Utilizing machine learning to facilitate the early diagnosis of posterior circulation stroke.

BACKGROUND: Posterior Circulation Syndrome (PCS) presents a diagnostic challenge characterized by its variable and nonspecific symptoms. Timely and accurate diagnosis is crucial for improving patient outcomes. This study aims to enhance the early diagnosis of PCS by employing clinical and demographic data and machine learning. This approach targets a significant research gap in the field of stroke diagnosis and management. METHODS: We collected and analyzed data from a large national Stroke Registry spanning from January 2014 to July 2022. The dataset included 15,859 adult patients admitted with a primary diagnosis of stroke. Five machine learning models were trained: XGBoost, Random Forest, Support Vector Machine, Classification and Regression Trees, and Logistic Regression. Multiple performance metrics, such as accuracy, precision, recall, F1-score, AUC, Matthew's correlation coefficient, log loss, and Brier score, were utilized to evaluate model performance. RESULTS: The XGBoost model emerged as the top performer with an AUC of 0.81, accuracy of 0.79, precision of 0.5, recall of 0.62, and F1-score of 0.55. SHAP (SHapley Additive exPlanations) analysis identified key variables associated with PCS, including Body Mass Index, Random Blood Sugar, ataxia, dysarthria, and diastolic blood pressure and body temperature. These variables played a significant role in facilitating the early diagnosis of PCS, emphasizing their diagnostic value. CONCLUSION: This study pioneers the use of clinical data and machine learning models to facilitate the early diagnosis of PCS, filling a crucial gap in stroke research. Using simple clinical metrics such as BMI, RBS, ataxia, dysarthria, DBP, and body temperature will help clinicians diagnose PCS early. Despite limitations, such as data biases and regional specificity, our research contributes to advancing PCS understanding, potentially enhancing clinical decision-making and patient outcomes early in the patient's clinical journey. Further investigations are warranted to elucidate the underlying physiological mechanisms and validate these findings in broader populations and healthcare settings.

Complications and visual outcomes following surgical resection of pediatric optic pathway/hypothalamic gliomas: a systematic review and meta-analysis.

Pediatric optic pathway/hypothalamic gliomas (OPHG) pose challenges in treatment due to their location and proximity to vital structures. Surgical resection plays a key role in the management of OPHG especially when the tumor exhibits mass effect and causes symptoms. However, data regarding outcomes and complications of surgical resection for OPHG remains heterogenous. The authors performed a systematic review on pediatric OPHG in four databases: PubMed, EMBASE, Cochrane Library, and Google Scholar. We included studies that reported on the visual outcomes and complications of OPHG resection. A meta-analysis was performed and reported per the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement. A total of 26 retrospective studies were included. Seven hundred ninety-seven pediatric patients with OPHG undergoing surgical resection were examined. A diagnosis of NF1 was confirmed in 9.7%. Gross total resection was achieved in 36.7%. Intraorbital optic pathway gliomas showed a significantly higher gross total resection rate compared to those located in the chiasmatic/hypothalamic region (75.8% vs. 9.6%). Postoperatively, visual acuity improved in 24.6%, remained unchanged in 68.2%, and worsened in 18.2%. Complications included hydrocephalus (35.4%), anterior pituitary dysfunction (19.6%), and transient diabetes insipidus (29%). Tumor progression post-resection occurred in 12.8%, through a mean follow-up of 53.5 months. Surgical resection remains an essential strategy for treating symptomatic and large pediatric OPHG and can result in favorable vision outcomes in most patients. Careful patient selection is critical. Patients should be monitored for hydrocephalus development postoperatively and followed up to assess for tumor progression and adjuvant treatment necessity.

Machine learning-based prognostication of mortality in stroke patients.

