DIGI Grant Winners

This project aims to identify previously unknown pulmonary emphysema in non-symptomatic patients using a combination of imaging and artificial intelligence. The goal of the project is to enable earlier treatment with improved prognosis and outcome.

Remote Patient Monitoring (RPM) is an innovative strategy designed to give patients and caregivers better access to clinical data about patients between visits to the clinic or hospital. This program seeks to optimize access to this technology for our UMass Memorial patients and streamline the process of onboarding and managing the flow of data. By integrating a multidisciplinary care team, including Mobile Integrated Health (MIH) Emergency Physicians and Primary Care Physicians, the hope is to make this resource more available and easier to access.

Limited resources on nights and weekends can significantly alter patient flow and length of stay for our Emergency Department patients and hospital inpatients. The objective of this project is to use sophisticated optimization algorithms to allocate limited resources on nights and weekends to minimize length of stay. This project is being conducted in coordination with our research colleagues within MIT's Operations Research Center.

Osteoporosis and its clinical outcome of fragility fracture is an ever-increasing public health problem due to our aging population. Despite a number of FDA-approved therapies for the treatment of osteoporosis, a significant number of individuals who would qualify for these treatments are not yet being prescribed them. The aim of this project is to use existing data within the electronic health record to identify individuals with osteoporosis who do not appear to have been treated and link them into treatment for fracture-risk reduction.

Despite continuous glucose monitoring (CGM) achieving a high level of accuracy and durability for glucose measurement in the outpatient setting, its use within the Intensive Care Unit (ICU), where dysglycemia is associated with increased morbidity and mortality, is not well validated. The current point-of-care glucose and plasma glucose testing have been the gold standard in the hospital setting, but these methods have significant limitations in that they capture only intermittent single-point-in-time data. CGM technology offers significant advantages in capturing blood glucose trends over time that could allow for earlier and timelier therapeutic interventions, especially with regard to hypoglycemia. This project aims to study the validity and efficacy of Continuous Glucose Monitoring in the ICU.

Given the ongoing high demand for inpatient services, tremendous efforts have been underway to improve patient flow across the care continuum. With the support of the DIGI Grant this project will use digital tools, such as existing EPIC infrastructure and Tableau data, to improve and track discharge processes. This project’s primary goal is to improve discharge orders before noon to >35% across the Hospital Medicine Division and Department of Medicine, more generally; and a secondary goal is to improve overall discharge from the hospital by 2 pm.

Sickle cell disease (SCD) is an inherited, complex, lifelong blood disorder with varying genotypes and a spectrum of phenotypes. Uncontrolled pain secondary to vaso-occlusion is the most common reason for hospital admission for patients with SCD. However, the complex nature of sickle cell-related pain complicates the care of this vulnerable and underserved patient population. While opioids are standard pharmacologic treatments for vaso-occlusive crises, their use may be limited by significant side effects. Growing awareness of the multidimensional aspects of pain has led to the increasing use of adjunctive non-pharmacologic interventions to address the physical, psychological, social and environmental factors that contribute to the overall experience of pain. This project will examine the usability and utility of immersive virtual reality as an adjunctive modality in managing sickle cell disease-related pain for pediatric patients at UMass Memorial Medical Center.

The Transitional Care Team is responsible for accepting and tracking patients for admission or transfer to the Department of Medicine and Medicine subspecialties. This project will leverage the current capabilities of EPIC to overhaul the patient tracking and triaging system, improving patient safety and reducing the workload of members on the Transitional Care Team.

Reduced availability in weekend clinical services (blood bank, endoscopy, cardiac interventions, interpreters, IV/PICC placement, laboratory medicine, medical and surgical consultations, pathology, pharmacy, physical therapy, radiology, speech and swallow, social work, and transportation) results in unnecessary bed days and discharge delays. Deciding how best to expand upon these services in a way that will maximally improve probability of weekend discharge while minimizing additional expense is challenging with traditional methods of analysis. Yet, creation of an optimization model utilizing the robust, clinical data available in our EMR holds the potential to efficiently arrive at such a solution. Such methods have been employed in retail, energy and transportation sectors. Translating such work to UMass Memorial Medical Center holds not only significant academic promise worthy of publication but also hopes to increase equitable access to care for Central Massachusetts.

This program seeks to identify patients with cardiometabolic disorders who will benefit from advanced therapeutic interventions using SGLT2 inhibitors. Patients with chronic kidney disease, heart failure and diabetes are enrolled and followed with the goal of reducing morbidity and mortality.

Emergency departments are experiencing a severe demand/capacity mismatch, which leads to increased risk. The post-COVID era has significantly impacted the workforce leading to a significant increase in turnover of nursing staff, decreasing institutional knowledge and experience. The nurse performing triage only has a few minutes with each patient, therefore this project aims to implement a tool to support staff and ensure accurate triaging of our patients. KATE AI uses machine learning and Clinical Natural Language Processing to improve acuity assignments, patient safety, and reduce bias, errors and risk.

The effect of low-density lipoprotein (LDL) on atherosclerotic cardiovascular disease (ASCVD) risk is both causal and cumulative over time, thus lowering plasma LDL is critical to reducing ASCVD. Only one-third of patients at high risk for ASCVD have LDL at levels recommended by the American Heart Association and American College of Cardiology. This project aims to reduce ASCVD risk in this vulnerable population by improving access to advanced lipid-lowering therapeutics.