Deep Learning-Derived High-Level Neuroimaging Features Predict Clinical Outcomes for Large Vessel Occlusion.

Background and Purpose- For patients with large vessel occlusion, neuroimaging biomarkers that evaluate the changes in brain tissue are important for determining the indications for mechanical thrombectomy. In this study, we applied deep learning to derive imaging features from pretreatment diffusion-weighted image data and evaluated the ability of these features in predicting clinical outcomes for patients with large vessel occlusion. Methods- This multicenter retrospective study included patients with anterior circulation large vessel occlusion treated with mechanical thrombectomy between 2013 and 2018. We designed a 2-output deep learning model based on convolutional neural networks (the convolutional neural network model). This model employed encoder-decoder architecture for the ischemic lesion segmentation, which automatically extracted high-level feature maps in its middle layers, and used its information to predict the clinical outcome. Its performance was internally validated with 5-fold cross-validation, externally validated, and the results compared with those from the standard neuroimaging biomarkers Alberta Stroke Program Early CT Score and ischemic core volume. The prediction target was a good clinical outcome, defined as a modified Rankin Scale score at 90-day follow-up of 0 to 2. Results- The derivation cohort included 250 patients, and the validation cohort included 74 patients. The convolutional neural network model showed the highest area under the receiver operating characteristic curve: 0.81±0.06 compared with 0.63±0.05 and 0.64±0.05 for the Alberta Stroke Program Early CT Score and ischemic core volume models, respectively. In the external validation, the area under the curve for the convolutional neural network model was significantly superior to those for the other 2 models. Conclusions- Compared with the standard neuroimaging biomarkers, our deep learning model derived a greater amount of prognostic information from pretreatment neuroimaging data. Although a confirmatory prospective evaluation is needed, the high-level imaging features derived by deep learning may offer an effective prognostic imaging biomarker.

Cerebral Venous Thrombosis after Red Blood Cell Transfusion for the Treatment of Iron Deficiency Anemia.

A 57-year-old male presented with generalized seizure who received red blood cell (RBC) transfusion for the treatment of iron deficiency anemia (IDA). Neuroradiological findings revealed cerebral venous thrombosis (CVT) on the left frontal vein. He received anticoagulants, anticonvulsants, and iron supplements. He discharged without any neurological deficit. It should be noted that RBC transfusion might increase the risk of CVT in patients with IDA.

Predicting Clinical Outcomes of Large Vessel Occlusion Before Mechanical Thrombectomy Using Machine Learning.

Background and Purpose- The clinical course of acute ischemic stroke with large vessel occlusion (LVO) is a multifactorial process with various prognostic factors. We aimed to model this process with machine learning and predict the long-term clinical outcome of LVO before endovascular treatment and to compare our method with previously developed pretreatment scoring methods. Methods- The derivation cohort included 387 LVO patients, and the external validation cohort included 115 LVO patients with anterior circulation who were treated with mechanical thrombectomy. The statistical model with logistic regression without regularization and machine learning algorithms, such as regularized logistic regression, linear support vector machine, and random forest, were used to predict good clinical outcome (modified Rankin Scale score of 0-2 at 90 days) with standard and multiple pretreatment clinical variables. Five previously reported pretreatment scoring methods (the Pittsburgh Response to Endovascular Therapy score, the Stroke Prognostication Using Age and National Institutes of Health Stroke Scale index, the Totaled Health Risks in Vascular Events score, the Houston Intra-Arterial Therapy score, and the Houston Intra-Arterial Therapy 2 score) were compared with these models for the area under the receiver operating characteristic curve. Results- The area under the receiver operating characteristic curve of random forest, which was the worst among the machine learning algorithms, was significantly higher than those of the standard statistical model and the best model among the previously reported pretreatment scoring methods in the derivation (the area under the receiver operating characteristic curve were 0.85±0.07 for random forest, 0.78±0.08 for logistic regression without regularization, and 0.77±0.09 for Stroke Prognostication using Age and National Institutes of Health Stroke Scale) and validation cohorts (the area under the receiver operating characteristic curve were 0.87±0.01 for random forest, 0.56±0.07 for logistic regression without regularization, and 0.83±0.00 for Pittsburgh Response to Endovascular Therapy). Conclusions- Machine learning methods with multiple pretreatment clinical variables can predict clinical outcomes of patients with anterior circulation LVO who undergo mechanical thrombectomy more accurately than previously developed pretreatment scoring methods.

Mobility of an in-stent plaque protrusion detected with carotid duplex during the carotid artery stenting procedure: a case report.

