Predicting symptomatic cerebral vasospasm after aneurysmal subarachnoid hemorrhage with an artificial neural network in a pediatric population

Intracranial Arterial Aneurysms in Childhood

Intracranial arterial aneurysms in children are rare. They differ from adult aneurysms in their etiology, natural history, and management approach. Unruptured asymptomatic aneurysms in children can often be observed for growth over time. Endovascular treatment has become the primary interventional modality in children with intracranial aneurysms. The authors discuss the management approach to pediatric intracranial aneurysms.

Need for Transparency and Clinical Interpretability in Hemorrhagic Stroke Artificial Intelligence Research: Promoting Effective Clinical Application

PURPOSE

This study aimed to evaluate the quality of artificial intelligence (AI)/machine learning (ML) studies on hemorrhagic stroke using the Minimum Information for Medical AI Reporting (MINIMAR) and Minimum Information About Clinical Artificial Intelligence Modeling (MI-CLAIM) frameworks to promote clinical application.

MATERIALS AND METHODS

PubMed, MEDLINE, and Embase were searched for AI/ML studies on hemorrhagic stroke. Out of the 531 articles found, 29 relevant original research articles were included. MINIMAR and MI-CLAIM scores were assigned by two experienced radiologists to assess the quality of the studies.

RESULTS

We analyzed 29 investigations that utilized AI/ML in the field of hemorrhagic stroke, involving a median of 224.5 patients. The majority of studies focused on diagnostic outcomes using computed tomography scans (89.7%) and were published in computer science journals (48.3%). The overall adherence rates to reporting guidelines, as assessed through the MINIMAR and MI-CLAIM frameworks, were 47.6% and 46.0%, respectively. In MINIMAR, none of the studies reported the socioeconomic status of the patients or how missing values had been addressed. In MI-CLAIM, only two studies applied model-examination techniques to improve model interpretability. Transparency and reproducibility were limited, as only 10.3% of the studies had publicly shared their code. Cohen's kappa between the two radiologists was 0.811 and 0.779 for MINIMAR and MI-CLAIM, respectively.

CONCLUSION

The overall reporting quality of published AI/ML studies on hemorrhagic stroke is suboptimal. It is necessary to incorporate model examination techniques for interpretability and promote code openness to enhance transparency and increase the clinical applicability of AI/ML studies.

Machine learning predicts cerebral vasospasm in patients with subarachnoid haemorrhage

BACKGROUND

Cerebral vasospasm (CV) is a feared complication which occurs after 20-40% of subarachnoid haemorrhage (SAH). It is standard practice to admit patients with SAH to intensive care for an extended period of resource-intensive monitoring. We used machine learning to predict CV requiring verapamil (CVRV) in the largest and only multi-center study to date.

METHODS

Patients with SAH admitted to UCLA from 2013 to 2022 and a validation cohort from VUMC from 2018 to 2023 were included. For each patient, 172 unique intensive care unit (ICU) variables were extracted through the primary endpoint, namely first verapamil administration or no verapamil. At each institution, a light gradient boosting machine (LightGBM) was trained using five-fold cross validation to predict the primary endpoint at various hospitalization timepoints.

FINDINGS

A total of 1750 patients were included from UCLA, 125 receiving verapamil. LightGBM achieved an area under the ROC (AUC) of 0.88 > 1 week in advance and ruled out 8% of non-verapamil patients with zero false negatives. Our models predicted "no CVRV" vs "CVRV within three days" vs "CVRV after three days" with AUCs = 0.88, 0.83, and 0.88, respectively. From VUMC, 1654 patients were included, 75 receiving verapamil. VUMC predictions averaged within 0.01 AUC points of UCLA predictions.

INTERPRETATION

We present an accurate and early predictor of CVRV using machine learning with multi-center validation. This represents a significant step towards optimized clinical management and resource allocation in patients with SAH.

FUNDING

Robert E. Freundlich is supported by National Center for Advancing Translational Sciences federal grant UL1TR002243 and National Heart, Lung, and Blood Institute federal grant K23HL148640; these funders did not play any role in this study. The National Institutes of Health supports Vanderbilt University Medical Center which indirectly supported these research efforts. Neither this study nor any other authors personally received financial support for the research presented in this manuscript. No support from pharmaceutical companies was received.

Machine Learning Predicts Cerebral Vasospasm in Subarachnoid Hemorrhage Patients

Background

Cerebral vasospasm (CV) is a feared complication occurring in 20-40% of patients following subarachnoid hemorrhage (SAH) and is known to contribute to delayed cerebral ischemia. It is standard practice to admit SAH patients to intensive care for an extended period of vigilant, resource-intensive, clinical monitoring. We used machine learning to predict CV requiring verapamil (CVRV) in the largest and only multi-center study to date.

