Resident Workflow Reimagined: A Generative-AI–Driven Chatbot to Improve Clinical Operations and Trainee Efficiency

Presentation Type: Oral Presentation - Regular
Presentation Time: 02:00 PM - 02:12 PM

Abstract Keywords: Generative AI in Clinical Workflow: Ambient Listening, Chart Summarization, Automated Response with LLM, Clinician Well-Being, Clinical Decision Support and Care Pathways, Human Factors and Usability, Education and Training, Outcomes Improvement and Equity
Primary Track: Advancing Wellness for Providers and Community with Consideration of Human Factors


Residency workflows often rely on unstructured documents and informal communication, creating inefficiencies and inconsistent practices. We describe the development of a retrieval-augmented generative-AI chatbot designed to deliver real-time, document-linked guidance from a curated internal medicine knowledge base. This presentation outlines system architecture, document engineering, pre-pilot testing, and lessons learned, highlighting how AI tools can surface and standardize the hidden operational curriculum.

Speaker(s):
Daniel Yang, MD
UCI Irvine Medical Center

Author(s):
Daniel Yang, MD - UCI Irvine Medical Center; Vishnu Bharani, MD - University of California - Irvine; Yawen Guo, MISM - University of California - Irvine; Thuy Nguyen, Undergraduate Student - University of California - Irvine; Jiacheng Zhuo, Master of Software Engineering - University of California - Irvine; Monil Patel, MD - UC Irvine Health;

---

From Flood to Focus: Inpatient Insights AI Summarization

Presentation Type: Oral Presentation - Regular
Presentation Time: 02:12 PM - 02:24 PM

Abstract Keywords: Generative AI in Clinical Workflow: Ambient Listening, Chart Summarization, Automated Response with LLM, Change Management, Clinician Well-Being, Human Factors and Usability
Primary Track: Implementing Real-World Change, Digital Engagement, and Connected Health


Clinicians face an overwhelming volume of inpatient data dispersed across notes, labs, and documentation. Inova Health System implemented Inpatient Insights, an Epic-integrated generative AI tool that synthesizes clinically relevant summaries to reduce cognitive burden and accelerate chart review. A phased validation approach involving 22 inpatient clinicians demonstrated improved NASA-TLX cognitive workload scores, higher perceived efficiency, and strong user satisfaction (NPS +58). Ongoing randomized stepped-wedge expansion to 222 providers will quantify impact and inform broader AI governance and adoption strategies.

Speaker(s):
Sameh Saleh, MD, MBMI
Inova Health System

Author(s):
Markos Kashiouris, MD, MPH - Inova;

---

From Summarization to Custom Prompts: Real-World Implementation and Evolution of an EHR-Native Generative AI Assistant

Presentation Type: Oral Presentation - Regular
Presentation Time: 02:24 PM - 02:36 PM

Abstract Keywords: Generative AI in Clinical Workflow: Ambient Listening, Chart Summarization, Automated Response with LLM, Clinician Well-Being, Human Factors and Usability, Workforce Automation, Communication, and Workflow Efficiency
Primary Track: Implementing Real-World Change, Digital Engagement, and Connected Health


We evaluated an EHR-native generative AI assistant among 73 clinicians to address documentation burden and explore real-world use. The tool achieved a 75% acceptance rate, with 56% of users reporting reduced burden. The addition of custom prompts shifted usage from clinical summarization to more complex cognitive tasks, such as creating handoffs and extracting billing/medical complexity. These findings demonstrate that flexible, clinician-driven implementation is essential for maximizing the utility of AI in clinical workflows.

