Leveraging artificial intelligence in ischemic stroke imaging.

Artificial intelligence (AI) is having a disruptive and transformative effect on clinical medicine. Prompt clinical diagnosis and imaging are critical for minimizing the morbidity and mortality associated with ischemic strokes. Clinicians must understand the current strengths and limitations of AI to provide optimal patient care. Ischemic stroke is one of the medical fields that have been extensively evaluated by artificial intelligence. Presented herein is a review of artificial intelligence applied to clinical management of stroke, geared toward clinicians. In this review, we explain the basic concept of AI and machine learning. This review is without coding and mathematical details and targets the clinicians involved in stroke management without any computer or mathematics' background. Here the AI application in ischemic stroke is summarized and classified into stroke imaging (automated diagnosis of brain infarction, automated ASPECT score calculation, infarction segmentation), prognosis prediction, and patients' selection for treatment.

The Role of Machine Learning and Radiomics for Treatment Response Prediction in Idiopathic Normal Pressure Hydrocephalus.

Introduction Ventricular shunting remains the standard of care for patients with idiopathic normal pressure hydrocephalus (iNPH); however, not all patients benefit from the shunting. Prediction of response in advance can result in improved patient selection for ventricular shunting. This study aims to develop a machine learning predictive model for treatment response after shunt placement using the clinical and radiomics features. Methods In this retrospective pilot study, the medical records of iNPH patients who underwent ventricular shunting were evaluated. In each patient, the "idiopathic normal pressure hydrocephalus grading scale" (iNPHGS) and a "Modified Rankin Scale" were calculated before and after surgery. The subsequent treatment response was calculated as the difference between the iNPHGS scores before and after surgery. iNPHGS score reduction of two or more than two were considered as treatment response. The presurgical MRI scans were evaluated by radiologists, the ventricular systems were segmented on the T2-weighted images, and the radiomics features were extracted from the segmented ventricular system. Using Orange data mining open-source platform, different machine learning models were then developed based on the presurgical clinical features and the selected radiomics features to predict treatment response after shunt placement. Results After the implementation of the inclusion criteria, 78 patients were included in this study. One hundred twenty radiomics features were extracted, and the 12 best predictive radiomics features were selected. Using only clinical data (iNPHGS and Modified Rankin Scale), the random forest model achieved the best performance in treatment prediction with an area under the curve (AUC) of 0.71. Adding the Radiomics analysis to the clinical data improved the prediction performance, with the support vector machine (SVM) achieving the highest rank in treatment prediction with an AUC of 0.8. Adding age and sex to the analysis did not improve the prediction. Conclusion Using machine learning models for treatment response prediction in patients with iNPH is feasible with acceptable accuracy. Adding the Radiomics analysis to the clinical features can further improve the predictive performance. SVM is likely the best model for this task.

Heavily calcified parapharyngeal space mesenchymal chondrosarcoma: Imaging and pathological findings and a review of the literature.

Parapharyngeal space (PPS) masses are relatively rare lesions of the head and neck, and account for 0.5-1.5% of head and neck lesions. The most common lesion to occur in the PPS is a benign salivary neoplasm, typically pleomorphic adenoma either from the deep parotid or from ectopic parotid tissue rests within the PPS. A calcified or ossified mass in this location is exceedingly rare, but a calcified variant of pleomorphic adenoma has been reported. In this study, we present a patient with a heavily calcified PPS mesenchymal chondrosarcoma with an unusual presentation. We discuss the imaging and pathologic findings followed by a review of the current literature.

Emerging Applications of Radiomics in Neurological Disorders: A Review.

Radiomics has achieved significant momentum in radiology research and can reveal image information invisible to radiologists' eyes. Radiomics first evolved for oncologic imaging. Oncologic applications (histopathology, tumor grading, gene mutation analysis, patient survival, and treatment response prediction) of radiomics are widespread. However, it is not limited to oncologic analysis, and any digital medical images can benefit from radiomics analysis. This article reviews the current literature on radiomics in non-oncologic, neurological disorders including ischemic strokes, hemorrhagic stroke, cerebral aneurysms, and demyelinating disorders.

Critical illness-associated cerebral microbleed in a patient with sickle cell disease: A case report and review of the literature.

Critical illness-associated cerebral microbleed (CICM) is a relatively rare and newly described condition. It can occur in critically ill ICU patients and can be secondary to many underlying etiologies. CICM is associated with high mortality and permanent neurologic deficits in surviving patients. Distribution of cerebral microhemorrhages in neuroimaging findings is critical for accurate diagnosis of this condition. Here, we present an ICU admitted patient with sickle cell disease crisis and CICM and will discuss their clinically and radiologically distinct phenomenon followed by a review of current literature.

A hypothesis about the role of fetal hemoglobin in COVID-19.

COVID-19 infection is less common in children (with higher fetal hemoglobin levels). In our preliminary study, we also observed a low prevalence and fatality of COVID-19 in countries with high rate of hemoglobinopathy carries. Given these two facts, the hemoglobin structure can play a role in the physiopathology of COVID-19 disease. Several drugs are known to increase fetal hemoglobin in adults. Adding these drugs to COVID-19 clinical trials may improve the patients' outcomes.

