This manuscript will review emerging applications of artificial intelligence, specifically deep learning, and its application to glioblastoma multiforme (GBM), the most common primary malignant brain tumor. Current deep learning approaches, commonly convolutional neural networks (CNNs), that take input data from MR images to grade gliomas (high grade from low grade) and predict overall survival will be shown. There will be more in-depth review of recent articles that have applied different CNNs to predict the genetics of glioma on pre-operative MR images, specifically 1p19q codeletion, MGMT promoter, and IDH mutations, which are important criteria for the diagnosis, treatment management, and prognostication of patients with GBM. Finally, there will be a brief mention of current challenges with DL techniques and their application to image analysis in GBM.

BACKGROUND AND PURPOSE: Coronavirus disease 2019 (COVID-19) appears to be an independent risk factor for stroke. We hypothesize that patients who develop stroke while hospitalized for severe COVID-19 will have higher inflammatory markers and distinct stroke imaging patterns compared with patients positive for COVID-19 with out-of-hospital stroke onset and milder or no COVID-19 symptoms.MATERIALS AND METHODS: This is a retrospective case series of patients positive for COVID-19 on polymerase chain reaction testing with imaging-confirmed stroke treated within a large health care network in New York City and Long Island between March 14 and April 26, 2020. Clinical and laboratory data collected retrospectively included complete blood counts and creatinine, alanine aminotransferase, lactate dehydrogenase, C-reactive protein, ferritin, and D-dimer levels. All CT and MR imaging studies were independently reviewed by 2 neuroradiologists who recorded stroke subtype and patterns of infarction and intracranial hemorrhage. RESULTS:Compared with patients with COVID-19 with outside-of-hospital stroke onset and milder or no COVID-19 symptoms (n ¼ 45, 52.3%), patients with stroke already hospitalized for severe COVID-19 (n ¼ 41, 47.7%) had significantly more frequent infarctions (95.1% versus 73.3%, P ¼ .006), with multivascular distributions (56.4% versus 33.3%, P ¼ .022) and associated hemorrhage (31.7% versus 4.4%, P ¼ .001). Patients with stroke admitted with more severe COVID-19 had significantly higher C-reactive protein and ferritin levels, elevated D-dimer levels, and more frequent lymphopenia and renal and hepatic injury (all, P , .003). CONCLUSIONS:Patients with stroke hospitalized with severe COVID-19 are characterized by higher inflammatory, coagulopathy, and tissue-damage biomarkers, supporting proposed pathogenic mechanisms of hyperinflammation activating a prothrombotic state. Cautious balancing of thrombosis and the risk of hemorrhagic transformation is warranted when considering anticoagulation.

No abstract