Speaker(s):
Peter Zhang, MD, MS
Children's Hospital of Philadelphia / University of Pennsylvania

Author(s):
Loukya Kanakamedala, DO - CHOP; Kai Xu, MD - University of Pennsylvania;

---

Results from a Pilot Implementation of a Generative AI Tool for Drafting Hospital Courses at a Large Academic Children’s Hospital

Presentation Type: Oral Presentation - Regular
Presentation Time: 02:36 PM - 02:48 PM

Abstract Keywords: Generative AI in Clinical Workflow: Ambient Listening, Chart Summarization, Automated Response with LLM, Change Management, Clinician Well-Being, Leadership and Strategy
Primary Track: Implementing Real-World Change, Digital Engagement, and Connected Health


Our institution, a large quaternary academic children’s hospital, piloted a Draft Hospital Course Summary Tool available within our electronic health record (EHR). Clinicians reported increased efficiency and decreased cognitive burden, with 96.8% reporting that the tool provided a helpful starting point for writing their hospital courses. Our findings offer early evidence that such a tool can support discharge documentation workflows in a pediatric academic medical center.

Speaker(s):
Alex Ruan, MD
Children's Hospital of Philadelphia

Author(s):
Osvaldo Mercado, MD - Children's Hospital of Philadelphia; Eric Shelov, MD MBI - The Children's Hospital of Philadelphia/ Perelman School of Medicine, Univ. of Pennsylvania; Maria Ricci, BS - Children's Hospital of Philadelphia; Stephon Proctor, PhD, MBI, ABPP - Children's Hospital of Philadelphia;

---

Real-World Experience from a Pediatric Hospital Using EHR-Integrated AI Summarization Tools

Presentation Type: Oral Presentation - Regular
Presentation Time: 02:48 PM - 03:00 PM

Abstract Keywords: Generative AI in Clinical Workflow: Ambient Listening, Chart Summarization, Automated Response with LLM, Workforce Automation, Communication, and Workflow Efficiency, Analytics, Registries, and the Digital Command Center
Primary Track: Advancing Wellness for Providers and Community with Consideration of Human Factors


This study examines real-world usage an electronic health record-integrated artificial intelligence text assistant. Rapid adoption and acceptance from clinicians suggest great interest, especially for features that support formal, accurate clinical notes. Our results underscore the potential of artificial intelligence for future healthcare implementations.

Speaker(s):
Jonathan Hron, MD
Boston Children's Hospital

Author(s):
Jonathan Hron, MD - Boston Children's Hospital; Danielle Perley, MSN, RN, NI-BC - Boston Children's Hospital; Daniel Kats, MD - Boston Children's Hospital; Alon Geva, MD, MPH - Boston Children's Hospital; Chase Parsons, DO, MBI - Boston Children's Hospital;

---

Implementing Chart Hero: A SMART-on-FHIR Integrated Assistant to Support Clinicians

Presentation Type: Oral Presentation - Regular
Presentation Time: 03:00 PM - 03:12 PM

Abstract Keywords: Innovation Partnerships, Implementation Science, and Learning Health Systems, Leadership and Strategy, Generative AI in Clinical Workflow: Ambient Listening, Chart Summarization, Automated Response with LLM
Primary Track: Implementing Real-World Change, Digital Engagement, and Connected Health


We developed and implemented Chart Hero, a SMART-on-FHIR integrated AI assistant that provides context-aware answers to clinician questions using structured and unstructured EHR data. We outline the system’s design, governance, and iterative deployment, and report results from a 3.5-month pilot with 20 clinicians. We describe real-world query patterns, usability outcomes, and key challenges for safely deploying AI assistants within clinical workflows.

Speaker(s):
Yevgeniy Gitelman, MD


Author(s):
Yevgeniy Gitelman, MD; Sabrina Meng, BS - Perelman School of Medicine - University of Pennsylvania; Emeka Anyanwu, MD - Penn Medicine; Damien Leri, MSHP - Penn Medicine; Laura Kavanaugh, MSCI - Penn Medicine; Kevin Xie, PhD - Penn Medicine; Srinath Adusumali, MD, MSHP - Penn Medicine; Kathleen Lee, MD - Penn Medicine; Raina Merchant, MD - Penn Medicine; Mitchell Schnall, MD, PhD - Penn Medicine; Srinivas Sridhara, PhD - Penn Medicine; Anna Schoenbaum, DNP - Penn Medicine; Lauren Hahn, MBA - Penn Medicine;

---