Artificial Intelligence Empowers Radiologists to Differentiate Pneumonia Induced by COVID-19 versus Influenza Viruses.

BACKGROUND: Given the current pandemic, differentiation between pneumonia induced by COVID-19 or influenza viruses is of utmost clinical significance in the patients' management. For this purpose, this study was conducted to develop sensitive artificial intelligence (AI) models to assist radiologists to decisively differentiate pneumonia due to COVID-19 versus influenza viruses. METHODS: Cross sectional chest CT images (N=12744) from well-evaluated cases of pneumonias induced by COVID-19 or H1N1 Influenza viruses, and normal individuals were collected. We examined the computer tomographic (CT) chest images from 137 individuals. Various pre-trained convolutional neural network models, such as ResNet-50, InceptionV3, Wide ResNet, SqueezNet, VGG 16 and VGG 19 were fine-tuned on our datasets. The datasets were used for training (60%), validation (20%), and testing (20%) of the final models. Also, the predictive power and means of precision and recall were determined for each model. RESULTS: Fine-tuned ResNet-50 model differentiated the pneumonia due to COVID-19 or H1N1 influenza virus with accuracies of 96.7% and 92%, respectively This model outperformed all others, i.e., InceptionV3, Wide ResNet, SqueezNet, VGG 16 and VGG 19. CONCLUSION: Fine-tuned and pre-trained image classifying models of AI enable radiologists to reliably differentiate the pneumonia induced by COVID-19 versus H1N1 influenza virus. For this purpose, ResNet-50 followed by InceptionV3 models proved more promising than other AI models. Also in the supplements, we share the source codes and our fine-tuned models for use by researchers and clinicians globally toward the critical task of image differentiation of patients infected with COVID-19 versus H1N1 Influenza viruses.

Pseudohypoxic Brain Swelling: Report of 2 Cases and Introduction of the Lentiform Rim Sign as Potential MRI Marker.

BACKGROUND: A rare but important complication related to otherwise uneventful brain and spine surgery is becoming more recognized and more frequently reported in the medical literature. This has been variably labeled as pseudohypoxic brain swelling or postoperative hypotension-associated venous congestion. This poorly understood condition occurs in the setting of surgical intervention and is thought to be related to cerebrospinal fluid leak or evacuation, decreased intracranial pressure, and subsequent development of deep venous congestion affecting the basal ganglia, thalami, and cerebellum. Clinically, patients may have global neurologic deficit and outcomes range from full recovery to vegetative state or death. The imaging correlate includes atypical edema, infarction, or hemorrhage and can overlap the appearance of diffuse hypoxic injury, for which this condition can be mistaken both clinically and radiologically. Although this deep brain tissue edema can be associated with other signs of cerebrospinal fluid hypotension such as dural thickening, brain sagging, and cerebellar herniation, it can be isolated, making the diagnosis challenging. CASE DESCRIPTION: We present 2 cases of unexpected clinical deterioration occurring in patients with otherwise uncomplicated neurosurgery, 1 with craniotomy and the other with lumbar spine intervention. Both patients exhibit similar appearing edema in the deep gray structures on postoperative magnetic resonance imaging scans. In addition to reviewing the prior literature and imaging findings, we evaluate the imaging findings to determine if there are unique features or signatures that might allow differentiation of PHBS from hypoxic-ischemic encephalopathy. CONCLUSIONS: The lentiform rim sign can be helpful for differentiation of pseudohypoxic brain swelling versus hypoxic-ischemic encephalopathy.

Superior ophthalmic vein thrombosis: What radiologist and clinician must know?

PURPOSE: Superior ophthalmic vein thrombosis (SOVT) is an extremely rare condition. Few studies have been published about clinical aspects of this condition. In this study, we have studied the symptoms, underlying etiologies, treatment, pathogenesis and complication of the SOVT and we tried to classify it based on the etiology, treatment, and prognosis. METHODS: We reviewed the patients' data from a tertiary academic referral center. Each patient with SOVT was then reviewed for symptoms associated with SOVT, underlying etiology, treatment protocol, treatment response, complications, possible pathogens, and final outcome. RESULTS: Twenty-four cases of SOVT were included in this study. Overall, 13 cases were diagnosed as right-sided SOVT, out of which, eight had simultaneous right-sided cavernous sinus thrombosis (CST). Eighteen cases were diagnosed to have left-sided SOVT, out of which, 11 had simultaneous left-sided CST. CONCLUSIONS: The SOVT can be secondary to different mechanisms. The SOVT secondary to trauma, recent surgery and coagulopathy are mostly non-aggressive, and can be managed by conservative therapy and anticoagulation. The SOVT in patients with orbital cellulitis, history of active sinusitis or paranasal sinus surgery are aggressive presenting with acute orbital swelling, abscess and visual loss. This type of SOVT can be complicated by extension to the cavernous sinus and intracranial structures. These patients require urgent antibiotics therapy and sinus surgery. The most severe type of SOVT is caused by mucormycosis which may also extend intracranially resulting in stroke and is often life-threatening.