Objectives: Predicting stroke mortality is crucial for personalized care. This study aims to design and evaluate a machine learning model to predict one-year mortality after a stroke. Materials and methods: Data from the National Multiethnic Stroke Registry was utilized. Eight machine learning (ML) models were trained and evaluated using various metrics. SHapley Additive exPlanations (SHAP) analysis was used to identify the influential predictors. Results: The final analysis included 9840 patients diagnosed with stroke were included in the study. The XGBoost algorithm exhibited optimal performance with high accuracy (94.5%) and AUC (87.3%). Core predictors encompassed National Institutes of Health Stroke Scale (NIHSS) at admission, age, hospital length of stay, mode of arrival, heart rate, and blood pressure. Increased NIHSS, age, and longer stay correlated with higher mortality. Ambulance arrival and lower diastolic blood pressure and lower body mass index predicted poorer outcomes. Conclusions: This model's predictive capacity emphasizes the significance of NIHSS, age, hospital stay, arrival mode, heart rate, blood pressure, and BMI in stroke mortality prediction. Specific findings suggest avenues for data quality enhancement, registry expansion, and real-world validation. The study underscores machine learning's potential for early mortality prediction, improving risk assessment, and personalized care. The potential transformation of care delivery through robust ML predictive tools for Stroke outcomes could revolutionize patient care, allowing for personalized plans and improved preventive strategies for stroke patients. However, it is imperative to conduct prospective validation to evaluate its practical clinical effectiveness and ensure its successful adoption across various healthcare environments.

Pediatric nasal chondromesenchymal hamartomas: a case series.

PURPOSE: Nasal chondromesenchymal hamartomas (NCMH) are rare, predominantly benign tumors of the sinonasal tract. The distinction from higher grade malignancy may be challenging based on imaging features alone. To increase the awareness of this entity among radiologists, we present a multi-institutional case series of pediatric NCMH patients showing the varied imaging presentation. METHODS: Descriptive assessment of imaging appearances of the lesions on computed tomography (CT) and magnetic resonance imaging (MRI) was performed. In addition, we reviewed demographic information, clinical data, results of genetic testing, management, and follow-up data. RESULTS: Our case series consisted of 10 patients, with a median age of 0.5 months. Intraorbital and intracranial extensions were both observed in two cases. Common CT findings included bony remodeling, calcifications, and bony erosions. MRI showed heterogeneous expansile lesion with predominantly hyperintense T2 signal and heterogenous post-contrast enhancement in the majority of cases. Most lesions exhibited increased diffusivity on diffusion weighted imaging and showed signal drop-out on susceptibility weighted images in the areas of calcifications. Genetic testing was conducted in 4 patients, revealing the presence of DICER1 pathogenic variant in three cases. Surgery was performed in all cases, with one recurrence in two cases and two recurrences in one case on follow-up. CONCLUSION: NCMHs are predominantly benign tumors of the sinonasal tract, typically associated with DICER1 pathogenic variants and most commonly affecting pediatric population. They may mimic aggressive behavior on imaging; therefore, awareness of this pathology is important. MRI and CT have complementary roles in the diagnosis of this entity.

Adjuvant Tranexamic Acid for Reducing Postoperative Recurrence of Chronic Subdural Hematoma in the Elderly: A Systematic Review and Meta-Analysis.

BACKGROUND: Chronic subdural hematoma (CSDH) is a frequently encountered neurosurgical disease among the elderly. The mainstay treatment involves surgical evacuation, but recurrence rates of approximately 13% pose complications. Adjuvant treatments, including tranexamic acid (TXA), have been explored, yet consensus on their efficacy and safety in elderly patients remains uncertain. The study aims to examine the role of TXA as adjunctive therapy in reducing CSDH recurrence and explore any potential association between TXA use and thrombotic events in this patient population. METHODS: The systematic review and meta-analysis adhered to Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines and Cochrane Handbook standards, searching databases up to July 2023 for randomized controlled trials and propensity-matched cohorts evaluating adjuvant TXA. The primary outcome was CSDH recurrence, and the secondary outcome was thrombosis risk, measured as relative risks. RESULTS: A total of 6 studies were included, comprising 1403 patients with CSDH who underwent surgical treatment. Four studies were randomized controlled trials, while the other 2 were propensity-matched cohorts. The overall pooled relative risk for CSDH recurrence in the TXA group compared to the control group was 0.41 (95% confidence interval [0.29-0.59], P < 0.01), indicating a significant reduction in recurrence with TXA treatment. CONCLUSIONS: In conclusion, our study indicates that adjuvant TXA may help reduce CSDH recurrence in elderly patients undergoing surgical treatment. However, the study has limitations and there is a need for further research to validate these findings.

