Research Area 1 - Integrating eXtended Reality and Biosensors for Non-Pharmacological Therapies and Digital Twins in Healthcare
My research leverages Extended Reality (XR) technologies in combination with biosensors to develop innovative solutions for both non-pharmacological therapies and healthcare digital twins. This dual approach addresses physical and psychological health challenges while enhancing patient care systems through predictive modeling and real-time data integration.
In therapeutic contexts, I utilize XR and biosensors to manage pain, anxiety, and addiction recovery. By incorporating biofeedback, such as brain and cardiac activity, into immersive XR environments, I create adaptive interventions tailored to individual physiological responses. For instance, VR alongside biosensors has been used during ultrasound-guided breast biopsies to alleviate patient anxiety, offering a personalized, real-time intervention that complements traditional pain management strategies. My work also tackles the opioid crisis by incorporating Art Therapy and Cognitive Behavioral Therapy (CBT) into XR tools, providing accessible therapy for patients recovering from Opioid Use Disorder (OUD).
Beyond therapeutic applications, my research explores using VR-based digital twins in healthcare. These digital twins create virtual replicas of healthcare environments, systems, and processes, integrating real-time data from biosensors to simulate and optimize patient outcomes. For example, I have developed VR simulations of surgical environments where digital twins provide insights into patient stress and recovery, informing strategies to improve short-term and long-term care. This research expands the scope of digital twins to include operative planning, post-surgery recovery, and predictive modeling for personalized healthcare.
​
Research Area 2 - Adaptive Digital Twins in Healthcare: Integrating Agent-Based Simulations and Reinforcement Learning for Decision-Making
My second research area focuses on developing adaptive digital twins in healthcare by integrating agent-based modeling (ABM) with reinforcement learning (RL) for optimized decision-making and personalized care. By using ABM, I model complex healthcare environments—such as emergency departments or surgical settings—where individual agents (patients, healthcare professionals, resources) interact in dynamic and unpredictable ways. This approach allows us to simulate patient pathways, resource allocation, and system capacity, uncovering bottlenecks and inefficiencies. The addition of reinforcement learning enables these digital twins to continuously adapt, learning from real-time data to optimize processes such as reducing patient onboarding times, improving resource utilization, and enhancing treatment outcomes.
In personalized medicine, I use digital twins to simulate individualized treatment plans for chronic conditions, such as opioid addiction or diabetes. By incorporating real-time biosensor data and applying RL, the digital twin can adjust treatment protocols in response to patient-specific changes, improving both outcomes and efficiency. During the COVID-19 pandemic, these techniques were applied to optimize staffing policies, helping reduce burnout and infection risk among healthcare workers.
Beyond operational challenges, my research expands into areas like surgical planning and preventive care, where digital twins can simulate different scenarios to enhance patient safety and provide personalized, evolving lifestyle interventions. Through the integration of mathematical modeling and machine learning, my work aims to create intelligent, adaptive healthcare systems that drive data-driven, personalized decision-making across various healthcare domains.​​
​​​​​​
Selected Completed Projects
​
The Impact of Virtual Reality on Pain and Anxiety during Ultrasound Guided Breast Biopsies
Over 1.6 million patients undergo an image-guided biopsy to obtain a breast tissue diagnosis in the United States annually. Prior studies report that women anticipate these biopsies to be painful. Local anesthesia has been the mainstay for pain management during image-guided biopsies for decades, and most patients report significantly lower pain levels with local anesthesia. However, despite local analgesia, a common accompaniment of pain during these procedures is anxiety. Studies investigating the factors contributing to patient anxiety relating to a percutaneous breast biopsy report a multitude of factors, including patient concerns over biopsy results, level of education, age of the patient, communication with radiologists, information available, and the number of relatives with breast cancer. The purpose of this study was to investigate the efficacy of immersive VR in combination with conventional pharmacological treatment for mitigating pain and anxiety experienced during ultrasound-guided breast biopsies compared to conventional pharmacological treatment alone in the outpatient setting.
