<img height="1" width="1" style="display:none;" alt="" src="https://dc.ads.linkedin.com/collect/?pid=54935&amp;fmt=gif">

Latest Posts

Beyond theory: Building on the success of radiology AI

Our AI models are at work today prioritizing critical cases, improving reporting accuracy and assisting billing compliance. But we’re just getting started. AI will permeate the practice and business of medical imaging to empower radiologists, conquer mountains of administrative burden and strengthen health care delivery.

Following are some strategies we are using to optimize the performance of AI currently in use, while realizing the potential of radiological AI to reduce the cost of care and mitigate risk.

Enhance sensitivity and specificity of current models

Radiologists currently working with MEDNAX Radiology AI applications “in the wild” are providing critical data and insights that are shaping and defining the next generation of AI tools. We continually compare the models’ results with actual case reports. If a model prioritizes a suspected case of pulmonary embolism, the radiologist report should validate that finding. However, if the algorithm generates a false positive or false negative based on the radiologist report, the report data can be used to improve future performance.

Expand the list of specific pathologies prioritized by AI

At vRad, we are working with a prioritized list of pathologies that present urgent threats to patients at a relatively high rate of frequency. We are currently training individual algorithms to accurately identify these pathologies. As the algorithms achieve acceptable levels of sensitivity and specificity, they will be integrated into the MEDNAX Imaging Platform used by our physicians.

Empower radiologists to meet present and future challenges

Radiologists are recognizing AI not as a substitute for their skills, but as an essential tool to help them provide the best quality patient care in a rapidly changing environment. As the medical industry continues to morph, MEDNAX is advancing AI to help radiologists respond.

Where increasing demand for studies is driving higher image volumes — AI will help cull inconclusive images, enabling radiologists to focus on the critical scans.

Where a flat-to-declining number of radiologists in the field is placing more pressure on individuals — AI can promote more efficient study assignment and reporting, empowering radiologists to handle larger caseloads.

Where spiraling regulatory requirements can distract radiologists from patients — AI will reduce administrative burdens, enabling radiologists to focus “eyes on images.”

Risk mitigation through AI

We are already demonstrating that AI can improve diagnostic reporting quality, accuracy and turnaround times. Consider the cumulative impact over time of fewer negative outcomes and reduced exposure to litigation. Risk mitigation as a result of AI can benefit the entire health system by reducing legal incidents and deflating medical malpractice liability reserves. Already, we are seeing third-party payer engagement on the issue of reimbursement and demonstrated quality improvements. We anticipate such engagement to continue.

Whether you’re a radiologist wanting the power of AI at your back or a health care administrator seeking new advantages for your organization and patient population, I invite you to learn more about how vRad and MEDNAX Radiology Solutions is building the future of radiology.

Author Benjamin W. Strong, MD

Chief Medical Officer, Education Committee Chair. Dr. Strong is at the forefront of efforts to expand access to quality, affordable care through telemedicine. As CMO for the nation’s largest radiology practice, he collaborates with radiologist and hospital partners, uncovering opportunities to enhance the practice environment. Dr. Strong first completed residency in internal and emergency medicine, and later was drawn to the fast-paced flow of diagnostic puzzles that is the practice of radiology.

    Related Posts