Cellular senescence in brain aging and cognitive decline.

Cellular senescence is a biological aging hallmark that plays a key role in the development of neurodegenerative diseases. Clinical trials are currently underway to evaluate the effectiveness of senotherapies for these diseases. However, the impact of senescence on brain aging and cognitive decline in the absence of neurodegeneration remains uncertain. Moreover, patient populations like cancer survivors, traumatic brain injury survivors, obese individuals, obstructive sleep apnea patients, and chronic kidney disease patients can suffer age-related brain changes like cognitive decline prematurely, suggesting that they may suffer accelerated senescence in the brain. Understanding the role of senescence in neurocognitive deficits linked to these conditions is crucial, especially considering the rapidly evolving field of senotherapeutics. Such treatments could help alleviate early brain aging in these patients, significantly reducing patient morbidity and healthcare costs. This review provides a translational perspective on how cellular senescence plays a role in brain aging and age-related cognitive decline. We also discuss important caveats surrounding mainstream senotherapies like senolytics and senomorphics, and present emerging evidence of hyperbaric oxygen therapy and immune-directed therapies as viable modalities for reducing senescent cell burden.

Fluorescence-guided resection of intradural spinal tumors: a systematic review and meta-analysis.

Intradural spinal tumors present significant challenges due to involvement of critical motor and sensory tracts. Achieving maximal resection while preserving functional tissue is therefore crucial. Fluorescence-guided surgery aims to improve resection accuracy and is well studied for brain tumors, but its efficacy has not been fully assessed for spinal tumors. This meta-analysis aims to delineate the efficacy of fluorescence guidance in intradural spinal tumor resection. The authors performed a systematic review in four databases. We included studies that have utilized fluorescence agents, 5-aminolevulinic acid (5-ALA) or sodium fluorescein, for the resection of intradural spinal tumors. A meta-analysis was performed in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement. A total of 12 studies involving 552 patients undergoing fluorescence-guided intradural spinal tumor resection were included. Meningiomas demonstrated a 98% fluorescence rate and were associated with a homogenous florescence pattern; however, astrocytomas had variable fluorescence rate with pooled proportion of 70%. There was no significant difference in gross total resection (GTR) rates between fluorescein and 5-ALA (94% vs 84%, p = .22). Pre-operative contrast enhancement was significantly associated with intraoperative fluorescence with fluorescein. Intramedullary tumors with positive intraoperative fluorescence were significantly associated with higher GTR rates (96% vs 73%, p = .03). Utilizing fluorescence guidance during intradural spinal tumor resection holds promise of improving intraoperative visualization for specific intradural spinal tumors. Meningiomas and ependymomas have the highest fluorescence rates especially with sodium fluorescein; on the other hand, astrocytomas have variable fluorescence rates with no superiority of either agent. Positive fluorescence of intramedullary tumors is associated with a higher degree of resection.

Predicting 90-Day Prognosis in Ischemic Stroke Patients Post Thrombolysis Using Machine Learning.

