFS) showed no statistical differences in the extent of resection for survival, recurrence, or progression. Univariate and multivariate analyses showed that Ki-67 was significantly associated with tumor recurrence and was an independent hazard factor (p=0.02). CONCLUSION: The CCGS can help neurosurgeons anticipate surgical outcomes. Pathological results are important in evaluating the possibility of tumor recurrence, and postoperative radiotherapy improves overall survival and RFS/PFS.

Identification of Mir-9 in Glioma Diagnosis and Prognosis.

BACKGROUND: Mir-9 was recognized as a potential tumor suppressor gene or oncogene in varies of diseases; however, its roles in glioma have not been investigated. Our study was to identify the mRNA expression of mir-9 gene in glioma and correlate abnormal versions of microRNAs with clinicopathological features and investigate the impact of mir-9 in glioma prognosis. METHODS: Seventy-one glioma samples, which included 22 grade II, 24 grade III, and 25 glioblastoma tumors of previously untreated patients who underwent surgical excision at Qing Hai Province People's Hospital from 2011 to 2014, were included. In addition, 2 epilepsy patients with normal brain tissue were included as a control group. The expression of mir-9 was examined by RT-PCR in formalin-fixed paraffin embedded primary tissue specimens. The clinicopathologic features, the survival rate of glioma patients were also discussed. The RNA expression of mir-9 and glioma prognosis was evaluated using a Chi-square test, Cox regression model, and GraphPad Prism survival curve analysis. RESULTS: There were 16, 20, 21 cases which showed high expression of mir-9 in grade II and III gliomas and glio-blastoma, 16/22 (72.7%), 20/24 (83.3%), 21/25 (84.0%), respectively. The expression of mir-9 was correlated with the grade of glioma. The mir-9 RNA expression in glioblastoma were higher than grade II and III gliomas (p < 0.05). The higher the grade, the higher the expression, and the difference was significant (p < 0.05), while it was not correlated with patient nationality, gender, or location (p > 0.05). Univariate analysis determined that high expression of mir-9, grade, chemotherapy, radiation therapy, and patient onset age were associated with glioma patient prognosis (p < 0.05). CONCLUSIONS: The expression of mir-9 plays a role in glioma progression, and may be used as a prognostic marker in glioma.

Curcumin accelerates reendothelialization and ameliorates intimal hyperplasia in balloon-injured rat carotid artery via the upregulation of endothelial cell autophagy.

Delayed reendothelialization and intimal hyper-plasia (IH) contribute to the failure of vascular interventions. Curcumin (Cur) has been used for various types of diseases with antioxidant, antiproliferative and anti­inflammatory effects. However, investigations involving the application of Cur in inhibiting IH are limited. The aim of the present study was to evaluate the potential therapeutic effects of Cur and its underlying mechanisms on a rat model of carotid artery (CA) intimal injury. In vitro, an endothelial cell (EC) migration assay was conducted using cultured primary human umbilical vein endothelial cells (HUVECs) that were exposed to Cur. In vivo, CA angioplasty injury was used to generate a rat model of intimal injury. CAs were collected at 3 days, and 1 and 4 weeks after injury, respectively, for western blot analysis and double-immunofluorescence analyses, terminal deoxynucleotidyl transferase-mediated dUTP nick end-labeling staining, oxidative stress indicator analysis and hematoxylin and eosin staining of the neointima. In vivo, Cur significantly enhanced the migration and healing of HUVECs and simultaneously promoted microtubule-associated protein light chain 3-II (LC3-II) expression when HUVECs were subjected to an artificial scratch. In vitro, endangium from the Cur-treated rats exhibited a significantly reduced number of apoptotic ECs and oxidative stress level compared to that of the sham group. In addition, Cur treatment markedly improved quantification of the LC3-II concomitant with the downregulation of p62 in the injured CA. At 1 week following injury, sizable neointimal lesions had developed, although prominent intima thickening was not observed. At 4 weeks, apparent hemadostenosis occurred resulting from the exorbitance IH. Cur treatment markedly reduced the thickness of the neointimal lesion. It is noteworthy that high-dose Cur may have exerted more significant effects than low-dose Cur. Cur can potentially become a therapeutic drug for angiostenosis by imparting a protective effect that accelerates reendothelialization and ameliorates IH and was mediated by its pro-autophagic effect.

Expression and clinical significance of non-phagocytic cell oxidase 2 and 4 after human traumatic brain injury.

The goal of this study was to examine NOX2 and NOX4 expression in clinical samples of patients with traumatic brain injury (TBI), and to explore the correlation of NOX2 and NOX4 expression with the severity of injury, duration of injury, and prognosis. Brain samples of 20 TBI patients within 1 cm of the contusion site were collected and grouped based on duration of injury, Glasgow Coma Scale (GCS) and Glasgow Outcome Scale (GOS), and immunofluorescence staining were performed to examine the expression levels of NOX2 and NOX4 in the neurons and astrocytes. We found that the expression level of NOX2 in neurons and positive rate of NOX2 expression in astrocytes peaked at 12-24 and 6-12 h after injury, respectively. The expression level of NOX4 in neurons peaked at 24-48 h, and the positive rate of NOX4 expression gradually increased with prolonged injury. We also found that the higher the GCS score, the lower the expression levels of NOX2 and NOX4 in neurons was, while higher the GCS score, the lower the positive rate of NOX4 expression in astrocytes was and the higher the GOS grade, the lower the positive rate of expression in astrocytes was. In conclusion, NOX2 and NOX4 expressions significantly increase at an early stage after TBI, and abnormal expressions of NOX2 and NOX4 are correlated to patient prognosis to certain extent.

Alterations in the time course of expression of the Nox family in the brain in a rat experimental cerebral ischemia and reperfusion model: effects of melatonin.

Ischemia-reperfusion (I/R) injury induces the generation of reactive oxygen species (ROS), which results in a poor prognosis for ischemic stroke patients. This study was designed to evaluate the time course of expression of the Nox family, a major source of ROS, and whether melatonin, a potent scavenger of ROS, influences these parameters in a rat model of cerebral I/R caused by middle cerebral artery occlusion (MCAO). After 2-hr occlusion, the filament was withdrawn to allow reperfusion. At 0, 3, 6, 12, 24, and 48 hr after reperfusion, brain tissue samples were obtained for assays. Among the Nox family, the mRNA and protein levels of Nox2 and Nox4 were increased both in the ischemic hemisphere and contralateral counterpart in the experimental I/R rats at 0 hr after reperfusion, peaked at 3 hr, and then returned to the basal level at 24 hr. Double-immunofluorescence staining further confirmed the expressions of Nox2 and Nox4 in three major types of brain cells, including neurons, astrocytes, and endothelial cells. In addition, melatonin (5 mg/kg) or its vehicle was injected intraperitoneally at 0.5 hr before MCAO. Compared with I/R + vehicle group, melatonin pretreatment diminished the increased expression of Nox2 and Nox4, reduced ROS levels, and inhibited cell apoptosis. Our findings suggested that the inhibition of Nox2 and Nox4 expressions by melatonin may essentially contribute to its antioxidant and anti-apoptotic effects during brain I/R.