Search

Malignant cerebral edema occurs when brain swelling displaces and compresses vital midline structures within the first week of a large middle cerebral artery stroke. Early interventions such as hyperosmolar therapy or surgical decompression may reverse secondary injury but must be administered judiciously. To optimize treatment and reduce secondary damage, clinicians need strategies to frequently and quantitatively assess the trajectory of edema using updated, relevant information. However, existing risk assessment tools are limited by the absence of structured records capturing the evolution of edema and typically estimate risk at a single time point early in the admission, therefore failing to account for changes in variables over the following hours or days. To address this, we developed and validated dynamic machine learning models capable of accurately predicting the severity of midline structure displacement, an established indicator of malignant edema, in real-time. Our models can provide updated estimations as frequently as every hour, using data from structured time-varying patient records, radiographic text, and human-curated neurological characteristics. Our work resulted in two novel multi-class classification models, collectively named Hybrid Ensemble Learning Models for Edema Trajectory (HELMET), predicting the progression of midline shift over 8-hour (HELMET-8) and 24-hour windows (HELMET-24), respectively. HELMET combines transformer-based large language models with supervised ensemble learning, demonstrating the value of merging human expertise and multimodal health records in developing clinical risk scores. Both models were trained on a retrospective cohort of 15,696 observations from 623 patients hospitalized with large middle cerebral artery ischemic stroke and were externally validated using 3,713 observations from 60 patients at a separate hospital system. Our HELMET models are accurate and generalize effectively to diverse populations, achieving a cross-validated mean area under the receiver operating characteristic score of 96.6% in the derivation cohort and 92.5% in the external validation cohort. Moreover, our approach provides a framework for developing hybrid risk prediction models that integrate both human-extracted and algorithm-derived multi-modal inputs. Our work enables accurate estimation of complex, dynamic, and highly specific clinical targets, such as midline shift, in real-time, even when relevant structured information is limited in electronic health record databases., Competing Interests: Competing Interests All authors declare no financial or non-financial competing interests.

Background: Recent studies have shown that follow-up head CT is a strong predictor of functional outcomes in patients with middle cerebral artery stroke and mechanical thrombectomy. We sought to determine whether total and/or regional follow-up Alberta Stroke Program Early CT Score (ASPECTS fu ) are associated with important clinical outcomes during hospitalization and improve the performance of clinical prediction models of potentially lethal malignant edema (PLME)., Methods: We conducted a retrospective study of patients at three medical centers in a major North American metropolitan area with baseline and follow-up head CTs after large middle cerebral artery stroke between 2006 and 2022. We used multivariable logistic regression to test the association of total and regional ASPECTS fu with PLME (cerebral edema related death or surgery), adjusting for total baseline ASPECTS, age, sex, admission glucose, tissue plasminogen activator, and mechanical thrombectomy. We compared existing clinical risk models with and without total or regional ASPECTS fu using area under the curve., Results: In our 560 patient cohort, lower total ASPECTS fu was significantly associated with higher odds of PLME when adjusting for confounders (OR 1.69, 95% CI 1.49 to 2.0), and improved model discrimination compared with existing models and models using baseline ASPECTS. Deep territory involvement (OR 2.46, 95% CI 1.53 to 4.01) and anterior territory involvement (OR 3.23, 95% CI 1.88 to 5.71) were significantly associated with PLME., Conclusions: Lower ASPECTS fu and certain locations on regional ASPECTS fu , including deep and anterior areas, were significantly associated with PLME. Including ASPECTS fu information improved discrimination of established edema prediction models and could be used immediately to help facilitate clinical management decisions and prognostication., Competing Interests: Competing interests: None declared., (© Author(s) (or their employer(s)) 2023. No commercial re-use. See rights and permissions. Published by BMJ.)

Methods on rendering neurons in the central nervous system to be light responsive has led to a boom in using optical neuromodulation as a new approach for controlling brain states and understanding neural circuits. In addition to the developing versatility to "optogenetically" labeling of neural cells and their subtypes by microbiological methods, parallel efforts are under way to design and implement optoelectronic devices to achieve simultaneous optical neuromodulation and electrophysiological recording with high spatial and temporal resolution. Such new device-based technologies need to be developed for full exploitation of the promise of optogenetics. In this paper we present single- and multi-element optoelectronic devices developed in our laboratories. The single-unit element, namely the coaxial optrode, was utilized to characterize the neural responses in optogenetically modified rodent and primate models. Furthermore, the multi-element device, integrating the optrode with a 6×6 microelectrode array, was used to characterize the spatiotemporal spread of neural activity in response to single-site optical stimulation in freely moving rats. We suggest that the particular approaches we employed can lead to the emergence of methods where spatio-temporal optical modulation is integrated with real-time read out from neural populations.