(1) Objective: This study aimed to construct a machine learning model for predicting the prognosis of ischemic stroke patients who underwent thrombolysis, assessed through the modified Rankin Scale (mRS) score 90 days after discharge. (2) Methods: Data were sourced from Qatar's stroke registry covering January 2014 to June 2022. A total of 723 patients with ischemic stroke who had received thrombolysis were included. Clinical variables were examined, encompassing demographics, stroke severity indices, comorbidities, laboratory results, admission vital signs, and hospital-acquired complications. The predictive capabilities of five distinct machine learning models were rigorously evaluated using a comprehensive set of metrics. The SHAP analysis was deployed to uncover the most influential predictors. (3) Results: The Support Vector Machine (SVM) model emerged as the standout performer, achieving an area under the curve (AUC) of 0.72. Key determinants of patient outcomes included stroke severity at admission; admission systolic and diastolic blood pressure; baseline comorbidities, notably hypertension (HTN) and coronary artery disease (CAD); stroke subtype, particularly strokes of undetermined origin (SUO); and hospital-acquired urinary tract infections (UTIs). (4) Conclusions: Machine learning can improve early prognosis prediction in ischemic stroke, especially after thrombolysis. The SVM model is a promising tool for empowering clinicians to create individualized treatment plans. Despite limitations, this study contributes to our knowledge and encourages future research to integrate more comprehensive data. Ultimately, it offers a pathway to improve personalized stroke care and enhance the quality of life for stroke survivors.

Neutrophil extracellular traps and long COVID.

Post-acute COVID-19 sequelae, commonly known as long COVID, encompasses a range of systemic symptoms experienced by a significant number of COVID-19 survivors. The underlying pathophysiology of long COVID has become a topic of intense research discussion. While chronic inflammation in long COVID has received considerable attention, the role of neutrophils, which are the most abundant of all immune cells and primary responders to inflammation, has been unfortunately overlooked, perhaps due to their short lifespan. In this review, we discuss the emerging role of neutrophil extracellular traps (NETs) in the persistent inflammatory response observed in long COVID patients. We present early evidence linking the persistence of NETs to pulmonary fibrosis, cardiovascular abnormalities, and neurological dysfunction in long COVID. Several uncertainties require investigation in future studies. These include the mechanisms by which SARS-CoV-2 brings about sustained neutrophil activation phenotypes after infection resolution; whether the heterogeneity of neutrophils seen in acute SARS-CoV-2 infection persists into the chronic phase; whether the presence of autoantibodies in long COVID can induce NETs and protect them from degradation; whether NETs exert differential, organ-specific effects; specifically which NET components contribute to organ-specific pathologies, such as pulmonary fibrosis; and whether senescent cells can drive NET formation through their pro-inflammatory secretome in long COVID. Answering these questions may pave the way for the development of clinically applicable strategies targeting NETs, providing relief for this emerging health crisis.

Using artificial intelligence to improve body iron quantification: A scoping review.

This scoping review explores the potential of artificial intelligence (AI) in enhancing the screening, diagnosis, and monitoring of disorders related to body iron levels. A systematic search was performed to identify studies that utilize machine learning in iron-related disorders. The search revealed a wide range of machine learning algorithms used by different studies. Notably, most studies used a single data type. The studies varied in terms of sample sizes, participant ages, and geographical locations. AI's role in quantifying iron concentration is still in its early stages, yet its potential is significant. The question is whether AI-based diagnostic biomarkers can offer innovative approaches for screening, diagnosing, and monitoring of iron overload and anemia.

Long-term functional outcomes and complications of microsurgical resection of brainstem cavernous malformations: a systematic review and meta-analysis.

