Search

Segmentation of intracerebral hemorrhage (ICH) helps improve the quality of diagnosis, draft the desired treatment methods, and clinically observe the variations with healthy patients. The clinical utilization of various ICH progression scoring systems has limitations due to the systems' modest predictive value. This paper proposes a single pipeline of a multi-task model for end-to-end hemorrhage segmentation and risk estimation. We introduce a 3D spatial attention unit and integrate it into the state-of-the-art segmentation architecture, UNet, to enhance the accuracy by bootstrapping the global spatial representation. We further extract the geometric features from the segmented hemorrhage volume and fuse them with clinical features such as CT angiography (CTA) spot, Glasgow Coma Scale (GCS), and age to predict the ICH stability. Several state-of-the-art machine learning techniques such as multilayer perceptron (MLP), support vector machine (SVM), gradient boosting, and random forests are applied to train stability estimation and to compare the performances. To align clinical intuition with model learning, we determine the shapely values (SHAP) and explain the most significant features for the ICH risk scoring system. A total of 79 patients are included, of which 20 are found in critical condition. Our proposed single pipeline model achieves a segmentation accuracy of 86.3%, stability prediction accuracy of 78.3%, and precision of 82.9%; the mean square error of exact expansion rate regression is observed to be 0.46. The SHAP analysis reveals that CTA spot sign, age, solidity, location, and length of the first axis of the ICH volume are the most critical characteristics that help define the stability of the stroke lesion. We also show that integrating significant geometric features with clinical features can improve the ICH progression scoring by predicting long-term outcomes. Graphical abstract Overview of our proposed method comprising of spatial attention and feature extraction mechanisms. The architecture is trained on the input CT images, and the first step output is the predicted segmentation of the hemorrhagic region. The output is fed into a geometric feature extractor and is fused with clinical features to estimate ICH stability using a multilayer perceptron (MLP).

Chronic subdural haemorrhage (CSDH) is a common neurosurgical entity with complex pathophysiological pathways. The generally favourable surgical outcome may be affected by its associated risks including recurrence rates. We performed a prospective randomized multi-center clinical trial comparing the addition of tranexamic acid (TXA) to standard neurosurgical procedures for patients with symptomatic CSDH. The primary endpoint was CSDH requiring repeat surgery within 6-month post-operatively. Secondary endpoints were comparison of post-operative volumes between the treatment arms and safety evaluation of the dosing regime. 90 patients were analyzed with 49 patients in the observation arm and 41 patients in the TXA arm. The observation arm had five (10.2%) recurrences compared to two (4.8%, p = 0.221) in the TXA arm. Patients in the TXA arm demonstrated a greater reduction of their CSDH volume at 6 weeks follow up (36.6%) compared to the observation arm (23.3%, p = 0.6648). There were no reportable serious adverse events recorded in the observation arm, compared to 4 (9.8%) patients in the TXA arm. The addition of TXA treatment to standard surgical drainage of CSH did not significantly reduce symptomatic post-operative recurrence. Patients in the TXA arm had a delay in the CSDH recurrence with a comparative reduction of residual hematoma volume at the 6-week follow up although the effect was unsustained. Larger randomized trials with dose adjustments should be considered to investigate subgroups of patients that may benefit from this medical adjunct.

OBJECTIVENeurosurgery presents unique surgical challenges arising from delicate neural structures, limited accessibility, and the risk of CSF leakage that can lead to CNS infections. Sutures and staples may have limited applicability in the complex anatomical constraints of cranial and spinal surgeries, especially in trauma settings when time is of the essence. Surgical bioadhesives are emerging as attractive alternatives because they avoid traumatic application methods, provide a stress-distributed fixation, and provide good cosmesis and outcomes. This article presents the history of the development of surgical bioadhesives, and is also a review of current applications of commercial surgical bioadhesives within neurosurgical procedures and the unmet clinical needs that should be addressed in bioadhesives technologies.METHODSA PubMed literature search was performed using the terms “(glue OR bioadhesive OR fibrin OR tisseel OR evicel OR tachosil OR cyanoacrylate OR duraseal OR bioglue) AND (neurosurgery OR spine OR spinal OR dural OR microvascular decompression OR transsphenoidal OR endovascular).” Of 2433 records screened, 168 studies were identified that described the use of bioadhesives in neurosurgical procedures.RESULTSThe greatest number of studies describing the use of bioadhesives in neurosurgery were identified for endovascular embolization, followed by dural closure and transsphenoidal surgeries. Other common areas of application were for microvascular decompression, skin closure, peripheral nerve repair, and other novel uses. Numerous case reports were also identified describing complications associated with bioadhesive use.CONCLUSIONSDespite the paucity of approved indications, surgical bioadhesive use in neurosurgical procedures is prevalent. However, current bioadhesives still each have their own limitations and research is intense in the development of novel solutions.

Deep brain stimulation of the subthalamic nucleus (STN DBS) has been shown to reduce antiparkinsonian medication in Parkinson’s disease. We aimed to investigate the changes in long-term medication profile with STN DBS. Antiparkinsonian medication data for 56 patients were collected from as early as 3 years before STN DBS up to 10 years after. Cost spending on medication changes was analyzed. Mean levodopa equivalent daily dose (LEDD) was projected 10 years into the future based on preoperative data to create a comparator group wherein the patients did not undergo STN DBS. Use of neuroleptics and antidepressants was also recorded. LEDD requirement was significantly reduced by a mean of 31 ± 2% over 10 years after DBS, from 1049 ± 381 mg at pre-DBS baseline, to 713 ± 392 mg at 1 year post-DBS, and 712 ± 385 mg at 10 years post-DBS. This was associated with a mean reduction of 35 ± 3% in medicine cost. Modeled LEDD requirements for not having STN DBS were in the range of 1489 mg to 2721 mg at 10 years post-DBS (109–282% higher than the observed mean LEDD in DBS cohort). The proportion of patients increased from 5% before STN DBS to 14% at 10 year post-DBS for neuroleptics, and 11–23% for antidepressants. STN DBS led to LEDD reduction and antiparkinsonian medication cost savings in our South-East Asian cohort. Medication reduction with STN DBS in our cohort over the 10-year period was comparable to those reported in Western populations.

