Introduction to Artificial Intelligence in Healthcare
Before it began making use of artificial intelligence applications, Jefferson Einstein, part of the Jefferson Health health system, faced a growing challenge in managing acute pulmonary embolism cases.
The Challenge
Rising imaging volumes and manual workflows – issues common across many health systems – limited the speed and efficiency of identifying and triaging PE cases, especially when time was critical. Traditional processes often struggled to keep pace with increasing imaging demands, which spanned multiple disciplines and modalities. Staff recognized AI had the potential to bridge the gaps in these workflows, ultimately helping achieve the goal of delivering the highest standard of patient care.
The Proposal
"At Jefferson Einstein, our culture of innovation and commitment to staying ‘on the bleeding edge’ of advancements drove us to explore how AI could revolutionize PE management by streamlining workflows, enhancing diagnostic accuracy and improving patient care," said Dr. Avi Sharma, director and associate chair of AI at Jefferson Einstein. Technology from health IT vendor Aidoc emerged as the ideal tool to address this critical need. "As global partners of ARC – The Center for Digital Innovation at Sheba Medical Center – and early adopters of AI in radiology, we had already implemented Aidoc’s platform for diagnostic radiology systems," he continued.
Meeting the Challenge
Implementing the AI technology revolutionized how Jefferson Einstein manages PE cases. The PERT now receives notifications through mobile alerts, allowing interventional radiologists to assess patient imaging and lab results from anywhere and at any time. Real-time updates ensure seamless collaboration across departments, significantly reducing delays. "The AI triggers a streamlined process: It alerts interventional radiologists, who then mobilize necessary resources to triage qualified patients to the operating room," Sharma explained.
Results
Hard results to date can be broken down into three major areas. First, an increase in clinically appropriate interventions. PERT intervention rates have increased by 73.8%, rising from 0.84% (17 interventions/2,022 CTPAs) pre-AI to 1.46% (32 interventions/2,191 CTPAs). Second, a reduction in time to treatment. "In a recent study, accepted to SIR 2025, we evaluated the impact of the system on reducing exam-to-needle time for patients with acute pulmonary embolism undergoing percutaneous thrombectomy," Sharma explained. And third, increased efficiency in radiology workflows. "Our radiology department has seen measurable gains in critical result communication turnaround times," he noted.
Advice for Others
Sharma said if his peers at other hospitals and health systems are considering AI, they should jump right in. "The best time to start using AI was yesterday," he quipped. "Focus on identifying the pain points in your workflows and select AI systems that integrate seamlessly with your existing systems, such as PACS and EHR platforms." Building a culture of trust and curiosity is critical, he added.
Conclusion
The implementation of AI technology at Jefferson Einstein has been a game-changer in managing acute pulmonary embolism cases. With increased clinically appropriate interventions, reduced time to treatment, and increased efficiency in radiology workflows, the results speak for themselves. As Dr. Sharma noted, AI is a long-term investment in patient care, and it’s not just about ROI, but about raising the standard of care across the board.
FAQs
Q: What is PE and why is it important to manage it effectively?
A: PE stands for pulmonary embolism, a blockage of an artery in the lungs that can be life-threatening if not treated promptly.
Q: How does AI help in managing PE cases?
A: AI helps in identifying and triaging PE cases more efficiently and accurately, allowing for timely treatment and reducing delays.
Q: What are the benefits of using AI in healthcare?
A: The benefits of using AI in healthcare include improved diagnostic accuracy, streamlined workflows, and enhanced patient care.
Q: What advice would you give to other healthcare organizations considering AI implementation?
A: Focus on identifying pain points in workflows, select AI systems that integrate seamlessly with existing systems, and build a culture of trust and curiosity.