​
Extended Reality-Based Art Therapy and Cognitive Behavioral Therapy for Opioid Use Disorder and Recovery
Opioid abuse and related overdoses, characterized as the "US opioid epidemic," are one of the nation's leading causes of injury deaths. In December 2021, opioid overdose deaths reached a record high in the US, with over 290 daily deaths. While opioid misuse has significantly increased over the years, digital health and technology-based tools for managing opioid cravings, recovery, and relapse have not kept pace. This calls for an urgent need to design and develop adjunct programs and tools beyond conventional therapies that are accessible, affordable, and convenient for those recovering from Opioid Use Disorder (OUD). This research intends to address the critical public health challenge of the opioid epidemic by integrating Art Therapy and Cognitive Behavioral Therapy with Extended Reality (XR) technologies to develop a digital tool that is convenient, affordable, and accessible to OUD patients. This research will be carried out by a diverse faculty and practitioners with expertise in immersive technologies, human-computer interaction, psychotherapy, behavioral health, and addiction services.
​​
Biofeedback Enhanced Adaptive Virtual Reality to Mitigate Surgical Pain and Anxiety
Pain and anxiety are common accompaniments of surgery, and opioids have been the mainstay of pain management for decades. Approximately 80% of the patients undergoing surgery leave the hospital with an opioid prescription. Moreover, patients receiving an opioid prescription after short-stay surgeries have a 44% increased risk of long-term opioid use, and about one in 16 surgical patients becomes a long-term user. We aimed to develop a virtual reality experience based on Attention Restoration Theory and integrate the user's heart rate variability biofeedback to create an adaptive environment to mitigate preoperative anxiety and postoperative pain.
​
Virtual Reality to Mitigate Pain and Anxiety in an Operative Setting
Research demonstrates that ten percent of the population becomes addicted to opioids from exposure to narcotics in the operative setting. The abuse and addiction to these drugs have placed the US at the center of an “opioid epidemic.” A variety of programs and interventions are being explored to treat the pain associated with surgery while minimizing or eliminating the need for opioids. Our goal was to assess the efficacy of virtual reality in mitigating surgical pain and anxiety among patients undergoing total knee arthroplasty.
​
Virtual Reality to Mitigate Anxiety among AYA Cancer Patients
Studies have reported that adolescent and young adult (AYA) patients are specifically vulnerable to distress because of the intersection of disease and age. Compared to older cancer patients, AYA cancer patients indicate a more negative psychosocial outcome. Recent studies have focused on improving the quality of life of AYA patients by providing adjunctive non-pharmacological interventions. Our longitudinal study analyzed the efficacy of VR in the AYA population and sought to understand their preferences in virtual environments.
​
Stress and Burnout among Attending and Resident Physicians in the Emergency Department
Emergency department physicians work in complex care settings and are frequently exposed to stressful conditions. Burnout among physicians is increasing each year, and one of the most prone groups is emergency medicine providers. This research sought to understand if a difference in stress levels and burnout rate exists between attending and resident physicians working in an academic Level 1 trauma center emergency department on the same shift. Stress levels were estimated using physiological measures, including heart rate variability and electrodermal activity. Burnout scores and workload index were collected using the Maslach Burnout Inventory-Human Services Survey and the NASA-TLX survey.
​
A Data-Driven Modeling Approach to Allocating Resources in Emergency Departments
Over 145 million people visit US Emergency Departments (ED) annually. The diverse nature and overwhelming volume of patient visits make the ED one of the most complicated settings in healthcare to study. ED overcrowding is a recognized worldwide public health problem, and its negative impacts include an increased patient length of stay, medical errors, patients left without being seen, ambulance diversions, and health system finances. The critical research task is to understand how to identify “optimized” policies that reduce onboarding time, reduce the length of stay, and increase throughput with minimum investment. The research team will identify and collect patient-level and system-level data to support data-driven analytical modeling of GMH ED operations. With these data, the team will build a simulation/optimization model of the ED representing patient care pathways, system capacities, and operational policies.
​
Team-Based, Risk-Adjusted Staffing during a Pandemic: An Agent-Based Approach
Since the World Health Organization declared the novel coronavirus disease a pandemic, more than 30 million cases of infections and 950,000 deaths have been reported across the world. Specialty physicians are now working as frontline workers due to hospital overcrowding and a lack of providers, and this places them as a high-risk target of the epidemic. Within these specialties, anesthesiologists are one of the most vulnerable groups as they come in close contact with the patient's airway. An agent-based simulation model was developed to test various staffing policies, and we demonstrate the benefits of a restricted, no-mixing shift policy, which segregates the anesthesiologists as groups and assigns them to a shift within a single hospital.