Neurologic disorders impose a heavy burden on healthcare in Singapore. To date, no data on the willingness to pay (WTP) for neurologic treatments has been reported in the local population.We aimed to quantify the value of various health domains to neuroscience patients and their caregivers by comparing their WTP for different types of treatments.A questionnaire using a mixed open-ended and closed-ended contingent valuation method was developed to elicit WTP and self-administered by 112 visitors to a neuroscience outpatient clinic. The WTP for treatments in 3 health domains (advanced restoration of function, life extension, and cosmesis) was evaluated and compared. Subgroup regression analysis was performed to investigate the impact of demographic and socioeconomic factors.Treatment that improved cosmesis had the highest median WTP of Singapore dollar (SGD) 35 000, followed by treatment that provided 1 year of life extension (SGD 20 000) and 1 year of advanced restored function (SGD 10 000; P.001). Respondents with a university education were willing to pay as much as 2 to 3 times of those without across all health domains.This is the first study to provide data on how different health domains are valued by neuroscience patients and caregivers in our population. Respondents valued treatment that restored or improved their physical appearances the most. These findings could contribute to future policies on the improvement of neuroscience care.

Background and Purpose— Selective serotonin reuptake inhibitors (SSRIs) have a well-established association with bleeding complications and conflicting reports on outcome after stroke. We sought to evaluate whether pre–intracerebral hemorrhage (ICH) SSRI use increased ICH risk and post-ICH SSRI use improved ICH outcome. Methods— Through post hoc analysis of the ERICH study (Ethnic/Racial Variations of Intracerebral Hemorrhage), SSRI use was categorized into no use, pre-ICH only, pre- and post-ICH use (termed “continuous”), and post-ICH only (termed “new”). Using multivariable modeling, associations were sought between pre-ICH SSRI use and ICH risk in the case-control set, and associations between post-ICH SSRI use and 3-month outcome were analyzed in the ICH case set. Exploratory analyses sought to assess influence of race/ethnicity in models. Results— The final study cohort consisted of 2287 ICH cases and 2895 controls. Pre-ICH SSRI use was not associated with ICH risk (odds ratio, 0.824 [95% CI, 0.632–1.074]) nor potentiation of ICH risk with anticoagulant or antiplatelet use. New post-ICH SSRI use was associated with unfavorable modified Rankin Scale score at 3 months after ICH (odds ratio, 1.673 [95% CI, 1.162–2.408]; P =0.006) in multivariable analyses. Additional propensity score analysis indicated a similar trend but did not reach statistical significance ( P =0.107). When stratified by race/ethnicity, multivariable modeling demonstrated reduced ICH risk with pre-ICH SSRI use in Hispanics (odds ratio, 0.513 [95% CI, 0.301–0.875]; P =0.014), but not non-Hispanic whites or blacks, and no associations between post-ICH SSRI use and 3-month outcome in any racial/ethnic group. Conclusions— In a large multiethnic cohort, pre-ICH SSRI use was not associated with increased ICH risk, but post-ICH SSRI use was associated with unfavorable 3-month neurological outcome after ICH. Registration— URL: https://www.clinicaltrials.gov ; Unique identifier: NCT01202864.

This article presents a robotic needle positioning approach with a novel 4-DoF parallel positioner, two back-driveable 2-channel linear/rotational driver, and a machine–body interface. Being more compact and flexible, Ni–Ti alloy wire-based Bowden cable transmission is introduced to connect the positioner and the drivers. For this new mechanism, forward and inverse kinematics are derived, and the workspace is analyzed. The mechanism’s maximum incidence angle is 45 $^{\circ }$ , and the stroke of the slider is 41 mm. The structural size of one positioner is $\varnothing$ 70 mm × 45 mm. The compact size and lightweight of the positioner make it readily mountable on the skull or attached to other parts of the body. We fabricated a 3D printed proof-of-concept prototype of the system and tested its performance. The open-loop positioning accuracy of the slider and rotor is within $\pm$ 1 mm and from $-0.8^{\circ }$ to 1.5 $^{\circ }$ , respectively. The trajectory tracking error of the system is within 1.71 mm. The stiffness of the positioner in the x - and y -directions can be calculated as 9.03 and 11.91 N/mm, respectively. Finally, an image-guided navigation framework based on electromagnetic tracking demonstrates the feasibility of the proposed system on a phantom study.

This paper presents a disposable fiber Bragg grating (FBG) based tridirectional force/torque sensor mountable at the end of the tip of the aspiration instruments for detecting instrument–tissue interaction force/torque. The sensor utilizes an inexpensive three-dimensional (3-D) printed elastomer with an annular diaphragm, which is sensitive to axial force (Fz) and torques ( Mx and My ). Four suspended optical fibers inscribed with an FBG each have been fixed on two ends of the elastomer to simultaneously sense and decouple force/torque-induced diaphragm deformation. The associated sensing theoretical model and force-torque decoupling principle with temperature compensation have been derived. Calibration experiments demonstrate a high resolution of 8.8 mN within the range of 0–2 N for the axial force detection. Dynamic performances of the designed sensor have been evaluated in comparison with the commercial force sensor. The average relative errors of force/torque components are less than 4.4% of full-scale range. The maximum temperature drift-caused errors of the force/torque components within the range of 30 °C–45 °C are –0.1340 N, 0.0084 N·mm, and 0.0300 N·mm, respectively. In vivo experiments on a cadaver head have been performed to further validate the feasibility and effectiveness of the designed sensor for real-time monitoring of the interaction force.

Background Posterior fossa surgery is the established treatment for large cerebellar strokes with brainstem compression. Despite this, there is a paucity of data for long-term outcomes. Methods A retrospective analysis of patients who underwent posterior fossa surgery for cerebellar hemorrhages and infarcts was performed to compare their difference in 6-month outcomes and to identify factors that affect outcomes. Patients were dichotomized into groups with good outcomes (modified Rankin scale [mRS] score 0–3) or poor outcomes (mRS score 4–6). Sex, age, preoperative Glasgow Coma Scale score, Charleston comorbidity index, time to surgery, intraventricular hemorrhage, surgical complications, length of intensive care unit and hospital stay, shunt dependence, and tracheostomy rates were analyzed. Results In total, 126 patients were recruited: 76 in hemorrhage group and 50 in infarct group. There was a greater mortality in the hemorrhage group (P = 0.0730). At 6 months, more patients in the hemorrhage group had poor outcomes (P = 0.0074, odds ratio 3.04) and greater mortality (P = 0.0730, odds ratio 2.20). More patients in the hemorrhage group required a tracheostomy (P = 0.0245). Factors predictive of poor outcome include older age (P = 0.0108), Glasgow Coma Scale score ≤8 (P = 0.0011), and tracheostomy (P = 0.0269). A total of 69.2% of patients had improvements in mRS scores at 6 months. Shorter length of stay (P = 0.0003) and discharge to a rehabilitation hospital (P = 0.0001) were predictive of functional improvement. Conclusions Patients who underwent posterior fossa surgery for cerebellar hemorrhage had worse outcomes compared with patients with cerebellar infarcts and were more likely to require a tracheostomy. Rehabilitation helped to improved outcomes.

