Ambulatory Setting

This research develops and evaluates an artificial intelligence-enhanced pretreatment peer-review process in radiation oncology, aiming to improve patient safety by reducing variability among providers in treatment planning, minimizing clinical errors, and enhancing overall treatment outcomes.

This research aims to improve the early detection of venous thromboembolism in primary and urgent care by using mixed methods (stakeholder interviews and surveys, electronic health records, and machine learning) to better understand missed and delayed diagnoses, identify risk factors, and develop tools to enhance patient safety.

This research evaluates the adoption and impact of three electronic health record-optimization interventions—scribes, advanced team-based inbox management, and artificial intelligence-assisted messaging support—on primary care physicians' time, wellbeing, and patient outcomes, with the goal of identifying effective strategies to improve physician satisfaction and care quality and to reduce healthcare costs.

The study will develop and implement a validated machine learning model to optimize voiding cystourethrogram timing and use for diagnosing vesicoureteral reflux (VUR) in children, aiming to reduce the significant health and economic impacts of VUR and recurrent febrile urinary tract infections (fUTIs) by standardizing practices, minimizing unnecessary procedures, and ensuring timely diagnosis for those at highest risk, ultimately seeking to prevent renal injury from fUTIs.

This study will develop and establish the efficacy of an actionable predictive model to identify pregnant individuals at high risk for postpartum hemorrhage which can be used in combination with a clinical decision support tool to reduce the risk of hemorrhage-related morbidity and improve maternal health outcomes.

The study will create and implement digital tools using the SMART on FHIR framework to support Age-Friendly care in clinical practice and institutional decision-making.

This research will develop and validate a clinician-facing longitudinal risk-prediction tool using self-reported data from US military service members and veterans, to assist clinicians in screening, evaluation, and timely intervention in suicide ideation and suicide attempts.

This research will implement, evaluate, and disseminate DDInteract, a shared decision making clinical decision support tool that aims to reduce harm from oral anticoagulant drug-drug interactions.

This research will enhance, refine, and evaluate an artificial intelligence-supported tool that collects patient-reported outcomes to inform care and improve glycemic control for patients with poorly controlled diabetes.

This research will develop, validate, and evaluate a decision support tool using a machine learning algorithm to standardize the approach to discontinuing disease-modifying therapies for multiple sclerosis.