Summary
Jefferson Einstein uses AI to improve pulmonary embolism management, resulting in faster treatment and better patient outcomes. The AI system flags potential PE cases, allowing for quick intervention and improved collaboration among medical teams. This technology has significantly reduced time-to-treatment and increased clinically appropriate interventions.
Healthcare data growth can be overwhelming scale effortlessly with TrueNAS by Esdebe.
** Main Story**
AI-Powered Revolution in Pulmonary Embolism Management at Jefferson Einstein
Jefferson Einstein, part of the Jefferson Health system, has implemented a cutting-edge AI solution to address the growing challenge of managing acute pulmonary embolism (PE) cases. Before adopting this technology, the hospital faced increasing imaging volumes and relied on manual workflows, which limited the speed and efficiency of identifying and triaging PE cases, especially in time-critical situations. Recognizing the potential of AI to bridge these gaps, Jefferson Einstein partnered with Aidoc, a health IT vendor, to implement an AI-powered platform that has revolutionized their PE management process.
Real-Time Alerts and Streamlined Workflows
The AI system developed by Aidoc flags suspected PE cases in real-time, automatically identifying potential cases from CTPA scans and notifying the Pulmonary Embolism Response Team (PERT) through a mobile application. This proactive approach ensures timely treatment and facilitates rapid mobilization of the multidisciplinary PERT for urgent interventions. Moreover, the technology seamlessly integrates into Jefferson Einstein’s existing systems, including Fuji Synapse PACS and Cerner EHR, minimizing disruption to workflows and empowering both diagnostic and interventional radiologists to work more efficiently. This integration has been crucial in optimizing the PERT workflow, significantly reducing delays in diagnosis and treatment.
Improved Patient Outcomes and Reduced Treatment Times
The implementation of this AI technology has yielded remarkable results in three key areas:
-
Increased Clinically Appropriate Interventions: PERT intervention rates have increased by 73.8%, demonstrating the AI’s effectiveness in efficiently triaging critical cases and ensuring high-risk PE patients receive timely care. The increase from 0.84% (17 interventions/2,022 CTPAs) pre-AI to 1.46% (32 interventions/2,191 CTPAs) underscores the impact of AI in identifying patients requiring intervention.
-
Reduction in Time to Treatment: A recent study accepted to SIR 2025 demonstrated that the AI alert system reduced overall exam-to-needle time for patients with acute PE undergoing percutaneous thrombectomy by 20% – from 148 minutes to 119 minutes. This reduction in time to treatment is crucial as it can significantly impact patient outcomes and minimize the risk of long-term complications. The improved interdisciplinary collaboration among radiology, emergency and critical care teams is a testament to the effectiveness of the AI-powered system.
-
Increased Efficiency in Radiology Workflows: The mobile alerts generated by the AI platform allow interventional radiologists to access patient imaging and lab results from anywhere at any time. The seamless communication and real-time updates foster collaboration between departments, reducing delays and enabling prompt decision-making. The integration of AI has streamlined the process, allowing interventional radiologists to mobilize resources quickly and efficiently.
A Model for AI Integration in Healthcare
Jefferson Einstein’s success with AI-powered PE management serves as a compelling example of how AI can transform healthcare. The hospital has not only achieved significant improvements in patient care and operational efficiency but has also incorporated AI into its educational efforts, training radiology residents on the system during orientation. This forward-thinking approach prepares future healthcare professionals for the growing role of AI in medicine. As AI continues to evolve, initiatives like this will pave the way for wider adoption of AI-driven solutions across healthcare systems, ultimately benefiting both patients and medical professionals alike.
Exam-to-needle time reduced by 20%? So, if I understand correctly, AI is now faster at sticking needles in people than humans are? Does this mean my doctor’s office will soon be staffed entirely by robots with excellent bedside manner algorithms? I’m only asking for a friend… who is *definitely* not scared of needles!
That’s a funny take! While the AI isn’t *actually* sticking the needles, it’s speeding up the process of identifying patients who need intervention and alerting the right teams faster. Think of it as AI helping doctors be even more efficient, not replacing them (yet!). Perhaps slightly better bedside manner is achievable from an algorithm. Thanks for the comment!
Editor: MedTechNews.Uk
Thank you to our Sponsor Esdebe
PERT intervention rates up 73.8%? Suddenly feeling very inadequate about my own intervention rates with pizza and Netflix. I wonder if there’s an AI for that? Asking for a friend… who may or may not be wearing elasticated trousers.
Given the increased efficiency in radiology workflows due to mobile alerts, could this AI integration model be adapted to improve response times and coordination in other time-sensitive medical emergencies, such as stroke or cardiac arrest?
Given the significant improvement in clinically appropriate interventions, how does the AI system differentiate between PE cases requiring intervention versus those that can be managed conservatively, and what data points are most influential in that determination?