Glioblastoma multiforme (GBM) is a rapidly growing tumor associated with poor prognosis. This study evaluates the effectiveness of thirteen tumor shape features for overall survival (OS) prognosis in GBM patients. Shape features were extracted from the abnormality regions of the GBM tumor visible on the fluid attenuated inversion recovery (FLAIR) and T1-weighted contrast enhanced (T1CE) MR images of GBM patients. Survival analysis was performed using univariate and multivariate (with clinical features) Cox proportional hazards regression analysis. Kaplan-Meier survival curves were plotted and observed for the shape features which were found to be significant from the Cox regression analysis. Three 3D shape features: Bounding ellipsoid volume ratio (BEVR), sphericity and spherical disproportion, computed from both the abnormality regions were found to be significant for OS prognosis in GBM patients.

Brain needle intervention is a specific diagnosis and therapy procedure in brain disorders, such as brain tumors and Parkinson’s disease. Preoperative needle path planning is a vital step to guarantee the patient’s safety and reduce lesions. For positioning accuracy in the CT/MRI environment, we have developed a novel needle intervention robot in our previous work. Because the robot is currently designed for the rigid needle, the task of preoperative path-planning is to search for an optimal Remote Center of Motion (RCM) for needle insertion. Therefore, this work proposes an RCM recommendation system using deep reinforcement learning. Considering the robot kinematics, this system takes the following criteria/constraints into consideration: clinical obstacle (blood vessels, tissues) avoidance (COA), mechanically inverse kinematics (MIK) and mechanically less motion (MLM) for the robot. We design a reward function to combine the above three criteria based on their corresponding importance level and utilize proximal policy optimization (PPO) as the main agent of reinforcement learning (RL). RL methods are proved to be competent in searching the RCM, which satisfies the above criteria simultaneously. On the one hand, the results present that RL agents obtain the success rate of finishing the designed task at 93%, which has reached the human level in the tests. On the other hand, the RL agents have the remarkable capability of combining more complex criteria/constraints in future work.

IntroductionCurrent guidelines recommend blood pressure (BP) lowering in patients after acute intracerebral haemorrhage (ICH) without guidance on initial choice of antihypertensive class. This study sought to determine if initial antihypertensive class differentially effects acute BP lowering in a large multiethnic ICH cohort.MethodsSubjects enrolled in the Ethnic/Racial Variations in ICH study between August 2010 and August 2017 with elevated admission BP and who received labetalol, nicardipine or hydralazine monotherapy as initial antihypertensive were analysed. Primary outcomes were systolic and diastolic BP changes from baseline to first BP measurement after initial antihypertensive treatment. Secondary outcomes included haematoma expansion (HE), hospital length of stay (LOS) and modified Rankin Score (mRS) up to 12 months after ICH. Exploratory outcomes assessed effects of race/ethnicity. Linear and logistic regression analyses, adjusted for relevant covariates, were performed to determine associations of antihypertensive class with outcomes.ResultsIn total, 1156 cases were used in analyses. Antihypertensive class was associated with diastolic BP change (p=0.003), but not systolic BP change (p=0.419). Initial dosing with nicardipine lowered acute diastolic BP than labetalol (least square mean difference (labetalol-nicardipine)=5.47 (2.37, 8.57), pConclusionIn this large acute ICH cohort, initial antihypertensive class was associated with acute diastolic, but not systolic, BP-lowering suggesting differential effects of antihypertensive agents.Trial registration numberNCT01202864.

Segmentation of intracerebral hemorrhage (ICH) helps improve the quality of diagnosis, draft the desired treatment methods, and clinically observe the variations with healthy patients. The clinical utilization of various ICH progression scoring systems has limitations due to the systems' modest predictive value. This paper proposes a single pipeline of a multi-task model for end-to-end hemorrhage segmentation and risk estimation. We introduce a 3D spatial attention unit and integrate it into the state-of-the-art segmentation architecture, UNet, to enhance the accuracy by bootstrapping the global spatial representation. We further extract the geometric features from the segmented hemorrhage volume and fuse them with clinical features such as CT angiography (CTA) spot, Glasgow Coma Scale (GCS), and age to predict the ICH stability. Several state-of-the-art machine learning techniques such as multilayer perceptron (MLP), support vector machine (SVM), gradient boosting, and random forests are applied to train stability estimation and to compare the performances. To align clinical intuition with model learning, we determine the shapely values (SHAP) and explain the most significant features for the ICH risk scoring system. A total of 79 patients are included, of which 20 are found in critical condition. Our proposed single pipeline model achieves a segmentation accuracy of 86.3%, stability prediction accuracy of 78.3%, and precision of 82.9%; the mean square error of exact expansion rate regression is observed to be 0.46. The SHAP analysis reveals that CTA spot sign, age, solidity, location, and length of the first axis of the ICH volume are the most critical characteristics that help define the stability of the stroke lesion. We also show that integrating significant geometric features with clinical features can improve the ICH progression scoring by predicting long-term outcomes. Graphical abstract Overview of our proposed method comprising of spatial attention and feature extraction mechanisms. The architecture is trained on the input CT images, and the first step output is the predicted segmentation of the hemorrhagic region. The output is fed into a geometric feature extractor and is fused with clinical features to estimate ICH stability using a multilayer perceptron (MLP).