Brainstem cavernous malformations (CMs) encompass up to 20% of all intracranial CMs and are considered more aggressive than cerebral CMs because of their high annual bleeding rates. Microsurgical resection remains the primary treatment modality for CMs, but long-term functional outcomes and complications are heterogenous in the literature. The authors performed a systematic review on brainstem CMs in 4 databases: PubMed, EMBASE, Cochrane library, and Google Scholar. We included studies that reported on the long-term functional outcomes and complications of brainstem CMs microsurgical resection. A meta-analysis was performed and reported in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement. The search yielded 4781 results, of which 19 studies met our inclusion criteria. Microsurgery was performed on 940 patients (mean age 35 years, 46.9% females). Most of the brainstem CMs were located in the pons (n = 475). The pooled proportions of improved, stable, and worsened functional outcomes after microsurgical resection of brainstem CMs were 56.7% (95% CI 48.4-64.6), 28.6% (95% CI 22.4-35.7), and 12.6% (95% CI 9.6-16.2), respectively. CMs located in the medulla were significantly (p = 0.003) associated with a higher proportion of improved outcome compared with those in the pons and midbrain. Complete resection was achieved in 93.3% (95% CI 89.8-95.7). The immediate postoperative complication rate was 37.2% (95% CI 29.3-45.9), with new-onset cranial nerve deficit being the most common complication. The permanent morbidity rate was 17.3% (95% CI 10.5-27.1), with a low mortality rate of 1% from the compiled study population during a mean follow-up of 58 months. Our analysis indicates that microsurgical resection of brainstem CMs can result in favorable long-term functional outcomes with transient complications in the majority of patients. Complete microsurgical resection of the CM is associated with a lower incidence of CM hemorrhage and the morbidity related to it.

Morbidity and etiology-based success rate of combined endoscopic ventriculostomy and choroid plexus cauterization: a systematic review and meta-analysis of 1918 infants.

Approaches to the treatment of infant hydrocephalus vary among centers. Standard shunting carries a significant infection rate, an unpredictable time-to-failure, and the life-long risk of recurrent failures. Combined choroid plexus cauterization (CPC) and endoscopic third ventriculostomy (ETV) have been increasingly employed over the past decade as an alternative approach in an attempt to avoid shunt dependency. We performed a systematic review and meta-analysis to explore the reported morbidity associated with ETV/CPC and its rate of success reported for specific etiologies of infant hydrocephalus. The protocol of this study was registered with the International prospective register of Systematic Reviews (PROSPERO) with the following registration number: CRD 42022343898. The study utilized four databases of medical literature to perform a systematic search following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. Several parameters were extracted from the included studies including authors, publication year, region, study design, sample size, mean age, success rate, complication rate, reported complications, hydrocephalus etiology, median time-to-failure, secondary management after failure, and mean follow-up time. The outcomes of interest, success, and complication rates were pooled using 95% confidence intervals (CI) and a random effects model. Heterogeneity was assessed using the I2 test. Twenty-eight studies met the inclusion criteria from an initial search result of 472 studies. The study included 1938 infants (1918 of which were included in the meta-analysis). The overall success rate of combined ETV/CPC is 0.59 (95% CI (0.53, 0.64), I2 = 82%). Etiology-based success rate is 0.71, 0.70, 0.64, and 0.52 for aqueductal stenosis, myelomeningocele, postinfectious hydrocephalus, and posthemorrhagic hydrocephalus, respectively. The overall complication rate is 0.04 (95% CI (0.02, 0.05), I2 = 14%). Our study presents a comprehensive analysis of the current evidence on the use of ETV/CPC for treating hydrocephalus in infants. The findings demonstrate the potential efficacy of this procedure; however, it is crucial to acknowledge the limitations inherent in the included studies, such as selection bias and limited follow-up, which could have impacted the reported outcomes.

Tackling the glial scar in spinal cord regeneration: new discoveries and future directions.

Axonal regeneration and functional recovery are poor after spinal cord injury (SCI), typified by the formation of an injury scar. While this scar was traditionally believed to be primarily responsible for axonal regeneration failure, current knowledge takes a more holistic approach that considers the intrinsic growth capacity of axons. Targeting the SCI scar has also not reproducibly yielded nearly the same efficacy in animal models compared to these neuron-directed approaches. These results suggest that the major reason behind central nervous system (CNS) regeneration failure is not the injury scar but a failure to stimulate axon growth adequately. These findings raise questions about whether targeting neuroinflammation and glial scarring still constitute viable translational avenues. We provide a comprehensive review of the dual role of neuroinflammation and scarring after SCI and how future research can produce therapeutic strategies targeting the hurdles to axonal regeneration posed by these processes without compromising neuroprotection.

