Faculty Bibliography

BACKGROUND:Early prediction of hematoma expansion (HE) following nontraumatic intracerebral hemorrhage (ICH) may inform preemptive therapeutic interventions. We sought to identify how accurately machine learning (ML) radiomics models predict HE compared with expert clinicians using head computed tomography (HCT). METHODS:We used data from 900 study participants with ICH enrolled in the Antihypertensive Treatment of Acute Cerebral Hemorrhage 2 Study. ML models were developed using baseline HCT images, as well as admission clinical data in a training cohort (n = 621), and their performance was evaluated in an independent test cohort (n = 279) to predict HE (defined as HE by 33% or > 6 mL at 24 h). We simultaneously surveyed expert clinicians and asked them to predict HE using the same initial HCT images and clinical data. Area under the receiver operating characteristic curve (AUC) were compared between clinician predictions, ML models using radiomic data only (a random forest classifier and a deep learning imaging model) and ML models using both radiomic and clinical data (three random forest classifier models using different feature combinations). Kappa values comparing interrater reliability among expert clinicians were calculated. The best performing model was compared with clinical predication. RESULTS:The AUC for expert clinician prediction of HE was 0.591, with a kappa of 0.156 for interrater variability, compared with ML models using radiomic data only (a deep learning model using image input, AUC 0.680) and using both radiomic and clinical data (a random forest model, AUC 0.677). The intraclass correlation coefficient for clinical judgment and the best performing ML model was 0.47 (95% confidence interval 0.23-0.75). CONCLUSIONS:We introduced supervised ML algorithms demonstrating that HE prediction may outperform practicing clinicians. Despite overall moderate AUCs, our results set a new relative benchmark for performance in these tasks that even expert clinicians find challenging. These results emphasize the need for continued improvements and further enhanced clinical decision support to optimally manage patients with ICH.

BACKGROUND:In many parts of the world including India, the prevalence of autoimmune inflammatory diseases such as neuromyelitis optica spectrum disorders (NMOSD), myelin oligodendrocyte glycoprotein antibody-associated disease (MOGAD), and multiple sclerosis (MS) is rising. A diagnosis is often delayed due to insufficient diagnostic tools. Machine learning (ML) models have accurately differentiated eyes of patients with MS from those of healthy controls (HCs) using optical coherence tomography (OCT)-based retinal images. Examining OCT characteristics may allow for early differentiation of these conditions. The objective of this study was to determine feasibility of ML analyses to distinguish between patients with different autoimmune inflammatory diseases, other ocular diseases, and HCs based on OCT measurements of the peripapillary retinal nerve fiber layer (pRNFL), ganglion cell-inner plexiform layer (GCIPL), and inner nuclear layers (INLs). METHODS:Eyes of people with MS (n = 99 patients), NMOSD (n = 40), MOGAD (n = 74), other ocular diseases (OTHER, n = 16), and HCs (n = 54) from the Mangalore Demyelinating Disease Registry were included. Support vector machine (SVM) classification models incorporating age, pRNFL, GCIPL, and INL were performed. Data were split into training (70%) and testing (30%) data and accounted for within-patient correlations. Cross-validation was used in training to choose the best parameters for the SVM model. Accuracy and area under receiver operating characteristic curves (AUROCs) were used to assess model performance. RESULTS:The SVM models distinguished between eyes of patients with each condition (i.e., MOGAD vs NMOSD, NMOSD vs HC, MS vs OTHER, etc) with strong discriminatory power demonstrated from the AUROCs for these comparisons ranging from 0.81 to 1.00. These models also performed with moderate to high accuracy, ranging from 0.66 to 0.81, with the exception of the MS vs NMOSD comparison, which had an accuracy of 0.53. CONCLUSIONS:ML models are useful for distinguishing between autoimmune inflammatory diseases and for distinguishing these from HCs and other ocular diseases based on OCT measures. This study lays the groundwork for future deep learning studies that use analyses of raw OCT images for identifying eyes of patients with such disorders and other etiologies of optic neuropathy.