Following the outbreak of coronavirus 2019 (COVID-19), there is strong evidence of neurological involvement in these patients. We aimed to determine the clinical characteristics of neurological manifestations in COVID-19.A systematic review of studies reporting neurological manifestations published between 1 December, 2019 and 11 May, 2020 was performed. Studies were grouped based on neurological manifestation. Pooled analyses of individual patient's clinical characteristics and olfactory and gustatory dysfunction prevalence were performed.Of 486 studies identified, 48 were included. 70 patients with 73 neurological manifestations were reported. 39 (53.4%) patients had stroke, 18 (24.7%) had Guillain-Barré syndrome and variants, 11 (15.1%) had meningitis, encephalitis, encephalopathy, or myelitis, and five (6.8%) had seizures. They had a mean age of 61.9 ± 17.7 years (60.6% male). Neurological disease occurred 8.1 ± 6.8 days from initial symptoms. Average mortality rate was 17.8%. Stroke has a mortality rate of 25.6%. Olfactory and gustatory dysfunction occurred in 59.9% and 57.5%, respectively.Stroke is the most frequently reported neurological manifestation in COVID-19 and has the highest mortality rate. Neurological manifestations tend to develop one to two weeks after the onset of respiratory disease. There is significant morbidity and mortality associated with COVID-19 neurological manifestations.

Background In the United States, nearly 25,000 patients annually undergo percutaneous ventriculostomy for the management of increased intracranial pressure with little consensus on extraventricular drain management. To characterize relationships between external ventricular drain management, permanent ventriculoperitoneal shunt placement, and hospital disposition, we hypothesized that patients requiring extended drainage would have greater association with ventriculoperitoneal shunt placement and unfavorable disposition. Methods Adult patients admitted to the Duke University Hospital Neuroscience Intensive Care Unit between 2008 and 2010 with extraventricular drains were analyzed. A total of 115 patient encounters were assessed to determine relative impact of age, sex, days of extraventricular placement, weaning attempts, cerebrospinal fluid drainage volumes, Glasgow Coma Scale, and physician’s experience on disposition at discharge and ventriculoperitoneal shunt placement. Univariate logistic regression was first used to test the effect of each variable on the outcome, followed by backward selection to determine a final multivariable logistic regression. Variables in the final model meeting p < 0.05 were declared as significant factors for the outcome. Results Increased extraventricular drain duration (odds ratio [OR] = 1.17, confidence interval [CI] = 1.05–1.30, p = 0.0049) was associated with ventriculoperitoneal shunt placement, while older age (OR = 1.05, CI = 1.02–1.08, p = 0.0027) and less physician extraventricular drain management experience (OR = 4.04, CI = 1.67–9.79, p = 0.0020) were associated with unfavorable disposition. Conclusion In a small cohort, exploratory analyses demonstrate potentially modifiable factors are associated with important clinical outcomes. These findings warrant further study to refine how such factors affect patient outcomes.

Background Gliomas consist of a heterogeneous group of tumors. This study aimed to report the incidences of O6-methylguanine-DNA-methyltransferase (MGMT) promoter methylation, 1p19q co-deletion, isocitrate dehydrogenase (IDH) gene mutations, and inactivating mutations of alpha-thalassemia/mental retardation syndrome X-linked (ATRX) in high-grade gliomas in an ethnically diverse population. Methods Records of patients who underwent surgery for high-grade gliomas from January 2013 to March 2017 at our institution were obtained. The patients’ age, gender, ethnicity, Karnofsky Performance Scale (KPS) score, ability to perform activities of daily living (ADLs), tumor location and biomarkers status were recorded. Data were analyzed using chi-square and Mann-Whitney U tests, Kaplan-Meier estimates and log-rank test. Results 181 patients were selected (56 with grade III gliomas, 125 with grade IV gliomas). In the grade III group, 55% had MGMT promoter methylation, 41% had 1p19q co-deletion, 35% had IDH1 mutation and none had ATRX loss. In the grade IV group, 30% had MGMT promoter methylation, 2% had 1p19q co-deletion, 15% had IDH1 mutation and 8% had ATRX loss. After adjusting for effects of age, surgery and pre-operative ADL statuses, only MGMT promoter methylation was found to be significantly associated with longer overall survival time in grade III (p = 0.024) and IV patients (p = 0.006). Conclusions The incidences of MGMT promoter methylation and IDH1 mutation were found to be comparable to globally reported rates, but those of 1p19q co-deletion and ATRX loss seemed to be lower in our cohort. MGMT promoter methylation was associated with increased overall survival in our cohort and might serve as favorable prognostic factor.

Accurate localization of subthalamic nucleus (STN) is a key prior in deep brain stimulation (DBS) surgery for the patients with advanced Parkinson’s disease (PD). Microelectrode recordings (MERs) along with preplanned trajectories are often employed for the STN localization and it remains challenging task. These MER signals are nonstationary and multicomponent in nature. In this study, we propose a system based on time-frequency features of MERs to differentiate the STN and non-STN regions. We assessed the system with 50 MER trajectories from 26 PD patients who have undergone DBS surgery. The signals are pre-processed and subjected to six-level wavelet decomposition. Then, the entropy is computed from the detailed and approximate coefficients. These features are fed to the random forest classifier and the model is evaluated by leave one patient out cross-validation. The results show that entropy associated with detailed wavelet coefficients (D 1 and D 2 ) are higher in STN where as it is lower in other wavelet scales. All extracted features except entropy from approximate coefficients are found to have significant difference between non-STN and STN (p

Background Surgery for brain metastases aims to reduce mass effect and achieve local control through maximizing resection. There is increasing recognition that awake craniotomy (AC) is especially relevant for resection of brain metastases in eloquent areas. This study seeks to examine the neurologic outcomes of using AC for brain metastases resection. Methods A systematic search for studies examining the role of AC in patients with brain metastases was conducted via PubMed without limitations on the year of publication, language, or study design, using the following search terms: (cerebral OR brain) and (metastases OR tumor) and (awake OR intraoperative OR eloquent OR supramarginal). Studies were included if patients underwent AC for resection of brain metastases and data on pre- and postoperative neurologic function were available. Results Seven studies published between 2001 and 2017 with 104 patients who underwent 107 ACs were identified. Gross total resection was achieved in 61% of patients, supramarginal resection was achieved in 32%, and subtotal resection was achieved in 7%. Immediately after operation, 73% of patients experienced no change or improvement in neurologic outcomes, whereas 27% experienced worsening. In the long term, 96% of those with postoperative worsening of function experienced improvements in neurologic function. Conclusions Most patients experienced improvements in neurologic function immediately after AC, and most patients that experienced short-term worsening of neurologic function after the procedure showed excellent recovery. AC should be considered as a technique to optimize outcomes in brain metastases in eloquent areas.

