RSNA 2019 is here and I’m excited to share what’s new from the teams at MEDNAX Radiology Solutions and vRad – especially in the area of AI.
After years of preparation, we’re helping patients today with artificial intelligence on a very large scale with over 20,000 studies processed on the MEDNAX Imaging Platform daily. For example, over 6,500 studies pass through our AI models for worklist prioritization on a typical day to identify potential pathologies like intracranial hemorrhage (ICH), pulmonary embolism, pneumoperitoneum, and pneumothorax for escalation to our radiologists, producing very exciting results.
But we’re just getting started.
AI will soon permeate the practice and business of medical imaging, to the benefit of patients, practitioners, and administrators. In the coming years, we will see AI empower radiologists to rapidly detect ever more subtle, serious and rare abnormalities; conquer mountains of administrative burden; and strengthen healthcare delivery through demonstrated quality improvements, such as faster turnaround times, intelligent prioritization of critical cases, greater reporting accuracy, and compliant billing practices.
How AI works for us today
AI worklist prioritization to accelerate care delivery in critical cases has been a major success for us. One model is used to evaluate x-rays of patients with suspected pneumothorax, for instance, while another scans non-contrast head CTs for signs of ICH. Our models are all developed using custom-trained convolutional neural networks and have achieved a specificity of at least 90%.
With AI prioritization, 14 critical cases each day are read an estimated 10 to 15 minutes sooner than if the AI models had not run and 15 patients per month are provided care up to 24 hours sooner when escalated from non-emergent.
For the radiologists using these AI models, the experience is relatively transparent. You work as you normally would, choosing the next item on your prioritized worklist, and you won’t even know if it was AI that shot the study to the top of the list. Upon learning that AI had elevated the case of one Iowa patient with acute ICH, vRad neuroradioloigist Joshua Morais stated, "For this patient, I have no doubt AI was a critical element for her survival."
We're also using our AI models to ensure reporting accuracy and compliance. Using natural language processing (NLP), the MEDNAX Imaging Platform reviews all radiology reports, flagging potential clinical improvements or compliance issues. For example, a radiologist will be alerted if the pathology has a specific reporting requirement – such as Fleischner criteria for a lung nodule, or a LI-RAD score for a liver lesion – or when intelligent validation detects left/right or male/female anatomic discrepancies.
“We didn’t just build an AI platform for the sake of doing it; we were thoughtful about using AI in a way that made sense,” says Ricardo C. Cury, MD, FACR, chief medical officer of MEDNAX Radiology Solutions. “We have spent years developing a level of AI expertise that does not exist anywhere else. Because of that groundwork, we are able to use AI to improve patient care and the results are astounding.”
An invitation to learn
We were honored to be invited into the IBM Theater at this year’s RSNA to co-present with one of our key AI development partners, IBM Watson Health Imaging. Although the presentation filled up fast, we’ve added three encore sessions in the MEDNAX/vRad booth and invite you to stop by. Benjamin W. Strong, MD, chief medical officer for vRad, will join me in sharing patient case studies, details on our AI development program, and a sneak peek at what’s coming.
During Wednesday’s session, we’ll be joined by Mandy Long, global head of product management for Watson Health Imaging’s strategy and business development division, who will spotlight how MEDNAX and IBM are working together to develop these powerful AI algorithms for radiology.
Let’s talk at RSNA
Do you have questions? Want to talk AI with some of the industry’s leading experts? Stop by booth #1729 in the South Hall and pay the MEDNAX Radiology Solutions and vRad teams a visit.
For information on our many career opportunities, visit our careers booth #1029—also in the South Hall.
For the complete schedule of activities and where to find us, visit www.mednax.com/rsna-19/.