Electroencephalography (EEG) has been instrumental in epilepsy research for the past century, both for basic and translational studies. Its contributions have advanced our understanding of epilepsy, shedding light on the pathophysiology and functional organization of epileptic networks, and the mechanisms underlying seizures. Here we re-examine the historical significance, ongoing relevance, and future trajectories of EEG in epilepsy research. We describe traditional approaches to record brain electrical activity and discuss novel cutting-edge, large-scale techniques using micro-electrode arrays. Contemporary EEG studies explore brain potentials beyond the traditional Berger frequencies to uncover underexplored mechanisms operating at ultra-slow and high frequencies, which have proven valuable in understanding the principles of ictogenesis, epileptogenesis, and endogenous epileptogenicity. Integrating EEG with modern techniques such as optogenetics, chemogenetics, and imaging provides a more comprehensive understanding of epilepsy. EEG has become an integral element in a powerful suite of tools for capturing epileptic network dynamics across various temporal and spatial scales, ranging from rapid pathological synchronization to the long-term processes of epileptogenesis or seizure cycles. Advancements in EEG recording techniques parallel the application of sophisticated mathematical analyses and algorithms, significantly augmenting the information yield of EEG recordings. Beyond seizures and interictal activity, EEG has been instrumental in elucidating the mechanisms underlying epilepsy-related cognitive deficits and other comorbidities. Although EEG remains a cornerstone in epilepsy research, persistent challenges such as limited spatial resolution, artifacts, and the difficulty of long-term recording highlight the ongoing need for refinement. Despite these challenges, EEG continues to be a fundamental research tool, playing a central role in unraveling disease mechanisms and drug discovery.

BACKGROUND:Brain arteriovenous malformations (AVMs) are abnormal connections between feeding arteries and draining veins, associated with significant risks of haemorrhage, seizures and other neurological deficits. Preoperative embolization is commonly used as an adjunct to microsurgical resection, with the aim of reducing intraoperative complications and improving outcomes. However, the efficacy and safety of this approach remain controversial. METHODS:This study is a subanalysis of the Multicenter International Study for Treatment of Brain AVMs consortium. We retrospectively analysed 486 patients with brain AVMs treated with microsurgical resection between January 2010 and December 2023. Patients were divided into two groups: those who underwent microsurgery alone (n=245) and those who received preoperative embolization, followed by microsurgery (n=241). Propensity score matching was employed, resulting in 288 matched patients (144 in each group). The primary outcomes were rates of complete AVM obliteration and functional outcomes (measured by the modified Rankin Scale (mRS)). Secondary outcomes included complication rates, mortality, hospital length of stay and postsurgical rupture. RESULTS:After matching, the complete obliteration rate was 97% with no significant difference between the microsurgery-only group and the preoperative embolization group (p=0.12). The proportion of patients with an mRS score of 0-2 at the last follow-up was similar in both groups (83% vs 84%; p=0.67). The median hospital stay was significantly longer for the embolisation group (9 days vs 7 days; p=0.017). Complication rates (24% vs 22%; p=0.57) and mortality rates (4.9% vs 2.1%; p=0.20) were comparable between the two groups. No significant differences were observed in postsurgical rupture, recurrence or retreatment rates. CONCLUSIONS:In this large multicentre study, preoperative embolization did not significantly improve AVM obliteration rates, functional outcomes or reduce complications compared with microsurgery alone.

BACKGROUND:Hemorrhagic stroke survivors may have cognitive impairment. We sought to identify preadmission and admission factors associated with cognitive impairment after hemorrhagic stroke. DESIGN/METHODS:Patients with nontraumatic intracerebral or subarachnoid hemorrhage (ICH or SAH) were assessed 3-months post-bleed using the Quality of Life in Neurological Disorders (Neuro-QoL) Cognitive Function short form. Univariate and multivariate analysis were used to evaluate the relationship between poor cognition (Neuro-QoL t-score ≤50) and preadmission and admission factors. RESULTS:Of 101 patients (62 ICH and 39 SAH), 51 (50 %) had poor cognition 3-months post-bleed. On univariate analysis, poor cognition was associated with (p < 0.05): age [66.0 years (52.0-77.0) vs. 54.5 years (40.8-66.3)]; private insurance (37.3 % vs. 74.0 %); BMI > 30 (13.7 % vs. 34.0 %); and admission mRS score > 0 (41.2 % vs. 14.0 %), NIHSS score [8.0 (2.0-17.0) vs. 0.5 (0.0-4.0)], and APACHE II score [16.0 (11.0-19.0) vs. 9.0 (6.0-14.3)]. On multivariate analysis, poor cognition was associated with mRS score > 0 [OR 4.97 (1.30-19.0), p = 0.019], NIHSS score [OR 1.14 (1.02-1.28), p = 0.026], private insurance [OR 0.21 (0.06-0.76), p = 0.017] and BMI > 30 [OR 0.13 (0.03-0.56), p = 0.006]. CONCLUSIONS:Cognitive impairment after hemorrhagic stroke is less common in patients with BMI > 30 and private insurance. Heightened surveillance for non-obese patients without private insurance is suggested. Additional investigation into the relationship between cognition and both BMI and insurance type is needed.