Hematoma expansion (HE) occurs in approximately one-third of patients with intracerebral hemorrhage (ICH) and is known to be a strong predictor of neurological deterioration as well as poor functional outcome. This study aims to externally validate three risk prediction models of HE (PREDICT, 9-point, and BRAIN scores) in an Asian population. A prospective cohort of 123 spontaneous ICH patients admitted to a tertiary hospital (certified stroke center) in Singapore was recruited. Logistic recalibrations were performed to obtain updated calibration slopes and intercepts for all models. The discrimination (c-statistic), calibration (Hosmer–Lemeshow test, le Cessie–van Houwelingen–Copas–Hosmer test, Akaike information criterion), overall performance (Brier score, R2), and clinical usefulness (decision curve analysis) of the risk prediction models were examined. Overall, the recalibrated PREDICT performed best among the three models in our study cohort based on the novel matrix comprising of Akaike information criterion and c-statistic. The PREDICT model had the highest R2 (0.26) and lowest Brier score (0.14). Decision curve analyses showed that recalibrated PREDICT was more clinically useful than 9-point and BRAIN models over the greatest range of threshold probabilities. The two scores (PREDICT and 9-point) which incorporated computed tomography (CT) angiography spot sign outperformed the one without (BRAIN). To our knowledge, this is the first study to validate HE scores, namely PREDICT, 9-Point and BRAIN, in a multi-ethnic Asian ICH patient population. The PREDICT score was the best performing model in our study cohort, based on the performance metrics employed in this study. Our findings also showed support for CT angiography spot sign as a predictor of outcome after ICH. Although the models assessed are sufficient for risk stratification, the discrimination and calibration are at best moderate and could be improved.

We report the development and implementation of an epi-detected spectral-focusing hyperspectral stimulated Raman scattering (SRS) imaging technique for label-free biomolecular subtyping of glioblastomas (GBMs). The hyperspectral SRS imaging technique developed generates SRS image stacks (from 2800 to 3020 cm–1 at 7 cm–1 intervals) within 30 s through controlling the time delay between the chirped pump and Stokes beams. SRS images at representative Raman shifts (e.g., 2845, 2885, and 2935 cm–1) delineate the biochemical variations and morphological differences between proneural and mesenchymal subtypes of GBMs. Multivariate curve resolution (MCR) analysis on hyperspectral SRS images enables the quantification of major biomolecule distributions in mesenchymal and proneural GBMs. Further principal component analysis (PCA) and linear discriminant analysis (LDA) together with leave-one SRS spectrum-out, cross-validation (LOOCV) yields a diagnostic sensitivity of 96.7% (29/30) and specificity of 88.9% (28/36) f...

Robotic systems play an increasingly important role in improving feasibility and effectiveness of minimally invasive neurosurgery (MIN). However, large footprint, bulky size, and complex mechanisms limit the clinical application of existing robotic neurosurgery solutions. This paper proposes a novel skull-mounted robot with a compact and lightweight parallel mechanism for positioning of surgical tools in MIN. The system serves as a mechanical guide for automatic positioning of needles, catheters, probes, or electrodes. A parallel mechanism with 4 degrees of freedom (DOFs) is adopted, with the aim of providing sufficient accuracy and load capacity. The volume of the robot is only 50 mm × 50 mm × 40 mm and the weight is 73 g. The miniature design allows the robot to be mounted on the skull easily without consuming space in the operating room while avoiding the patient's immobilization, simplifying the registration operation, and increasing patient comfort and tolerability. The mechanical design, kinematics and workspace are analyzed and described in detail. Three experiments on the prototype are conducted to test the stiffness, accuracy and performance. Results show that the deflection is less than 0.1 mm for holding common surgical tools and the tracking errors are less than 1.2 mm and 1.9° which is acceptable for MIN. The robot can be easily and firmly mounted on the skull model and cadaver head, and flexibly manipulated on the skull model.

Intraoperative frozen section assessment, to confirm acquisition of pathological tissues, is used in stereotactic brain biopsy to minimise sampling errors. Limitations include the dependence on dedicated neuro-oncology pathologists and an increase in operative duration. We investigated the use of intraoperative fluorescein sodium, and compared it to frozen section assessment, for confirming pathological tissue samples in the stereotactic biopsy of gadolinium-contrast-enhancing brain lesions. This prospective observational study consisted of 18 consecutive patients (12 men; median age, 63 years) who underwent stereotactic biopsy of gadolinium-contrast-enhancing brain lesions with intravenous fluorescein sodium administration. Twenty-three specimens were obtained and examined for the presence of fluorescence using a microscope with fluorescence visualisation capability. Positive and negative predictive values were calculated based on the fluorescence status of the biopsy samples with its corresponding intraoperative frozen section and definitive histopathological diagnosis. Nineteen specimens (83%) were fluorescent and four (17%) were non-fluorescent. All 19 fluorescent specimens were confirmed to be lesional on intraoperative frozen section assessment and were suitable for histopathological diagnosis. Three of the non-fluorescent specimens were confirmed to be lesional on intraoperative frozen section assessment. One non-fluorescent specimen was non-diagnostic on frozen section and histological assessments. The positive predictive value was 100% and the negative predictive value was 25%. Fluorescein sodium fluorescence is as accurate as frozen section assessment in confirming sampling of pathological tissue in the stereotactic biopsy of gadolinium-contrast-enhancing brain lesions. Fluorescein sodium fluorescence-guided stereotactic biopsy is a useful addition to the neurosurgical armamentarium.

