Penn Medicine

by Penn Medicine  · Based in United States → — Built on firsts. Driven by innovation.
Dermatology Emergency Medicine Neurology

Regulatory Status Disclosed

Overview

Penn Medicine has implemented several AI tools to enhance clinical workflows and patient care. One such tool is Chart Hero, a generative AI-powered, chat-based system designed to synthesize patient data from electronic health records (EHRs). This tool allows clinicians to quickly gather, organize, and interpret pertinent patient information, appearing as a sidebar within the EHR. Chart Hero is intended for use by clinicians across various specialties and care settings to streamline the pre-visit patient review process, potentially reducing the time spent sifting through extensive medical histories.

Another initiative involves integrating K Health’s AI-powered patient intake agents, initially in virtual urgent care (Penn Medicine OnDemand), with plans to expand into in-person primary care and specialties like cardiology and dermatology. These AI agents conduct an intake process by asking patients about symptoms and reasons for seeking care, then generate a draft clinical chart for the provider’s EHR before the appointment. This aims to reduce patient wait times, streamline clinical workflows, and support clinicians with visit preparation and documentation.

Penn Medicine also utilizes AI for other purposes, including an AI-enabled system to assist with replies to patient messages, where AI drafts are reviewed by clinicians before sending. Additionally, AI scribe technology is in use by over 1,200 providers for automated note-taking during patient visits. The institution is also exploring AI for medical education through the CRISP (Clinical Reasoning Insights for Shaping Performance) initiative, which uses AI to analyze clinical reasoning skills in medical students and trainees across specialties such as Internal Medicine, Emergency Medicine, Surgery, and Radiology.

Reviewed by Pouyan Golshani, MD — Interventional Radiologist

Key Features

  • Data sifting
  • Data synthesis
  • Patient data summarization
  • Key information highlighting
  • Supports diagnostic accuracy
  • Aids treatment planning

Use Cases

  • Quickly review patient history
  • Support clinical decision-making
  • Reduce physician data overload
  • Enhance efficiency in patient data review
  • Assist in emergency department settings
  • Facilitate consultations

What Physicians Need to Know

Evidence Base (guidelines, literature scope)
The AI tool, Chart Hero, is powered by a large language model that searches available data sources across a patient's chart, including images, scanned documents, notes, and summaries. It contextualizes this data with evidence from medical literature. Penn Medicine also collaborates on peer-reviewed research to expand the global evidence base for routine clinical AI automation.
Clinical Validation Studies
Chart Hero is currently in a testing phase with over 50 users across various inpatient and outpatient settings, including Radiology, Nuclear Medicine, Infectious Disease, Cardiology, and Emergency Medicine. In its first three months, it produced over 1,000 outputs and received largely positive feedback (68% positive) with a Net Promoter Score of 54. Penn Medicine plans to roll out AI tools in the next two years using validation studies and small test runs. The collaboration with K Health also includes co-development of peer-reviewed research examining the use of clinical AI in routine care settings.
Alert Fatigue Management
While the provided information does not specifically detail alert fatigue management for Chart Hero, Penn Medicine's broader AI initiatives, such as the AI scribe, aim to reduce documentation burden and allow clinicians to focus more on patient interaction, which can indirectly mitigate factors contributing to alert fatigue. Generally, AI systems are seen as a solution for delivering highly personalized and clinically relevant alerts to reduce alert fatigue.
Drug Interaction Checking
The AI tool is designed to answer a wide range of questions, including inquiries about whether specific medications can cause particular side effects. Penn researchers are also using AI and social media to assess drug side effects, analyzing large-scale, real-world data.
Differential Diagnosis Support
The AI tool assists clinicians in sifting through large volumes of patient data to quickly surface relevant information and contextualize it with medical literature. It can also suggest potential courses of action for patients, which doctors can then review. Penn Medicine is also developing AI that can give feedback on clinical reasoning, which is crucial for diagnostic processes.
Guideline Update Frequency
The AI field is rapidly evolving, and Penn's guidelines for AI use are anticipated to be updated regularly. The Penn Medicine Center for Evidence-based Practice (CEP) has an AI policy from February 2024 and regularly updates its methods for gathering and analyzing evidence.
Clinical Workflow Integration
Chart Hero appears as a sidebar in the electronic health record (EHR) system, allowing clinicians to gather, arrange, synthesize, and interpret pertinent patient information with simple queries. It is an Epic-integrated conversational AI assistant. The tool is designed to streamline clinical workflows, reduce patient wait times, and support clinicians with visit preparation and documentation tasks. Penn Medicine is also integrating K Health's AI agents directly into its existing EHR systems and patient-facing digital platforms.
Decision Audit Trail
Chart Hero responses provide citations that link directly to data sources within the patient's chart, allowing users to reference the source material. This feature supports transparency and accountability in the AI's outputs. The software functions as a continuous layer that prepares the provider visit, optimizes workflow coordination, and seamlessly routes patient inquiries into an auditable pathway.
Physician Tip

Leverage Chart Hero to rapidly synthesize patient data and generate summaries before visits, saving significant time on chart review. Always critically review the AI's suggestions and sourced information, as human oversight remains crucial for validating outputs and addressing potential 'hallucinations.' Utilize the tool's ability to suggest potential courses of action as a starting point for your clinical decision-making, combining AI insights with your expertise. Provide feedback on the tool's performance to help refine its accuracy and functionality.

