The Said lab is dedicated to advancing the field of biomedical informatics with a specific focus on pediatric critical care. Our lab is at the forefront of leveraging electronic health records (EHRs) and high-resolution data from intensive care units (ICUs) to develop predictive models that inform clinical decision-making, particularly in the pediatric ICU setting.
Our primary mission is to create clinically applicable decision support tools that aid healthcare professionals in determining the optimal timing and utilization of high-risk, resource-intensive therapies. These tools are designed to integrate seamlessly into existing clinical workflows, providing real-time insights that enhance the quality of care for critically ill pediatric patients. By focusing on interventions that improve both short-term and long-term neurological outcomes, our lab addresses some of the most pressing challenges in pediatric critical care.
Our research is interdisciplinary, drawing on expertise from clinical medicine, computer science, biostatistics, and engineering. We collaborate closely with pediatric intensivists, neurologists, biostatisticians, computing engineers, and other specialists to ensure that our models are grounded in clinical reality and are tailored to the unique needs of pediatric patients. Through this collaboration, we aim to translate complex data into actionable information, supporting clinicians in making informed, timely decisions that can significantly impact patient outcomes.
In addition to developing predictive models, our lab is committed to advancing the field of translational informatics. We work to bridge the gap between research and clinical practice by ensuring that our tools are not only scientifically rigorous but also user-friendly and practical for daily clinical use. Our goal is to accelerate the adoption of these innovations in pediatric ICUs, ultimately leading better long-term outcomes for children.
Through our innovative research and collaborative approach, the Said lab is poised to make a lasting impact on the field of pediatric critical care, improving the lives of children and their families.
Current projects
To develop a comprehensive database integrating high-resolution EHR and telemetry data from pediatric critical care settings to enable advanced predictive modeling. This project is at the forefront of translational biomedical informatics by creating a robust data infrastructure that supports the development of predictive models aimed at improving patient outcomes in pediatric critical care. By leveraging multisource data, it enhances the accuracy and applicability of predictive analytics, setting a new standard for data-driven clinical decision support in critical care.
To utilize high-resolution ICU data and EHR systems to create predictive models that inform decision support tools aimed at improving neurological outcomes in pediatric ECMO patients. This project bridges translational informatics and clinical care, offering the potential to significantly enhance the long-term neurological health of the most critically ill children. The integration of advanced data analytics with clinical decision-making sets a precedent for future innovations in pediatric critical care.
To analyze current pediatric anticoagulation practices to identify correlations with patient outcomes, with a focus on optimizing treatment strategies. This project aims to uncover key insights into how varying anticoagulation practices affect pediatric patient outcomes, providing evidence to guide more effective and individualized treatment strategies. The findings have the potential to significantly improve safety and efficacy in pediatric anticoagulation therapy.
To develop pilot machine learning models to predict outcomes based on pediatric ECMO patient characteristics, anticoagulation initiation, titration strategies and laboratory monitoring values. By optimizing anticoagulation management, this project aims to enhance patient outcomes and resource utilization in critical care settings. The interdisciplinary approach underscores its potential to influence precision health interventions, to improve patient outcomes.
To address the challenges of integrating ECLS data within the OMOP CDM framework to enhance data usability and interoperability for research and clinical purposes. This project is at the forefront of advancing data standardization and interoperability in critical care, with implications for improving data-driven research and patient outcomes in ECMO therapy.
To develop and implement bedside neuromonitoring technology using high-density optical tomography to provide functional MRI-equivalent data for pediatric ECMO patients. This collaborative project with advanced neuroimaging researchers pioneers the application of advanced optical imaging techniques in critical care, offering real-time insights into cerebral oxygenation and function at the bedside. The ability to monitor neurological status with MRI-like precision in the most critically ill pediatric patients, has the potential to revolutionize the management of pediatric ECMO patients, improving both immediate and long-term neurological outcomes.
Collaborations
Our lab is deeply engaged in a dynamic collaboration across the WashU campus.
Brain Light Laboratory
Washington University Mallickrodt Institute of Radiology
This close collaboration brings together the cutting-edge neuro-imaging technology to the most critically ill pediatric patients by providing the ability to continuously monitoring children supported on ECMO at the bedside with imaging modalities that provide functional MRI comparable data.
Cyber-Physical Systems Laboratory
Washington University McKelvey School of Engineering
This partnership leverages the complementary strengths of both labs to advance the integration of cyber-physical systems with biomedical informatics, driving innovation in predictive modeling and decision support systems for critical care. Together, we aim to push the boundaries of what’s possible in healthcare technology, enhancing our ability to deliver impactful, data-driven solutions.