OBJECTIVE This study aimed to characterize pyramidal tract shift in different regions of the brain during glioma resection and its association with head position and tumor location. METHODS From 2008-2013, 14 patients presenting at the National Neuroscience Institute with high-grade glioma (World Health Organization III or IV) underwent preoperative and intraoperative diffusion tensor imaging. A novel method of placing landmarks along the preoperative and intraoperative tracts, with anterior commissure as the origin, was used to determine pyramidal tract shift. Shift was evaluated in x (lateromedial), y (anteroposterior), and z (craniocaudal) directions for 3 brain regions: brainstem, around third and lateral ventricles, and above ventricles. Shift radius is calculated as the distance between preoperative and postoperative landmarks. RESULTS Mean shift radius was 2.72 ± 0.55, 2.98 ± 0.53, and 4.04 ± 0.58 mm at the brainstem, third and lateral ventricles, and above the ventricles, respectively (P

Gangliogliomas comprise of less than 1% of all primary brain tumours. Malignant transformation is rare. Correspondingly, secondary gliosarcomas, especially in the absence of previous irradiation exposure is too, exceedingly rare. Recent evidence supports the use of mutant IDH1 testing in suspected adult gangliogliomas as a distinguishing feature between ganglioglioma and early diffuse astrocytoma. The authors report a case of a ganglioglioma (WHO I) in the cingulate gyrus that underwent malignant transformation and propagated into a multi-lesional gliosarcoma (WHO IV glioblastoma variant). A previously well 59-year-old lady presented with first-onset focal seizures secondary to a left cingulate gyrus lesion. Stereotactic biopsy of the lesion demonstrated an IDH1 (R132H) ganglioglioma. The patient remained well until a routine follow-up MRI performed 36 months later showed significant progression of the initial lesion and a new, separate lesion in the left temporoparietal lobe. Surgical excision was performed. The final histology reported a gliosarcoma. In view of the unusual findings, the case is reviewed corroboratively with current disease understanding. Keywords: Ganglioglioma, Gliosarcoma, Isocitrate dehydrogenase 1

Background: The thalamus has been a surgical target for the treatment of various movement disorders. Commonly used therapeutic modalities include ablative and nonablative procedures. A major clinical side effect of thalamic surgery is the appearance of speech problems. Objective: This review summarizes the data on the development of speech problems after thalamic surgery. Methods: A systematic review and meta-analysis was performed using nine databases, including Medline, Web of Science, and Cochrane Library. We also checked for articles by searching citing and cited articles. We retrieved studies between 1960 and September 2014. Results: Of a total of 2,320 patients, 19.8% (confidence interval: 14.8–25.9) had speech difficulty after thalamotomy. Speech difficulty occurred in 15% (confidence interval: 9.8–22.2) of those treated with a unilaterally and 40.6% (confidence interval: 29.5–52.8) of those treated bilaterally. Speech impairment was noticed 2- to 3-fold more commonly after left-sided procedures (40.7% vs. 15.2%). Of the 572 patients that underwent DBS, 19.4% (confidence interval: 13.1–27.8) experienced speech difficulty. Subgroup analysis revealed that this complication occurs in 10.2% (confidence interval: 7.4–13.9) of patients treated unilaterally and 34.6% (confidence interval: 21.6–50.4) treated bilaterally. After thalamotomy, the risk was higher in Parkinson's patients compared to patients with essential tremor: 19.8% versus 4.5% in the unilateral group and 42.5% versus 13.9% in the bilateral group. After DBS, this rate was higher in essential tremor patients. Conclusion: Both lesioning and stimulation thalamic surgery produce adverse effects on speech. Left-sided and bilateral procedures are approximately 3-fold more likely to cause speech difficulty. This effect was higher after thalamotomy compared to DBS. In the thalamotomy group, the risk was higher in Parkinson's patients, whereas in the DBS group it was higher in patients with essential tremor. Understanding the pathophysiology of speech disturbance after thalamic procedures is a priority. © 2017 International Parkinson and Movement Disorder Society

The prevalence of asymptomatic unruptured intracranial aneurysms (AUIA) in a Southeast Asian population has not been previously studied. Knowing the disease burden and population at risk can assist us in making informed decisions when managing AUIAs. We aimed to determine if the local prevalence of AUIAs differed from other populations in the published literature.Magnetic resonance angiography radiology reports and images for 4572 patients between January 2013 and January 2014 were reviewed for AUIAs.The overall prevalence of AUIAs was 3.5% (160/4572). It was significantly higher in women (4.5%) compared with men (2.6%) (P0.001). The mean aneurysm size was 3.2 ± 1.7 mm. Of the aneurysms, 88.5% (146/165) measured less than 5 mm, 9.7% (16/165) measured 5-9 mm, and 1.8% (3/165) measured more than 9 mm. Most of the aneurysms were located in the cavernous segment (72/165, 43.6%) and ophthalmic segment (34/165, 20.6%) of the internal carotid artery, followed by the middle cerebral artery (16/165, 9.7%). The posterior circulation contributed 9.1% (15/165) of the AUIAs. During the study period, 1 aneurysm ruptured and the rest of the aneurysms did not show any growth.In this large cohort of a Southeast Asian population, the prevalence of AUIA was 3.5%. Most of the AUIAs were less than 5 mm and did not require intervention at time of discovery and did not progress within the year of follow-up.

Deep brain stimulation of the subthalamic nucleus (STN DBS) has been shown to reduce antiparkinsonian medication in Parkinson's disease. We aimed to investigate the changes in long-term medication profile with STN DBS.Antiparkinsonian medication data for 56 patients were collected from as early as 3 years before STN DBS up to 10 years after. Cost spending on medication changes was analyzed. Mean levodopa equivalent daily dose (LEDD) was projected 10 years into the future based on preoperative data to create a comparator group wherein the patients did not undergo STN DBS. Use of neuroleptics and antidepressants was also recorded.LEDD requirement was significantly reduced by a mean of 31 ± 2% over 10 years after DBS, from 1049 ± 381 mg at pre-DBS baseline, to 713 ± 392 mg at 1 year post-DBS, and 712 ± 385 mg at 10 years post-DBS. This was associated with a mean reduction of 35 ± 3% in medicine cost. Modeled LEDD requirements for not having STN DBS were in the range of 1489 mg to 2721 mg at 10 years post-DBS (109-282% higher than the observed mean LEDD in DBS cohort). The proportion of patients increased from 5% before STN DBS to 14% at 10 year post-DBS for neuroleptics, and 11-23% for antidepressants.STN DBS led to LEDD reduction and antiparkinsonian medication cost savings in our South-East Asian cohort. Medication reduction with STN DBS in our cohort over the 10-year period was comparable to those reported in Western populations.

We develop a deep learning approach for automated intracerebral hemorrhage (ICH) segmentation from 3D computed tomography (CT) scans. Our model, ICHNet, evolves by integrating dilated convolution neural network (CNN) with hypercolumn features where a modest number of pixels are sampled and corresponding features from multiple layers are concatenated. Due to freedom of sampling pixels rather than image patch, this model trains within the brain region and ignores the CT background padding. This boosts the convergence time and accuracy by learning only healthy and defected brain tissues. To overcome the class imbalance problem, we sample an equal number of pixels from each class. We also incorporate 3D conditional random field (3D CRF) to smoothen the predicted segmentation as a post-processing step. ICHNet demonstrates 87.6% Dice accuracy in hemorrhage segmentation, that is comparable to radiologists.