Methods

SAH patients admitted to UCLA from 2013-2022 and a validation cohort from VUMC from 2018-2023 were included. For each patient, 172 unique intensive care unit (ICU) variables were extracted through the primary endpoint, namely first verapamil administration or ICU downgrade. At each institution, a light gradient boosting machine (LightGBM) was trained using five- fold cross validation to predict the primary endpoint at various timepoints during hospital admission. Receiver-operator curves (ROC) and precision-recall (PR) curves were generated.

Results

A total of 1,750 patients were included from UCLA, 125 receiving verapamil. LightGBM achieved an area under the ROC (AUC) of 0.88 an average of over one week in advance, and successfully ruled out 8% of non-verapamil patients with zero false negatives. Minimum leukocyte count, maximum platelet count, and maximum intracranial pressure were the variables with highest predictive accuracy. Our models predicted "no CVRV" vs "CVRV within three days" vs "CVRV after three days" with AUCs=0.88, 0.83, and 0.88, respectively. For external validation at VUMC, 1,654 patients were included, 75 receiving verapamil. Predictive models at VUMC performed very similarly to those at UCLA, averaging 0.01 AUC points lower.

Conclusions

We present an accurate (AUC=0.88) and early (>1 week prior) predictor of CVRV using machine learning over two large cohorts of subarachnoid hemorrhage patients at separate institutions. This represents a significant step towards optimized clinical management and improved resource allocation in the intensive care setting of subarachnoid hemorrhage patients.

Vasospasm in Pediatric Subarachnoid Hemorrhage

Cerebral vasospasm (CV) is a common severe complication of subarachnoid hemorrhage (SAH), a severe type of intracranial bleeding that is uncommon in children. The purpose of this article is to review the current literature regarding this potentially devastating complication. CV may be asymptomatic and is less common in children compared to adults. Several molecular phenomena, including inflammatory ones, contribute to its pathophysiology. Better collateral circulation and higher cerebral blood flow are protective factors in children. When clinically apparent, CV may manifest as a change in the child's neurologic status or vital signs. CV can be diagnosed using brain vessel imaging, such as computed tomography angiography, magnetic resonance angiography, digital subtraction angiography, transcranial Doppler ultrasonography, and computed tomography perfusion. A reduction of < 50% in the artery's caliber confirms the diagnosis. Besides general supportive measures and causative treatment of SAH, CV management options include the administration of calcium channel blockers and neurointerventional approaches, such as intra-arterial vasodilators and balloon angioplasty. Long-term outcomes in children are usually favorable.

External Validation of an Extreme Gradient Boosting Model for Prediction of Delayed Cerebral Ischemia After Aneurysmal Subarachnoid Hemorrhage

BACKGROUND

Delayed cerebral ischemia (DCI) may significantly worsen the functional status of patients with aneurysmal subarachnoid hemorrhage (aSAH). Several authors have designed predictive models for early identification of patients at risk of post-aSAH DCI. In this study, we externally validate an extreme gradient boosting (EGB) forecasting model for post-aSAH DCI prediction.

METHODS

A 9-year institutional retrospective review of patients with aSAH was performed. Patients were included if they underwent surgical or endovascular treatment and had available follow-up data. DCI was diagnosed as new-onset neurologic deficits at 4-12 days after aneurysm rupture, defined as worsening Glasgow Coma Scale score for ≥2 points, and new ischemic infarcts at imaging.

RESULTS

We collected 267 patients with aSAH. At admission, median Hunt-Hess score was 2 (range, 1-5), median Fisher score 3 (range, 1-4), and median modified Fisher score 3 (range, 1-4). One-hundred and forty-five patients underwent external ventricular drainage placement for hydrocephalus (54.3%). The ruptured aneurysms were treated with clipping (64%), coiling (34.8%), and stent-assisted coiling (1.1%). Fifty-eight patients (21.7%) were diagnosed with clinical DCI and 82 (30.7%) with asymptomatic imaging vasospasm. The EGB classifier correctly predicted 19 cases of DCI (7.1%) and 154 cases of no-DCI (57.7%), achieving sensitivity of 32.76% and specificity of 73.68%. The calculated F1 score and accuracy were 0.288% and 64.8%, respectively.

