Publications by authors named "Isaac Sears"
Background: Venovenous extracorporeal membrane oxygenation (VV-ECMO) is associated with acute brain injury (ABI), including central nervous system (CNS) ischemia (defined as ischemic stroke or hypoxic-ischemic brain injury [HIBI]) and intracranial hemorrhage (ICH). Data on prediction models for neurologic outcomes in VV-ECMO are limited.
Methods: We analyzed adult (age ≥18 years) VV-ECMO patients in the Extracorporeal Life Support Organization (ELSO) Registry (2009-2021) from 676 centers.
Article Synopsis
- * We analyzed data from 35,855 adults who received ECMO between 2009 and 2021, finding that 7.7% experienced acute brain injuries. Various machine learning algorithms were used to evaluate predictive accuracy, with area under the curve values indicating moderate predictive capability.
- * Key factors linked to an increased risk of brain injury included longer ECMO duration, higher pump flow rates, and elevated oxygen levels during treatment, emphasizing the need for careful monitoring and management of these
Article Synopsis
- Researchers investigated how machine learning algorithms can predict complications after colectomy surgery for colonic neoplasia, aiming to enhance accuracy compared to traditional statistical methods.
- The study used data from the National Inpatient Sample database (2003-2017) and included 14,935 adult patients who underwent elective colectomy, analyzing outcomes such as anastomotic leaks and inpatient mortality.
- Among the machine learning models tested—decision tree, random forest, and artificial neural networks—the neural network performed best in predicting complications, achieving high accuracy with area under the receiver operating characteristic curve scores ranging from 0.84 to 0.93 for various outcomes.
Article Synopsis
- The study aimed to assess if machine learning can predict acute brain injury (ABI) in patients on venoarterial extracorporeal membrane oxygenation (VA-ECMO) and identify risk factors for such injuries.
- Using data from a large international registry (2009-2021) involving 35,855 VA-ECMO patients, it was found that 7.7% experienced ABI, with certain features like longer ECMO duration and higher pump flow linked to increased risk.
- For patients undergoing extracorporeal cardiopulmonary resuscitation (ECPR), 16.5% experienced ABI, and predictive accuracy for ABI was better than in VA-ECMO alone, with findings reinforcing the
Article Synopsis
- The study investigates the use of machine learning (ML) to predict acute brain injury (ABI) in patients undergoing venovenous extracorporeal membrane oxygenation (VV-ECMO), focusing on factors like CNS ischemia and intracranial hemorrhage (ICH).
- Data was analyzed from 37,473 VV-ECMO patients, revealing that 7.1% experienced ABI, with machine learning algorithms providing moderate predictive accuracy (around 67-70% area under the curve) for different types of ABI.
- Findings identified pre-ECMO cardiac arrest as the major risk factor for ABI, while longer ECMO duration and transplantation intentions were linked to a lower ABI risk, highlighting
In response to the 2023 George B. Moody PhysioNet Challenge, we propose an automated, unsupervised pre-training approach to boost the performance of models that predict neurologic outcomes after cardiac arrest. Our team, (BrownBAI), developed a model architecture consisting of three parts: a pre-processor to convert raw electroencephalograms (EEGs) into two-dimensional spectrograms, a three-layer convolutional neural network (CNN) encoder for unsupervised pre-training, and a time series transformer (TST) model.
View Article and Find Full Text PDFBackground A critical decrease in the number of healthcare providers in developing countries is one of the major burdens to healthcare access in these countries. Many factors contribute to the lack of healthcare providers, including low doctor-to-population ratio, emigration of doctors to other countries, long travel distances to hospitals, increasing cost of healthcare, and concentration of doctors in urban cities. Several measures have been taken by both governmental and nongovernmental organizations in these countries to mitigate this crisis with varying outcomes.
View Article and Find Full Text PDF