Objective: This study seeks to systematically review the selection of features and algorithms for machine learning and automation in deep brain stimulation surgery (DBS) for Parkinson’s disease. This will assist in consolidating current knowledge and accuracy levels to allow greater understanding and research to be performed in automating this process, which could lead to improved clinical outcomes. Methods: A systematic literature review search was conducted for all studies that utilized machine learning and DBS in Parkinson’s disease. Results: Ten studies were identified from 2006 utilizing machine learning in DBS surgery for Parkinson’s disease. Different combinations of both spike independent and spike dependent features have been utilized with different machine learning algorithms to attempt to delineate the subthalamic nucleus (STN) and its surrounding structures. Conclusion: The state-of-the-art algorithms achieve good accuracy and error rates with relatively short computing time, however, the currently achievable accuracy is not sufficiently robust enough for clinical practice. Moreover, further research is required for identifying subterritories of the STN. Significance: This is a comprehensive summary of current machine learning algorithms that discriminate the STN and its adjacent structures for DBS surgery in Parkinson’s disease. Ministry of Health (MOH) National Medical Research Council (NMRC) This research is supported by the Singapore Ministry of Health’s National Medical Research Council (NMRC/CNIG/1173/2017).

OBJECTIVEThe authors explored the feasibility of seizure detection and prediction using signals recorded from the anterior thalamic nucleus, a major target for deep brain stimulation (DBS) in the treatment of epilepsy.METHODSUsing data from 5 patients (13 seizures in total), the authors performed a feasibility study and analyzed the performance of a seizure prediction and detection algorithm applied to simultaneously acquired scalp and thalamic electroencephalography (EEG). The thalamic signal was obtained from DBS electrodes. The applied algorithm used the similarity index as a nonlinear measure for seizure identification, with patient-specific channel and threshold selection. Receiver operating characteristic (ROC) curves were calculated using data from all patients and channels to compare the performance between DBS and EEG recordings.RESULTSThalamic DBS recordings were associated with a mean prediction rate of 84%, detection rate of 97%, and false-alarm rate of 0.79/hr. In comparison, scalp EEG recordings were associated with a mean prediction rate of 71%, detection rate of 100%, and false-alarm rate of 1.01/hr. From the ROC curves, when considering all channels, DBS outperformed EEG for both detection and prediction of seizures.CONCLUSIONSThis is the first study to compare automated seizure detection and prediction from simultaneous thalamic and scalp EEG recordings. The authors have demonstrated that signals recorded from DBS leads are more robust than EEG recordings and can be used to predict and detect seizures. These results indicate feasibility for future designs of closed-loop anterior nucleus DBS systems for the treatment of epilepsy.

Background The ventral intermediate nucleus of the thalamus is not readily visible on structural magnetic resonance imaging. Therefore, a method for its visualization for stereotactic targeting is desirable. Objective The objective of this study was to define a tractography-based methodology for the stereotactic targeting of the ventral intermediate nucleus. Methods The lateral and posterior borders of the ventral intermediate nucleus were defined by tracking the pyramidal tract and medial lemniscus, respectively. A thalamic seed was then created 3 mm medial and anterior to these borders, and its structural connections were analyzed. The application of this method was assessed in an imaging cohort of 14 tremor patients and 15 healthy controls, in which we compared the tractography-based targeting to conventional targeting. In a separate surgical cohort (3 tremor and 3 tremor-dominant Parkinson's disease patients), we analyzed the accuracy of this method by correlating it with intraoperative neurophysiology. Results Tractography of the thalamic seed revealed the tracts corresponding to cerebellar input and motor cortical output fibers. The tractography-based target was more lateral (12.5 [1.2] mm vs 11.5 mm for conventional targeting) and anterior (8.5 [1.1] mm vs 6.7 [0.3] mm, anterior to the posterior commissure). In the surgical cohort, the Euclidian distance between the ventral intermediate nucleus identified by tractography and the surgical target was 1.6 [1.1] mm. The locations of the sensory thalamus, lemniscus, and pyramidal tracts were concordant within

Traumatic brain injury (TBI) is a global concern associated with high mortality and morbidity. Costs to individuals and society are extensive due to poor recovery, long-term disability and the young age group affected. Statins have emerged as potential therapeutic agents in TBI. This study aimed to investigate the protective effect of statins in severe TBI.This case-control study included adults with severe TBI. A sliding dichotomy approach was used to dichotomize mortality at 14-days and Glasgow Outcome Score (GOS) at 6 months. Logistic regression analysis was used to calculate the odds ratios (OR) for 14-day mortality and 6-month GOS.Equivalent cohorts of 59 age- and sex-matched statin and non-statin users were selected, resulting in population of 118 (mean age = 70.2 years, SD = 10.3), with a median Glasgow Coma Score of 5. Statins did not reduce the likelihood of mortality at 14 days (adjusted OR = 1.23, p = 0.68) or unfavourable outcome at 6 months (adjusted OR = 1.19, p = 0.78).Despite increasing evidence for benefit of statins in TBI, this study in an Asian population does not support this association, demonstrating no significant improvement in outcome for statin users. Further research is required to understand the mechanisms and impact of statins in TBI.