CONCLUSIONS

We validated that the EGB model is a potential assistant tool to predict post-aSAH DCI in clinical practice, finding moderate-high specificity but low sensitivity. Future research should investigate the underlying pathophysiology of DCI to allow the development of high-performing forecasting models.

Use of machine learning in pediatric surgical clinical prediction tools: A systematic review

PURPOSE

Clinical prediction tools (CPTs) are decision-making instruments utilizing patient data to predict specific clinical outcomes, risk-stratify patients, or suggest personalized diagnostic or therapeutic options. Recent advancements in artificial intelligence have resulted in a proliferation of CPTs created using machine learning (ML)-yet the clinical applicability of ML-based CPTs and their validation in clinical settings remain unclear. This systematic review aims to compare the validity and clinical efficacy of ML-based to traditional CPTs in pediatric surgery.

METHODS

Nine databases were searched from 2000 until July 9, 2021 to retrieve articles reporting on CPTs and ML for pediatric surgical conditions. PRISMA standards were followed, and screening was performed by two independent reviewers in Rayyan, with a third reviewer resolving conflicts. Risk of bias was assessed using the PROBAST.

RESULTS

Out of 8300 studies, 48 met the inclusion criteria. The most represented surgical specialties were pediatric general (14), neurosurgery (13) and cardiac surgery (12). Prognostic (26) CPTs were the most represented type of surgical pediatric CPTs followed by diagnostic (10), interventional (9), and risk stratifying (2). One study included a CPT for diagnostic, interventional and prognostic purposes. 81% of studies compared their CPT to ML-based CPTs, statistical CPTs, or the unaided clinician, but lacked external validation and/or evidence of clinical implementation.

CONCLUSIONS

While most studies claim significant potential improvements by incorporating ML-based CPTs in pediatric surgical decision-making, both external validation and clinical application remains limited. Further studies must focus on validating existing instruments or developing validated tools, and incorporating them in the clinical workflow.

TYPE OF STUDY

Systematic Review LEVEL OF EVIDENCE: Level III.

Closing the Gap in Pediatric Hemorrhagic Stroke: A Systematic Review

Pediatric hemorrhagic stroke (HS) accounts for a large proportion of childhood strokes, 1 of the top 10 causes of pediatric deaths. Morbidity and mortality lead to significant socio-economic and psychosocial burdens. To understand published data on recognizing and managing children with HS, we conducted a systematic review of the literature presented here. We searched PubMed, Embase, CINAHL and the Cochrane Library databases limited to English language and included 174 studies, most conducted in the USA (52%). Terminology used interchangeably for HS included intraparenchymal/intracranial hemorrhage, spontaneous ICH, and cerebrovascular accident (CVA). Key assessments informing prognosis and management included clinical scoring (Glasgow coma scale), and neuroimaging. HS etiologies reported were systemic coagulopathy (genetic, acquired pathologic, or iatrogenic), or focal cerebrovascular lesions (brain arteriovenous malformations, cavernous malformations, aneurysms, or tumor vascularity). Several scales were used to measure outcome: Glasgow outcome score (GOS), Kings outcome score for head injury (KOSCHI), modified Rankin scale (mRS) and pediatric stroke outcome measure (PSOM). Most studies described treatments of at-risk lesions. Few studies described neurocritical care management including raised ICP, seizures, vasospasm, or blood pressure. Predictors of poor outcome included ethnicity, comorbidity, location of bleed, and hematoma &gt;2% of total brain volume. Motor and cognitive outcomes followed independent patterns. Few studies reported on cognitive outcomes, rehabilitation, and transition of care models. Interdisciplinary approach to managing HS is urgently needed, informed by larger cohort studies targeting key clinical question (eg development of a field-guide for the clinician managing patients with HS that is reproducible internationally).

Diffuse Cerebral Vasospasm After Aneurysmal Subarachnoid Hemorrhage in a 15-Year-Old Girl: A Case Report

Diffuse cerebral vasospasm after subarachnoid hemorrhage (SAH) is a complication resulting in an ischemic condition presenting with altered mentality and followed by motor or speech impairment. It is uncommon in pediatric population and requires differential diagnosis from Moyamoya disease, which is relatively common in Korea. We report a case of a 15-year-old girl who was presented with a seizure and subsequent headache, poor oral intake, and altered mentality, who was finally diagnosed with sporadic vasospasm followed by multiple aneurysm ruptures. The patient had recurrent seizures and persistent headache at the time of transfer. On the second day after transfer, she showed focal motor weakness and dysarthria, and her symptoms gradually progressed, showing paraplegia and aphasia on the third hospitalization day. Brain magnetic resonance imaging and magnetic resonance angiography demonstrated diffuse narrowing cerebral vasospasm of bilateral middle cerebral arteries, anterior cerebral arteries, and distal internal carotid arteries and three unruptured aneurysms. The patient was treated with intravenous hydration and nimodipine to expand the narrowed vessels. After confirming that the vessels were enlarged, we successfully executed the endovascular coil embolization. Her neurological deficits were improved through medical, interventional, and rehabilitation treatments and fully restored 11 months after discharge.