In-stent plaque protrusion (ISP) is a predictor for ischemic complications in carotid artery stenting (CAS). Because its mobility would further increase the distal embolization risk, the intraprocedural detection of mobility is important but has not yet been reported. We present an 87-year-old male with symptomatic right carotid artery stenosis who underwent CAS. Intravascular ultrasound (IVUS) after stent deployment revealed a small ISP, and simultaneous extravascular carotid duplex disclosed its mobility. After coverage of the ISP with stent-in-stent implantation, no remaining protrusion was observed with either IVUS or carotid duplex. The patient experienced no periprocedural strokes. Carotid duplex is suited to evaluate targets' movement because of its excellent temporal resolution. Carotid duplex has potential value for intraprocedural risk estimation of ISP.

Multiple extra-ischemic hemorrhages following intravenous thrombolysis in a patient with Trousseau syndrome: case study.

BACKGROUND: Intracerebral hemorrhage is the most feared complication of intravenous thrombolysis for acute ischemic stroke. Such hemorrhage usually occurs within or at the margin of ischemic or manifestly infarcted brain tissue. A patient with Trousseau syndrome who developed multiple extra-ischemic hemorrhages following intravenous thrombolysis is described. CASE DESCRIPTION: An 80-year-old Japanese man with no other underlying disease was diagnosed with unresectable advanced lung cancer (stage IV) without brain metastasis and had not yet been treated. The patient suddenly presented with disturbance of consciousness, right hemiplegia, and total aphasia, and was admitted to our hospital. Magnetic resonance imaging revealed acute cerebral infarction extending from the basal ganglia to the corona radiata of the left cerebrum and multiple small areas of bilateral cerebral cortices. Cardiogenic cerebral embolism was considered among the differential diagnoses, but the brain natriuretic peptide level was within the normal range, and no arrhythmias such as atrial fibrillation were observed. With no other causes, the patient was diagnosed with Trousseau syndrome due to hypercoagulability associated with the advanced lung cancer. The patient received intravenous tissue plasminogen activator (t-PA) at 96 minutes after onset of symptoms. His symptoms partially improved, but they suddenly deteriorated at 84 minutes after the thrombolysis. A computed tomography (CT) scan immediately after the neurological deterioration revealed a subcortical hemorrhage in the left occipital lobe. A repeat CT scan the day after onset showed enlargement of the left occipital hemorrhage and two new subcortical hemorrhages in the right frontal and right temporal lobes. These hemorrhages were located in areas remote from the acute ischemic lesions. CONCLUSION: To the best of our knowledge, this is the first reported case of multiple extra-ischemic hemorrhages following intravenous thrombolysis in a patient with Trousseau syndrome. The course of this case suggests that intravenous t-PA administration for acute ischemic stroke with Trousseau syndrome may be associated with a higher risk of intracranial hemorrhage.

Thrombus Removal by Acute-phase Endovascular Reperfusion Therapy to Treat Cerebral Embolism Caused by Thrombus in the Pulmonary Vein Stump after Left Upper Pulmonary Lobectomy: Case Report.

Thrombus formation in a pulmonary vein stump after pulmonary lobectomy is extremely rare, but can trigger cerebral embolism of unknown cause. We encountered a case of cerebral embolism in a 58-year-old man 2 days after left upper lobectomy. Since intravenous administration of recombinant tissue plasminogen activator was contraindicated, thrombus removal by endovascular reperfusion therapy was performed. Cerebral angiography showed left internal carotid artery occlusion. Thrombus removal using a retrieval device was performed and complete recanalization of the left internal carotid artery was obtained. Although blood abnormalities or arrhythmia such as atrial fibrillation were not observed, thrombus in the left upper pulmonary vein stump was detected with contrast-enhanced computed tomography of the body trunk, which was therefore considered as the source of cerebral embolism. The patient is continuing on anticoagulant therapy to prevent embolism recurrence caused by thrombus formation in the pulmonary vein resection stump. To the best of our knowledge, this is the first report of thrombus removal by acute-phase endovascular reperfusion therapy to treat cerebral embolism likely caused by thrombus formation in the pulmonary vein stump after left upper lobectomy. When cerebral embolism of unknown cause develops after left upper lobectomy, thrombus formation in the pulmonary vein stump should be considered among the differential diagnoses. For acute-phase onset of cerebral embolism after pulmonary lobectomy, thrombus removal by endovascular reperfusion therapy may be considered as one of the therapies.

