11385
Blog
Marlee Ravid

Avi Sharma, MD, Thinks the Best Time to Have Started Using AI was Yesterday

In a series of enlightening conversations with Avi Sharma, MD CIIP, Director of AI at Jefferson Einstein Hospitals, we gained valuable insights into his experience as an early adopter of AI in radiology and the trajectory of its development. Dr. Sharma shared his expertise on creating an initial AI strategy, fostering continuous education and the role of opportunistic screening and AI care coordination.

Creating an Initial AI Strategy: Understanding the Imaging Lifecycle is Key

When embarking on the AI journey, Dr. Sharma emphasizes that there’s no one-size-fits-all approach. He recommends starting by examining the imaging lifecycle to identify gaps and needs that AI could address. This could include opportunities for optimizing with AI for exam ordering, scheduling, interpretation and reporting and follow-up. “It’s important to keep the framework of the imaging lifecycle in perspective and understand where your opportunities are for AI to make an impact,” Dr. Sharma explains.

He advises healthcare institutions to “dive right in” rather than getting caught up in comparing different solutions. “At its core, we’re better off with AI than without it, and the best time to have started using it was yesterday,” Dr. Sharma states, highlighting the maturity and reliability of current AI solutions in radiology.

Continuous AI Education and Feedback Loops

Implementing clinical AI is just the beginning. Dr. Sharma stresses the importance of continuous education and enablement to ensure proper use of AI tools. He employs a strategic approach at Jefferson Einstein Hospitals, categorizing users into quartiles based on their receptiveness to new technology.

To engage the crucial middle 50% of users in the radiology department, Dr. Sharma leverages data-driven success stories. By showcasing efficiency gains achieved by respected colleagues, he creates a positive feedback loop that encourages more radiologists to adopt AI solutions. 

This approach involves:

  • Analyzing pre- and post-AI implementation data
  • Presenting findings at staff meetings
  • Encouraging peer-to-peer discussions about AI benefits
  • Tailored Educational Approaches

Dr. Sharma also emphasizes the importance of offering both low-touch (emails, surveys) and high-touch (office hours, one-on-one sessions) educational opportunities to cater to different learning preferences.

AI Expansion and Opportunities for Impact: Opportunistic Screening and Care Coordination

Looking ahead, Dr. Sharma shared excitement about the potential of AI in radiology and healthcare at large. He sees opportunistic screening as a prime example of AI’s future impact and ROI potential. For instance, AI-powered flagging of coronary artery calcification in routine imaging can flag high-risk patients that may be at increased risk of a cardiovascular event. Aidoc’s solution then curates a list of patients with relevant findings by reviewing their electronic medical records to determine if they have been seen by a cardiologist. This streamlined process allows the cardiology department at Jefferson Einstein Hospitals to prioritize patients for timely treatment, improving outcomes and benefiting both healthcare systems and patients. 

“I think the ROI is pretty self explanatory” he shares. “These are all examples of where there’s a positive benefit given to the healthcare system and their subspecialties, as well as, most importantly, the patient. The patient now has the information they need to take action and get the appropriate care that they deserve.”

Connecting All Points Of Care For PE patients

The CareCo mobile app has revolutionized care coordination for the Pulmonary Embolism Response Team at Jefferson Einstein hospital by streamlining communication and decision-making. Interventional radiologists can quickly assess patient imaging and lab results, enabling them to determine if treatment is necessary and immediately contact the appropriate teams, such as critical care or the emergency department. 

“I’ve heard repeatedly from our radiologists and IR doctors on call that having this tool, which gives them an early look at potential interventions, significantly reduces the time needed to get patients onto the table” Dr. Sharma shared. “ We’re exploring the impact on intervention time, and some of our IR doctors estimate that it’s been nearly cut in half. We’re reviewing the data and plan to publish our findings.”

Charting a Path Forward With AI

Dr. Sharma’s insights provide a roadmap for healthcare institutions looking to harness the power of AI at a health system level, with radiology at the helm. By creating a thoughtful initial strategy, fostering continuous education and keeping an eye on future developments, healthcare systems and department leaders can utilize AI technology to put their patients first.

Explore the Latest AI Insights, Trends and Research

Marlee Ravid