Intracranial aneurysms in pediatric population: a two-center audit

OBJECTIVE

Intracranial aneurysms (IA) in children are rare, accounting for less than 5% of all IA. Due to their scarcity, the epidemiology is poorly understood and differs from adults in term of clinical presentation, size, location, and origin. Consequently, the treatment strategies are specific and cannot be only based on data from adult series. The aim of our study was to report the characteristics, management, and outcomes of children treated for IA in two university hospitals located in Normandy (France) over the last 17 years and to perform a literature review of this rare pathology.

METHODS

This retrospective study included 18 consecutive children (< 18 years old) admitted with cerebral aneurysm treated in two neurosurgery departments in Normandy, from 2001 to 2018. Computerized tomography and cerebral angiography established the diagnosis. Both endovascular and surgical procedures were discussed in all cases. Data focused on clinical condition at admission, characteristics of the IA, choice of the treatment modalities, and complications. The outcome at follow-up is based on Glasgow outcomes scale (GOS) at 1 year.

RESULTS

During the study period, 18 children (mean age: 12.6 years; sex ratio male/female: 2.3) were admitted with 21 IA. Aneurysms had a mean size of 13.6 mm with 4 giant aneurysms and were mostly located in the anterior circulation (16/21). Clinical presentations at onset were sudden symptoms related to a subarachnoid hemorrhage in 13 patients, headaches in 4 patients with giant aneurysm, and asymptomatic in one patient. Among the 13 patients with ruptured IA, 6 presented in poor preoperative condition (Hunt and Hess Grade ≥ 4). Treatment modalities consisted in embolization in 9 patients and surgery in 9 patients including 2 by-pass surgeries in fusiform aneurysms. Complications were similar in the two groups, but two cases of recanalization were observed in the endovascular group. At 1 year of follow-up, 14 children were in good condition (GOS Score > 4) and one died. Three children presented associated IA treated by the same technique as initial aneurysm.

CONCLUSIONS

Pediatric aneurysm is a different pathology compared with adults, occurring more frequently in male population with a higher proportion of giant aneurysms and aneurysms located in the internal carotid bifurcation. The use of endovascular techniques has progressed in the last years, but surgery was proposed for half of our population.

Can admission lipoprotein-associated phospholipase A2 predict the symptomatic cerebral vasospasm following aneurysmal subarachnoid hemorrhage?

Background

Inflammation has been believed to be related to the development of cerebral vasospasm following aneurysmal subarachnoid hemorrhage (aSAH). A potential biomarker for vascular inflammation that is well recognized is the lipoprotein-associated phospholipase A2 (Lp-PLA2). However, whether Lp-PLA2 can predict the occurrence of symptomatic cerebral vasospasm (SCV) in aSAH patients is still unknown. Thus, this study aimed to assess the value of Lp-PLA2 for predicting SCV in patients with aSAH.

Methods

Between March 2017 and April 2018, we evaluated 128 consecutive aSAH patients who were admitted in the First Affiliated Hospital of Fujian Medical University. Their Lp-PLA2 level was obtained within 24 h of the initial bleeding. Factors might be related to SCV were analyzed.

Results

Compared to patients without SCV, those with SCV (9.4%, 12/128) had significantly higher Lp-PLA2 level. Multivariate logistic analysis revealed that worse modified Fisher grade (OR = 10.08, 95% CI = 2.04-49.86, P = 0.005) and higher Lp-PLA2 level (OR = 6.66, 95% CI = 1.33-3.30, P = 0.021) were significantly associated with SCV, even after adjustment for confounders. Based on the best threshold, Lp-PLA2 had a sensitivity of 83.3% and a specificity of 51.7% for predicting SCV, as shown by the receiver operating characteristic curve analysis. In the poor World Federation of Neurosurgical Societies grade patient sub-group, patients with Lp-PLA2 > 200 μg/L had significantly higher SCV rate than that of patients having Lp-PLA2 ≤ 200 μg/L.

Conclusion

The admission Lp-PLA2 level might be a helpful predictor for SCV in aSAH.