Chart Hero is an Epic-integrated conversational AI assistant, appearing as a sidebar within the EHR. Penn Medicine is also undertaking a multi-year collaboration with K Health to deploy AI clinical agents across its network EHR platforms, with an initial focus on virtual urgent care and plans to expand into primary and specialty care. This integration aims to streamline pre-visit data loops and insulate physicians from charting fatigue.

Details

Category Clinical Decision Support & Reference, Documentation & Scribing, General Productivity (AI Assistants)
Pricing Unknown
DeploymentOn-premise (internal system within Penn Medicine's electronic health record)
Mobile App1
API Available No
Compliance
BAA Available Yes AI-estimated
HIPAA Compliant Yes AI-estimated
FDA Status Cleared AI-estimated
Integrations
EHR Not specified
Specialties Dermatology, Emergency Medicine, Neurology

Support & Reliability

Ratings & Reviews

No reviews yet. Be the first to review this tool!

Rate Penn Medicine

Clinical Value
Ease of Use
Integration
Support & Docs
Value for Money
No press coverage or publications on file yet. Know of some? Suggest an edit →

Videos

Product demos, reviews, and walkthroughs for Penn Medicine.

View all on YouTube

Frequently Asked Questions

Penn Medicine's primary CDS, PennPathways, is accessible through the UPHS & Entity Intranets, PennChart (via UPHS Links >> Algorithms >> PennPathways), and SCiP Phones. SELECT PennPathways are deeply integrated into PennChart, allowing users to place orders, review lab results, and write data to flowsheets directly from the pathway.
While specific compliance rates for physicians using Penn Medicine's CDS are not readily available, the Center for Evidence-based Practice (CEP) aims to integrate best evidence into decision-making to strengthen care quality and safety. Penn Medicine's Information Services department also implements best practice clinical alerts, decision support, and reminders to enhance caregiver decisions. Compliance in research settings is facilitated through a review program that identifies noncompliance and offers training.
Penn Medicine's PennPathways program recently transitioned to a new tool called AgileMD, which offers improved functionality. While the system focuses on Penn-approved clinical guidance, external clinical decision support tools like UpToDate are widely used and offer evidence-based solutions, drug information, and CME opportunities. The integration of external tools would likely depend on Penn Medicine's IT policies and system compatibility.
Information regarding specific pricing for Penn Medicine's internal clinical decision support tools for employed physicians is not publicly available. Generally, access to such internal systems is part of the institutional resources provided to staff. For patients, Penn Medicine offers price transparency for procedures and virtual second opinions, which are self-pay programs.
While Penn Medicine is actively leveraging AI to improve clinical reasoning and patient care, including initiatives like CRISP, the rapid adoption of AI in healthcare is outpacing governance frameworks. Potential limitations and risks of AI-driven CDS include errors, miscommunication, clinician deskilling, privacy issues, bias, and data security vulnerabilities. The question of whether clinicians should remain 'in the loop' or relinquish some control to AI algorithms is also a point of discussion.
Anyone can submit requests for PennPathways by using their custom intake form and selecting 'Option 2 - PennPathways Request for evidence-based algorithms and associated supporting documentation.'
Yes, Penn Medicine's Center for Personalized Diagnostics (CPD) utilizes massively parallel sequencing panels to evaluate multiple targetable genes in cancer specimens, providing information for individualized therapies. This supports the clinical oncology team in tailoring treatment courses and outcomes.

Related Tools

Stanford University
Stanford University
Clinical Decision Support & Reference
Stanford Online's "AI in Healthcare" program provides a comprehensive overview of artificial intelligence's applications and implications within the medical field.
RADIN Health
RADIN Health
Clinical Decision Support & Reference
RADINA is an AI-powered platform by RADIN Health designed to optimize radiology workflows, enhance diagnostic accuracy, and streamline reporting.
Med-PaLM 2
Google
Clinical Decision Support & Reference
Med-PaLM 2 is a large language model from Google, fine-tuned for healthcare to provide high-quality answers to medical questions and support clinical decision-making.
Pacmed Critical Discharge
Pacmed
Clinical Decision Support & Reference
Pacmed Critical Discharge is an AI tool that predicts the probability of death or readmission within seven days of ICU discharge, providing explainable insights for physicians.
AiZtech Labs
AiZtech Labs
Clinical Decision Support & Reference
AiZtech Labs provides enterprise-grade AI solutions for healthcare, focusing on operational efficiency, diagnostic accuracy, and patient care workflows for providers and health systems.
Stethophone Pro
Sparrow Acoustics
Cardiology AI
Stethophone Pro transforms smartphones into advanced cardiac screening tools, enabling clinicians to capture, interpret, and visualize heart sounds for early detection of cardiovascular diseases.

See all Clinical Decision Support & Reference tools →

Suggest an Edit → | Report an issue | Last Verified: 2026-07-03 | First Added: 2026-06-14
AI Tool Finder
AI-powered search. Results may not be comprehensive.