[Endovascular coil embolization for a ruptured distal anterior choroidal artery aneurysm showing definite short-term enlargement: a case report].

Distal anterior choroidal artery aneurysms are quite rare, and appropriate treatment timing and methods remain unclear. Direct surgery of these aneurysms is difficult due to their deep location, small size, and angioarchitecture;however, pseudoaneurysms might disappear spontaneously with conservative treatment. A 65-year-old man with a history of hypertension was admitted to our hospital with a 5-day history of sudden headache and nausea. Computed tomography revealed an intraventricular hematoma located mainly in the right lateral ventricle. Cerebral angiography 7 days after onset revealed a right distal anterior choroidal artery aneurysm and proximal right middle cerebral artery occlusion caused by atherosclerotic changes. Endovascular coil embolization was performed under general anesthesia 14 days after onset. Preoperative angiography demonstrated definite enlargement of the aneurysm and stasis of the contrast agent in the aneurysm in the venous phase. Detachable platinum coils were delivered into the aneurysm and parent artery. The patient was discharged neurologically intact after the procedure. Follow-up angiography 3 months after coil embolization showed complete occlusion of the aneurysm. In recent years, endovascular surgery has emerged as a less invasive treatment option. Early treatment should be considered for patients with ruptured distal anterior choroidal artery aneurysm because these aneurysms might grow and re-rupture in the short term.

Small Onodi cell mucocele causing chronic optic neuropathy: case report.

A 45-year-old woman presented with a rare case of small isolated Onodi cell mucocele manifesting as unilateral chronic optic neuropathy. The patient complained of gradual visual disturbance of her left eye with no other neurological abnormalities. Neuroimaging examinations showed a small cystic lesion located in the left posterior ethmoid extending into the anterior clinoid process. Surgery via the pterional epidural approach revealed a cystic lesion containing mucinous fluid isolated in the anterior clinoid process. Total removal of the lesion combined with unroofing of the left optic canal resulted in relief of the patient's symptom. Histological examination showed the lesion was a mucocele. Based on the radiological and surgical findings together with the histological examination, the final diagnosis was Onodi cell mucocele. Such small Onodi cell mucocele can cause optic neuropathy but is extremely unusual.

The Rho-mDia1 pathway regulates cell polarity and focal adhesion turnover in migrating cells through mobilizing Apc and c-Src.

Directed cell migration requires cell polarization and adhesion turnover, in which the actin cytoskeleton and microtubules work critically. The Rho GTPases induce specific types of actin cytoskeleton and regulate microtubule dynamics. In migrating cells, Cdc42 regulates cell polarity and Rac works in membrane protrusion. However, the role of Rho in migration is little known. Rho acts on two major effectors, ROCK and mDia1, among which mDia1 produces straight actin filaments and aligns microtubules. Here we depleted mDia1 by RNA interference and found that mDia1 depletion impaired directed migration of rat C6 glioma cells by inhibiting both cell polarization and adhesion turnover. Apc and active Cdc42, which work together for cell polarization, localized in the front of migrating cells, while active c-Src, which regulates adhesion turnover, localized in focal adhesions. mDia1 depletion impaired localization of these molecules at their respective sites. Conversely, expression of active mDia1 facilitated microtubule-dependent accumulation of Apc and active Cdc42 in the polar ends of the cells and actin-dependent recruitment of c-Src in adhesions. Thus, the Rho-mDia1 pathway regulates polarization and adhesion turnover by aligning microtubules and actin filaments and delivering Apc/Cdc42 and c-Src to their respective sites of action.

Quantitative analysis of topoisomerase IIalpha to rapidly evaluate cell proliferation in brain tumors.

Immunohistochemical cell proliferation analyses have come into wide use for evaluation of tumor malignancy. Topoisomerase IIalpha (topo IIalpha), an essential nuclear enzyme, has been known to have cell cycle coupled expression. We here show the usefulness of quantitative analysis of topo IIalpha mRNA to rapidly evaluate cell proliferation in brain tumors. A protocol to quantify topo IIalpha mRNA was developed with a real-time RT-PCR. It took only 3 h to quantify from a specimen. A total of 28 brain tumors were analyzed, and the level of topo IIalpha mRNA was significantly correlated with its immuno-staining index (p<0.0001, r=0.9077). Furthermore, it sharply detected that topo IIalpha mRNA decreased in growth-inhibited glioma cell. These results support that topo IIalpha mRNA may be a good and rapid indicator to evaluate cell proliferate potential in brain tumors.