Heart rate variability monitoring for the detection of delayed cerebral ischemia after aneurysmal subarachnoid hemorrhage

BACKGROUND

Delayed cerebral ischemia (DCI) is a major cause of impaired outcome after aneurysmal subarachnoidal hemorrhage (aSAH). In this observational cohort study we investigated whether changes in heart rate variability (HRV) that precede DCI could be detected.

METHODS

Sixty-four patients with aSAH were included. HRV data were collected for up to 10 days and analyzed offline. Correlation with clinical status and/or radiologic findings was investigated. A linear mixed model was used for the evaluation of HRV parameters over time in patients with and without DCI. Extended Glasgow outcome scale score was assessed after 1 year.

RESULTS

In 55 patients HRV data could be analyzed. Fifteen patients developed DCI. No changes in HRV parameters were observed 24 hours before onset of DCI. Mean of the HRV parameters in the first 48 hours did not correlate with the development of DCI. Low/high frequency (LF/HF) ratio increased more in patients developing DCI (β -0.07 (95% confidence interval, 0.12-0.01); P = .012). Lower STDRR (standard deviation of RR intervals), RMSSD (root mean square of the successive differences between adjacent RR intervals), and total power (P = .003, P = .007 and P = .004 respectively) in the first 48 hours were seen in patients who died within 1 year.

CONCLUSION

Impaired HRV correlated with 1-year mortality and LF/HF ratio increased more in patients developing DCI. Even though DCI could not be detected by the intermittent analysis of HRV used in this study, continuous HRV monitoring may have potential in the detection of DCI after aSAH using different methods of analysis.

Development of Decision Support System to Predict Neurofeedback Response in ADHD: an Artificial Neural Network Approach

INTRODUCTION

Clinical decision support system (CDSS) is an analytical tool that converts raw data into useful information to help clinicians make better decisions for patients.

AIM

The purpose of this study was to investigate the efficacy of neurofeedback (NF), in Attention Deficit Hyperactivity Disorder (ADHD) by the development of CDSS based on artificial neural network (ANN).

METHODS

This study analyzed 122 patients with ADHD who underwent NF in the Parand-Human Potential Empowerment Institute in Tehran. The patients were divided into two groups according to the effects of NF: effective and non-effective groups. The patients' record information was mined by data mining techniques to identify effective features. Based on unsaturated condition of data and imbalanced classes between the patient groups (patients with successful NF response and those without it), the SMOTE technique was applied on dataset. Using MATLAB 2014a, a modular program was designed to test both multiple architectures of neural networks and their performance. Selected architecture of the neural networks was then applied in the procedure.

RESULTS

Eleven features from 28 features of the initial dataset were selected as effective features. Using the SMOTE technique, number of the samples rose to around 300 samples. Based on the multiple neural networks architecture testing, a network by 11-20-16-2 neurons was selected (specify>00.91%, sensivity=100%) and applied in the software.

CONCLUSION

The ANN used in this study has led to good results in sensivity, specificity, and AUC. The ANN and other intelligent techniques can be used as supportive tools for decision making by healthcare providers.

Efficacy of CPP-ACP and CPP-ACPF for Prevention and Remineralization of White Spot Lesions in Orthodontic Patients: a Systematic Review of Randomized Controlled Clinical Trials

INTRODUCTION

Enamel subsurface lesions or white spot lesions (WSLs) are commonly found in orthodontic patients with a prevalence of 5% to 97%.

AIM

This systematic review aimed to evaluate the efficacy of casein phosphopeptide amorphous calcium phosphate (CPP-ACP) and casein phosphopeptide amorphous calcium phosphate fluoride (CPP-ACPF) for prevention and remineralization of WSLs in orthodontic patients in human randomized controlled clinical trials (RCTs).

METHODS

Relevant articles were retrieved by searching the Web of Science, Scopus, PubMed, and Cochrane Library databases up to November 2018 with no language or date restriction. The collected data included examination method, groups included in each study with number of patients in each group, study design, follow-up period and summary of important findings of each study. The risk of bias of each study was assessed according to the guidelines of the Cochrane Collaboration's tool.

RESULTS

Of 213 articles retrieved, 13 RCTs were included in this systematic review (none of them were included in the meta-analysis). Three articles showed superior efficacy of CPP-ACP for remineralization of WSLs while four studies reported the superior clinical efficacy of CPP-ACPF for this purpose.

CONCLUSION

Both CPP-ACP and CPP-ACPF can decrease the prevalence and increase the remineralization of WSLs during/after orthodontic treatment.