Pitch a Problem

The 9th Annual Johns Hopkins Research Symposium on Engineering in Healthcare. Human + AI: Redefining the Standard of Care in Medicine. Pitch A Problem.

Table leaders will present major high-impact challenges from their clinical domain. These are challenges that, if solved, could meaningfully improve patient outcomes, streamline clinical workflows, advance care delivery, and more. Participants will learn about each challenge and brainstorm concrete directions for collaborations to tackle those challenges.

The goal of this activity is to spark new collaborations, support the development of joint grant proposals, and lay the groundwork for faculty groups who wish to continue the conversation beyond the symposium.

Table 1: AI for Liver Transplant Decision-Making

Facilitated by Aly T. Strauss, MD, PhD

Assistant Professor of Gastroenterology and Hepatology, Johns Hopkins School of Medicine

Co-Director of the Clinical and Translational Research Unit, Division of Gastroenterology and Hepatology

In the U.S., 4.5 million adults have liver disease, and liver transplantation (LT) is the only curative treatment for those with cirrhosis. LT centers are charged assessing patient appropriateness for transplant, culminating in a decision to list for transplant or decline—but LT listing decision-making is variable. We will discuss ways in which machine learning and AI hold potential to strengthen objectivity and minimize bias in complex decision-making, and the challenges involved in doing so in practice.

Table 2: AI for Aging and Cognitive Health

Facilitated by Esther Oh, MD, PhD

Professor of Geriatric Medicine and Gerontology, School of Medicine

Co-Director of the Johns Hopkins Memory and Alzheimer’s Treatment Center

With an aging population, how do we manage the rise in aging-related cognitive impairment and related conditions? We will discuss the clinical challenges of anticipating and managing cognitive decline, and the specific areas in which AI can play a role in helping improve outcomes for the elderly.

Table 3: AI and the Radiologist

Facilitated by Haris Sair, MD

Professor of Radiology and Radiological Science, School of Medicine

Director of the Division of Neuroradiology

The U.S. faces a large and deepening shortage of radiologists as increasing demand for imaging studies runs up against a constrained workforce. We will discuss the opportunities that AI presents to close this gap, as well as the challenges facing broad-based deployment of AI in radiology.

Table 4: AI at the Bedside: Making Algorithms Teammates (Not New Burdens)

Facilitated by Jim Fackler, MD

Director of Safety, Quality, and Logistics, Johns Hopkins Pediatric Intensive Care Unit

and Kimia Ghobadi, PhD

John C. Malone Assistant Professor of Civil and Systems Engineering, Whiting School of Engineering

In a critical care setting, we believe AI-based algorithms have neither improved patient outcomes nor improved clinical effectiveness. Further, we believe AI will only improve care if it is used to improve workflows. This can only happen if the design of AI-based critical care systems is created cognizant of the current, and then improved, sociotechnical environment. We will explore two use cases: 1) mechanical ventilation management and 2) sepsis management.

Table 5: Reimagining Nursing Education in the AI Era

Facilitated by Kristen Brown, DNP, RN, CRNP, CPNP-AC, CHSE-A, FSSH, FAAN

Associate Professor, Johns Hopkins School of Nursing

Nurses play an increasingly prominent role in the U.S. healthcare system, and—with more than 4 million nurses in the U.S.—make up the largest share of the U.S. healthcare workforce. We will discuss challenges and opportunities for modernizing and personalizing nursing education to ensure a highly competent nursing workforce, including the potential of AI, competency-based education, and simulation.

Table 6: Using AI to Assist Digital Health Interventions

Facilitated by Liz Selvin, PhD, MPH

Professor of Epidemiology, Johns Hopkins Bloomberg School of Public Health

Director of the Welch Center for Prevention, Epidemiology, and Clinical Research

and Caitlin Hicks, MD, MS

Associate Professor of Surgery, School of Medicine

Vice Chair of Research in the Department of Surgery

We will discuss best practices for using AI to assist with digital health interventions in research and clinical practice. We will focus on issues related to engagement, health outcomes, and barriers to real-world deployment.

Table 7: AI for Decision Support in Infectious Diseases

Facilitated by Matthew Robinson, MD

Assistant Professor in the Division of Infectious Diseases, School of Medicine

Treating infectious diseases involves complex clinical decision-making and information-gathering, such as making judicious use of antibiotics to reduce the emergence of antibiotic-resistance bacteria, treating complex conditions like meningoencephalitis, and carefully sifting through complex patients’ medical records. We will discuss ways in which AI has made progress—or fallen short—in helping clinicians treat infectious diseases.

Table 8: Using AI to Measure and Predict Mental Health Conditions

Facilitated by Roy Adams, MS, PhD

Assistant Professor of Psychiatry and Computer Science, School of Medicine and Whiting School of Engineering

and Emily E. Haroz, PhD, MHS, MA

Associate Professor, Bloomberg School of Public Health

Co-Director of Research at the Center for Indigenous Health

Deputy Director of the Center for Suicide Prevention

Despite broad growth in diagnosis rates, mental health conditions remain among the least served by ML/AI-based decision support tools. With a focus on suicide prevention, we will discuss ways in which AI might be used to measure and predict these conditions, and how these tools might be integrated into care.