Authored by Arun Nagdev, VP of clinical services at Exo

The evolution of AI in healthcare

Artificial intelligence (AI) has become an integral aspect of healthcare. In medical imaging, AI has already begun to transform how clinicians capture and interpret images, so they can get the real-time information they need to make diagnoses. That’s especially true in point-of-care ultrasound (POCUS).

As technology evolves, we will continue to see the benefits of AI in both diagnostic care and even medical education. In this article, we will explore how AI can be an ideal tool for education for the growing number of POCUS users, such as medical students and the medical professionals of the future.

AI making it easier for medical education

The ultimate goal of any healthcare technology is to improve patient care, as well as equip healthcare workers with the ideal tools for clinical success.

In POCUS specifically, junior learners struggle with both image acquisition and image interpretation. AI can be invaluable aide helping an inexperienced learner defining what is an adequate view. Simply put, it tells the new learner that they have a good image. Once these images are obtained, AI can aide in clarifying a normal image versus pathology. This can change the slope of the learning curve and allow more clinicians the ability to incorporate POCUS into patient care.

Structuring workflow

Beyond image acquisition and interpretation, AI’s ability to recognize patterns can be tremendously impactful on improving workflows. POCUS directors spend significant amounts of time reviewing images for quality assurance and quality control purposes. AI can be a huge aide here as well. Rapidly recognizing normal from abnormal to allow for a more efficient workflow for the POCUS director.

POCUS experts can incrementally reduce QA/QC time and spend more time in clinical education, research and other higher-value work. This speaks to the use case for AI in medical imaging and healthcare more broadly, which is that AI can support repetitive tasks based on recognizing patterns so that staff can spend more time on activities that more directly benefit the patient.

Innovating medical imaging with AI

In recent surveys conducted by Pew Research Center, around 60 percent of Americans said they would be uncomfortable with clinicians relying on AI in their healthcare. Over 50 percent of Americans also believed using AI to diagnose diseases and suggest treatments would harm patient-provider relationships.

While these concerns are understandable, it’s important to grasp that AI is a tool that will help clinicians work more efficiently with clear oversight. The capacity to build operator independence, reproducibility, accuracy, and consistency across all ultrasound exams will immensely benefit the broader world of both caregivers and patients. Coinciding with the rise of other technological advancements, hospitals, clinicians, physicians, and anyone involved in the medical field should expect to see AI play a greater role in their work in the years to come.

If implemented correctly, that can be an overwhelmingly positive development for seasoned doctors, junior learners, and patients alike.

Verifying the right AI solution

One major requirement when evaluating an AI solution for an institution is whether it has FDA clearance. But I would also encourage readers to seek out AI that has been designed and trained using large test sample sizes from various high-quality and low-quality images. Sonographer-derived images that are “perfect” may not correlate well with the novice POCUS user who is learning to obtain ideal images.

Leading the way for AI ultrasound through education

In my opinion, for POCUS to scale, integrating AI into education is key. Allowing more novice providers to obtain rapid, real-time feedback from an integrated AI tool will aide in adoption, education and use. I am excited to share my vision for POCUS education and launch Exo U, Exo’s AI-driven educational coach for POCUS. It takes an actionable approach with TikTok-styled videos, all under 30 seconds long. The videos are highly accessible and embedded in the Exo Iris® app.

Not convinced about AI? Skepticism is reasonable and should be expected when using technology in patient care. Exo AI has been rigorously tested, and FDA-approved. To learn more about Exo’s AI-driven point-of-care ultrasound, check out https://www.exo.inc/ai.