Artificial Intelligence in the Management of Glioma: Era of Personalized Medicine.

Purpose: Artificial intelligence (AI) has accelerated novel discoveries across multiple disciplines including medicine. Clinical medicine suffers from a lack of AI-based applications, potentially due to lack of awareness of AI methodology. Future collaboration between computer scientists and clinicians is critical to maximize the benefits of transformative technology in this field for patients. To illustrate, we describe AI-based advances in the diagnosis and management of gliomas, the most common primary central nervous system (CNS) malignancy. Methods: Presented is a succinct description of foundational concepts of AI approaches and their relevance to clinical medicine, geared toward clinicians without computer science backgrounds. We also review novel AI approaches in the diagnosis and management of glioma. Results: Novel AI approaches in gliomas have been developed to predict the grading and genomics from imaging, automate the diagnosis from histopathology, and provide insight into prognosis. Conclusion: Novel AI approaches offer acceptable performance in gliomas. Further investigation is necessary to improve the methodology and determine the full clinical utility of these novel approaches.

Misleading CT perfusion in subacute ischemic stroke.

The RAPID© software is the most commonly used computed tomography perfusion (CTP) software in stroke centers. It is estimated that about 1300 hospitals in the world are using this software for decision-making in ischemic stroke. The software provides the estimated volume of infarction and ischemic penumbra, so it is the backbone of treatment planning in these patients. In this manuscript, we present two cases of subacute infarction with misleading CTP using RAPID© software. We believe that given the popularity of this software and increasing application of CTP in subacute infarction, this pitfall is likely underdiagnosed in many patients. In a subacute phase of infarction, we recommend diffusion-weighted imaging magnetic resonance imaging (DWI-MRI) for estimation of infarction to avoid this pitfall and possible mismanagement.

Fungal Peritonitis with Fungus Balls, a Complication of Trichobezoars and Rapunzel Syndrome.

BACKGROUND Rapunzel syndrome is a rare condition involving the extension of bezoars from the stomach to the distal gastrointestinal tract. Laparotomy remains the gold standard treatment for this condition because of the size of the bezoars. Although bacterial peritonitis is a known complication of laparotomy in Rapunzel syndrome, very few cases of post-surgical fungal peritonitis have been reported in these patients. CASE REPORT In this case report we present Rapunzel syndrome complicated by post-surgical fungal peritonitis and formation of fungus balls. To our knowledge, fungal peritonitis with fungus balls has never been reported as a Rapunzel syndrome complication.  CONCLUSIONS It is important to cover Candida and other fungi with an antifungal regimen in pediatric patients with Rapunzel syndrome pre- and post-surgery. In addition, prolonged fever and septic symptoms post-surgery warrant a search for peritoneal fungus balls that are not simply responsive to anti-fungal therapy and may necessitate repeat laparotomy.

Antidepressant-like effect of magnetic resonance imaging-based stimulation in mice.

INTRODUCTION: It has been reported that a novel type of magnetic resonance imaging (MRI) scan called echo planar magnetic resonance spectroscopic imaging (EP-MRSI) may show antidepressant effects. We examined whether the two routine diagnostic protocols of MRI [T1 and echo planar diffusion weighted imaging (EPI-DWI)], have antidepressant-like effects in an animal model of depression. METHODS: The effects of standard EPI-DWI and T1 MRI on immobility, swimming and climbing times in the modified forced swimming test (FST) in mice were examined. After exposure to the first session of modified forced swimming test, we randomly divided the mice into four groups. The first group (T1 MRI group, n=21) received a 15-minute stimulation of T1 sequence. The second group (EPI-DWI MRI group, n=21) received a 15-minute stimulation of EPI-DWI protocol. The third group (sham group, n=21) spent 15 min in a tunnel similar to the MRI gantry in terms of size, temperature and light intensity and received recorded sounds from a normal session of EPI-DWI with similar duration and intensity. The fourth group acted as controls (n=21). The second session of the modified FST was conducted twelve hours later. The mean of immobility, swimming and climbing times in this session were compared to the control group. RESULTS: T1 weighted and EPI-DWI MRI groups showed a reduction in immobility time compared to the control group (P value<0.002, P value<0.017 respectively). This effect is comparable to that seen in the FST after the administration of antidepressant agents. The climbing time in the group subjected to EPI-DWI MRI was longer than the control group (P value<0.035). Previous studies showed similar effects after the administration of antidepressant drugs affecting the catecholamine systems. The swimming time in the T1 MRI group was significantly longer than the control group (P value<0.037). Previous studies showed qualitatively similar effect to that of anti-depressant drugs affecting the serotoninergic systems. The swimming, climbing and immobility times in the sham and control groups showed no significant difference. CONCLUSIONS: Our findings raise the possibility that MRI-based stimulation may have antidepressant-like effects in mice. This is likely to be through different mechanisms in T1 weighted and EPI-DWI protocols. However the possible biological basis of this effect is not yet understood and we would advocate further studies of MRI-based stimulation effects on transmitters in the different organs in the body specially the brain.