Cancer of unknown primary (CUP) constitutes between 2% and 5% of human malignancies and is among the most common causes of cancer death in the United States. Brain metastases are often the first clinical presentation of CUP; despite extensive pathological and imaging studies, 20%-45% of CUP are never assigned a primary site. DNA methylation array profiling is a reliable method for tumor classification but tumor-type-specific classifier development requires many reference samples. This is difficult to accomplish for CUP as many cases are never assigned a specific diagnosis. Recent studies identified subsets of methylation quantitative trait loci (mQTLs) unique to specific organs, which could help increase classifier accuracy while requiring fewer samples. We performed a retrospective genome-wide methylation analysis of 759 carcinoma samples from formalin-fixed paraffin-embedded tissue samples using Illumina EPIC array. Utilizing mQTL specific for breast, lung, ovarian/gynecologic, colon, kidney, or testis (BLOCKT) (185k total probes), we developed a deep learning-based methylation classifier that achieved 93.12% average accuracy and 93.04% average F1-score across a 10-fold validation for BLOCKT organs. Our findings indicate that our organ-based DNA methylation classifier can assist pathologists in identifying the site of origin, providing oncologists insight on a diagnosis to administer appropriate therapy, improving patient outcomes.

OBJECTIVE:This study aimed to externally validate the Memory Assessment Clinics Scale for Epilepsy (MAC-E), a brief self-report measure of subjective memory complaints in adults with epilepsy. METHODS:A cross-sectional study was conducted including adults with focal pharmacoresistant epilepsy from three Level 4 epilepsy centers in the U.S., who completed the MAC-E as part of a clinical neuropsychological evaluation. Confirmatory factor analysis was conducted, and goodness-of-fit criteria were calculated to assess model fit: comparative fit index (CFI), root mean square error of approximation (RMSEA), and standardized root mean residual (SRMR). Item response theory models were constructed, and Mokken analysis was used to assess discrimination and unidimensionality. Internal consistency was evaluated with McDonald's Omega. RESULTS:values for each of the 5 factors (0.58-0.91 and 0.34-0.82, respectively). MAC-E items demonstrated high levels of discrimination as well as the ability to evaluate across the entirety of each latent trait. Score responses were uniformly distributed across latent traits, and unidimensionality was established by factor (all H coefficients > 0.4). Internal consistency was high across factors (omega range: 0.77-0.88). CONCLUSIONS:Results of this study demonstrate good external validation of the MAC-E in an independent, multicenter cohort of adults with epilepsy. These findings provide further support that the MAC-E is a psychometrically valid, self-report instrument to assess every-day memory abilities in adults with epilepsy in both clinical and research settings.

BACKGROUND:Life-threatening, space-occupying mass effect due to cerebral edema and/or hemorrhagic transformation is an early complication of patients with middle cerebral artery stroke. Little is known about longitudinal trajectories of laboratory and vital signs leading up to radiographic and clinical deterioration related to this mass effect. METHODS:We curated a retrospective data set of 635 patients with large middle cerebral artery stroke totaling 95,463 data points for 10 longitudinal covariates and 40 time-independent covariates. We assessed trajectories of the 10 longitudinal variables during the 72 h preceding three outcomes representative of life-threatening mass effect: midline shift ≥ 5 mm, pineal gland shift (PGS) > 4 mm, and decompressive hemicraniectomy (DHC). We used a "backward-looking" trajectory approach. Patients were aligned based on outcome occurrence time and the trajectory of each variable was assessed before that outcome by accounting for cases and noncases, adjusting for confounders. We evaluated longitudinal trajectories with Cox proportional time-dependent regression. RESULTS:Of 635 patients, 49.0% were female, and the mean age was 69 years. Thirty five percent of patients had midline shift ≥ 5 mm, 24.3% of patients had PGS > 4 mm, and 10.7% of patients underwent DHC. Backward-looking trajectories showed mild increases in white blood cell count (10-11 K/UL within 72 h), temperature (up to half a degree within 24 h), and sodium levels (1-3 mEq/L within 24 h) before the three outcomes of interest. We also observed a decrease in heart rate (75-65 beats per minute) 24 h before DHC. We found a significant association between increased white blood cell count with PGS > 4 mm (hazard ratio 1.05, p value 0.007). CONCLUSIONS:Longitudinal profiling adjusted for confounders demonstrated that white blood cell count, temperature, and sodium levels appear to increase before radiographic and clinical indicators of space-occupying mass effect. These findings will inform the development of multivariable dynamic risk models to aid prediction of life-threatening, space-occupying mass effect.