Current state of radiomics in pediatric neuro-oncology practice: a systematic review.

BACKGROUND: Radiomics is the process of converting radiological images into high-dimensional data that may be used to create machine learning models capable of predicting clinical outcomes, such as disease progression, treatment response and survival. Pediatric central nervous system (CNS) tumors differ from adult CNS tumors in terms of their tissue morphology, molecular subtype and textural features. We set out to appraise the current impact of this technology in clinical pediatric neuro-oncology practice. OBJECTIVES: The aims of the study were to assess radiomics' current impact and potential utility in pediatric neuro-oncology practice; to evaluate the accuracy of radiomics-based machine learning models and compare this to the current standard which is stereotactic brain biopsy; and finally, to identify the current limitations of radiomics applications in pediatric neuro-oncology. MATERIALS AND METHODS: Following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) standards, a systematic review of the literature was carried out with protocol number CRD42022372485 in the prospective register of systematic reviews (PROSPERO). We performed a systematic literature search via PubMed, Embase, Web of Science and Google Scholar. Studies involving CNS tumors, studies that utilized radiomics and studies involving pediatric patients (age<18 years) were included. Several parameters were collected including imaging modality, sample size, image segmentation technique, machine learning model used, tumor type, radiomics utility, model accuracy, radiomics quality score and reported limitations. RESULTS: The study included a total of 17 articles that underwent full-text review, after excluding duplicates, conference abstracts and studies that did not meet the inclusion criteria. The most commonly used machine learning models were support vector machines (n=7) and random forests (n=6), with an area under the curve (AUC) range of 0.60-0.94. The included studies investigated several pediatric CNS tumors, with ependymoma and medulloblastoma being the most frequently studied. Radiomics was primarily used for lesion identification, molecular subtyping, survival prognostication and metastasis prediction in pediatric neuro-oncology. The low sample size of studies was a commonly reported limitation. CONCLUSION: The current state of radiomics in pediatric neuro-oncology is promising, in terms of distinguishing between tumor types; however, its utility in response assessment requires further evaluation which, given the relatively low number of pediatric tumors, calls for multicenter collaboration.

Imaging of Abusive Head Trauma in Children.

In this article, we describe relevant anatomy, mechanisms of injury, and imaging findings of abusive head trauma (AHT). We also briefly address certain mimics of AHT, controversies, pearls, and pitfalls. Concepts of injury, its evolution, and complex nature of certain cases are highlighted with the help of case vignettes.

Unusual presentation of an aneurysmal bone cyst: A case report and literature review.

Aneurysmal bone cysts (ABCs) are non-neoplastic primary bone tumors, typically involving the long bones and vertebrae in the first 2 decades of life. ABCs require prompt diagnosis and intervention due to their rapidly expansile nature and ability to destroy the adjacent normal bone. ABCs rarely affect the rib. We report a case of a 51-year-old female presenting with chronic dry cough and right upper back pain. A chest X-ray and computed tomography scan revealed an expansile, lytic mass affecting the posterior aspect of the third right rib. The third right rib was resected using a posterolateral, Shaw-Paulson approach. Histopathology of the resected mass confirmed the diagnosis of ABC. There were no intra- or perioperative complications, and follow-up X-ray was normal.

Challenges and limitations of synthetic minority oversampling techniques in machine learning.

Oversampling is the most utilized approach to deal with class-imbalanced datasets, as seen by the plethora of oversampling methods developed in the last two decades. We argue in the following editorial the issues with oversampling that stem from the possibility of overfitting and the generation of synthetic cases that might not accurately represent the minority class. These limitations should be considered when using oversampling techniques. We also propose several alternate strategies for dealing with imbalanced data, as well as a future work perspective.