OBJECT Chronic subdural hemorrhage (SDH) or hematoma is a condition that affects elderly individuals. With advances in medical care, the number of nonagenarians and centenarians will increase. However, surgical treatments in this age group are associated with high rates of morbidity and mortality. Because no data are available on the rates of survival among elderly patients with chronic SDHs who undergo surgical drainage or receive only conservative care, the goal of this study was to determine survival rates in patients 90 years of age or older with symptomatic chronic SDHs. METHODS The authors conducted a retrospective analysis of patient data that were collected at 3 hospitals over a 13-year period (from January 2001 to June 2013). The data from patients 90 years or older with symptomatic chronic SDHs and who were offered surgical treatment were included in the analysis. Patients who underwent surgical treatment were included in the surgical group and patients who declined an operation were included in the conservative care group. The patients’ Charlson Comorbidity Index score, Karnofsky Performance Scale score, dates of death, presenting symptoms, Glasgow Coma Scale score, length of stay in the hospital, discharge location, side of the SDH, and neurological improvements at 30-day and 6-month follow-ups were recorded. Data were statistically analyzed with Fisher exact test, Kaplan-Meier curves, and logistic regression. RESULTS In total, 101 patients met the inclusion criteria of this study; 70 of these patients underwent surgical drainage, and 31 received conservative care. Patients in the surgical group had statistically significantly (p < 0.001) higher survival at both the 30-day and 6-month follow-ups, with 92.9% and 81.4% of the patients in this group surviving for at least 30 days and 6 months, respectively, versus 58.1% and 41.9%, respectively, in the conservative care group. Moreover, the mean overall length of survival of 34.4 ± 28.7 months was longer in the surgical group than it was in the conservative care group (11.3 ± 16.6 months). Overall, 95.7% of patients in the surgical group exhibited an improvement in neurological status after the SDH drainage, whereas none of the patients in the conservative care group showed any neurological improvement during their hospital stay. The surgical complication rate was 11.4%, and the overall rate of chronic SDH recurrence after surgery was 12.9%. CONCLUSIONS Surgical drainage of chronic SDHs in nonagenarians and centenarians is associated with lower incidence of inpatient death and higher 30-day and 6-month survival rates.

Intrathecal baclofen (ITB) therapy is a proven, effective treatment for disabling cortical spasticity. We describe the first local series of five patients with acquired brain injury (ABI) who received ITB and were followed up for 63.8 months.A retrospective review of medical and rehabilitation records of patients who received ITB therapy was carried out. Data studied included baseline demographic and injury variables, implantation data, spasticity and function, ITB dosage over time and complications.From 2006 to 2010, a total of five patients received ITB therapy via implanted pumps about 39.4 months after ABI. Four out of five patients experienced significant reductions in their lower limb spasticity scores and improvements in global function and dependency. One patient had minor adverse events associated with baclofen-related sedation. The mean ITB dose at one year was 182.7 ± 65.6 mcg/day.Our preliminary study showed encouraging long-term outcomes and safety for ITB therapy after ABI-related intractable spasticity. Individual ITB responses over time were variable, with gender differences. The outcomes experienced by our centre were comparable to those in the general ABI population, supporting the efficacy of ITB therapy for chronic disabling spasticity.

Background Deep brain stimulation (DBS) to the subthalamic nucleus (STN) is an effective neurosurgery that overcomes the motor system alternations of patients with advanced Parkinson’s disease. The most challenging aspect of DBS surgery is the accurate identification of STN and its borders. In general, it is performed manually by a neurophysiologist using the microelectrode recordings (MERs). This process is subjective, and tedious and further, interpretation of MERs is difficult because of its inherent nonstationary variations. New methods In this work, the wavelet-packet based features are proposed to automatically localize the STN and its subcortical structures using microelectrode recorded signals during DBS surgery. The study analyses 2904 MERs of 26 PD patients who underwent DBS implantation. The low and high order statistical parameters are extracted from the wavelet packet coefficients of MERs and used in the classifications, namely, non-STN vs. STN, pre-STN vs. STN and STN vs. post-STN. Results Most of the features are significantly different in STN and its subcortical regions, namely, pre-STN and post-STN. The proposed features achieve an average accuracy of 85 % in non-STN vs. STN, 87.2 % in pre-STN vs. STN and 77.7 % in STN vs. post-STN. The accuracy is improved by around 10 % in non-STN vs. STN and STN vs. post-STN when the transition error is 1 mm. Comparison with existing methods The proposed features are found to be better than the wavelet features. Conclusions The proposed approach could be a potential useful adjunct for the real-time rapid intraoperative identification of STN and its anatomical borders.

The use of anticoagulation and antiplatelet agents (ACAP) has steadily increased over recent years. However, the effects of ACAP on traumatic brain injuries (TBI) are not well investigated. The aim of this study was to investigate the effects of pre-injury ACAP use on clinical outcome and mortality in severe TBI. A retrospective case-control study was performed for all patients who presented with severe TBI (GCS 8) to the National Neuroscience Institute, Singapore, between 2006 and 2009. Patients with pre-injury ACAP use were compared to matched controls. Outcome measures were mortality at 14 days and 6 months, and Glasgow Outcome Score (GOS) at 6 months using a sliding dichotomy approach. Univariate analysis was performed using Chi-square and student's t-test and logistic regression was used to model the effect of ACAP on mortality rate. Forty-five patients with pre-injury use of ACAP were compared with matched controls. The mortality at 14 days (OR = 0.5, 95% CI 0.2-1.4) and 6 months (OR = 0.7, 95% CI 0.2-1.9) were not significantly different between the 2 groups. Using the sliding dichotomy approach, there was no difference in the odds for unfavorable functional outcomes at 6 months (OR = 1.2, 95% CI 0.4-3.7). In this case-control study, the use of ACAP did not have a significant effect on mortality and adverse outcomes in patients with severe TBI. This would suggest that in severe TBI, ACAP use may not contribute significantly to the overall prognosis.

Hematoma expansion (HE) occurs in approximately one-third of patients with intracerebral hemorrhage (ICH) and is known to be a strong predictor of neurological deterioration as well as poor functional outcome. This study aims to externally validate three risk prediction models of HE (PREDICT, 9-point, and BRAIN scores) in an Asian population.A prospective cohort of 123 spontaneous ICH patients admitted to a tertiary hospital (certified stroke center) in Singapore was recruited. Logistic recalibrations were performed to obtain updated calibration slopes and intercepts for all models. The discrimination (c-statistic), calibration (Hosmer-Lemeshow test, le Cessie-van Houwelingen-Copas-Hosmer test, Akaike information criterion), overall performance (Brier score, ROverall, the recalibrated PREDICT performed best among the three models in our study cohort based on the novel matrix comprising of Akaike information criterion and c-statistic. The PREDICT model had the highest RTo our knowledge, this is the first study to validate HE scores, namely PREDICT, 9-Point and BRAIN, in a multi-ethnic Asian ICH patient population. The PREDICT score was the best performing model in our study cohort, based on the performance metrics employed in this study. Our findings also showed support for CT angiography spot sign as a predictor of outcome after ICH. Although the models assessed are sufficient for risk stratification, the discrimination and calibration are at best moderate and could be improved.