Artificial intelligence improves the potential of intravascular OCT

Artificial intelligence‘s (AI) applicability in cardiac imaging is growing rapidly and was a major topic of discussion at this year’s EuroPCR 2022 meeting. Many session speakers discussed how they use AI tools in their day-to-day practice and in their research to improve decision-making and patient/research outcomes. However, it’s no secret that AI tools are only as good as the data sets and the thousands of expert opinions that use them.

Implementing AI applications in our day-to-day practice, from an operational point of view, could mean adapting clinicians’ workflow and freeing up time to set up and train the new systems. And from an efficacy standpoint, it leaves clinicians wary of the accuracy of the results, especially if they’re unsure of how good the data is used to power the technology. However, a good AI tool improves workflow, has a proven ROI, and is trained using data samples representative of our population and a variety of expert opinions. When it comes to interpreting and analyzing cardiac imaging, artificial intelligence must seamlessly provide clinicians with an immediately available second opinion, without jumping through hoops or doubting its accuracy. In front of intravascular optical coherence tomography (IVOCT), if applied correctly, AI has the potential to drive the adoption rate of imaging technology, improve decision-making and patient outcomes.

I started my business, Dyad Medical, to help address the challenges cardiology experts face. We develop clinical applications that leverage cloud-based AI technology to enable clinicians and researchers to interpret medical images more efficiently and accurately. We focus exclusively on imaging the heart, because more people die from cardiovascular disease than all forms of cancer, diabetes and accidents combined. Our goal is to arm clinicians and researchers with AI-powered tools that provide them with immediately available second opinion to improve their practice.

However, we would not develop these solutions without the challenges we experience in our field. Many studies suggest that applying imaging technology promotes better patient outcomes, but the main barriers are the need to learn specific image analysis skills and more time, which we simply do not have, to analyze images manually. I believe we can use AI to overcome these barriers, but we still need to fully understand why imaging modalities are not fully exploited and the limitations of AI in our field.

Intravascular OCT Adoption Rates

Today, high-resolution intravascular imaging helps reduce adverse cardiac events by optimizing stent implantation and reducing structural risk factors for stent failure. IVOCT offers better image quality and therefore very accurate measurements. The high resolution of IVOCT makes it easier to evaluate the placement of the stent support against vessel walls. It is the only imaging modality with the resolution and contrast that can optimize stent design.

Despite improved patient outcomes, intravascular imaging remains severely underused in clinical practice. IVOCT requires a great deal of interpretation skills to understand the nuances of cardiac images. Until now, the adoption of IVOCT imaging has been hampered, especially among less experienced clinicians, because image interpretation requires operators to be proficient in manual image analysis and able to interpret a large amount of data in real time. As a result of these challenges, clinical decisions about intravascular imaging are often based on insufficient and/or incomplete data.

The breaking point: a shortage of time and expertise

We are reaching a breaking point with imaging technology where the small improvements to today’s standards of manual analysis succumb to time constraints. In addition, disagreements among experts, diagnostic error rates and a shortage of professional experts are more common as burnout is rampant in labs, health systems and private practices.

Even for trained experts, today’s rapid clinical workflow precludes extensive evaluation of intravascular images during the procedure, a process that can require hours of manual labor. At present, there are no existing clinical software tools that can perform comprehensive automated intravascular image analysis.

Trust based on data

While we understand the practical and operational barriers to: image technology and the adoption of AI tools, above all, we need to be able to rely on the technology we use is evidence-based. AI programs must be trained on high-quality data, rigorously vetted, and based on expert insights. AI technology is certainly not perfect, and it is important to note that perfection is not the end goal. But to trust the AI ​​tool’s accuracy, we need to look deeper into the dataset and expert opinions driving image analysis.

Today’s fast-paced clinical environment requires us to treat more patients in less time. In our highly technical and highly specialized field, analyzing images quickly inevitably leads to errors, burnout and disagreements among experts. Advances in technology can ease the burden, but can we rely on these high-tech tools to produce better outcomes for our patients?

The uncertainty lies in not knowing the quality of the data used to train the AI ​​tools.

This all comes down to two criteria: 1) training the AI ​​tool with representative data samples; and 2) relying on many expert opinions.

First, our data samples must be representative of the patients we serve. For example, if the AI ​​imaging tool is intended for use in the United States, the images we analyze to power the tool must come from a diverse group of patients from different demographic backgrounds, similar to the makeup of the US population. This includes different age groups, physical characteristics such as height and weight, medical history such as surgery and medications, genetics such as family history of heart disease, and other variables that affect a patient’s overall health.

Second, the expert analyzes must come from many different sources. Ideally, we would be able to analyze images objectively and let no bias get in the way. The reality is that we all have a unique way of approaching our work and disagreements between experts will be inevitable. Therefore, it is paramount to rely on many opinions when training an AI algorithm. The more expert opinions driving the AI, the better the overall accuracy of the analytics.

AI is becoming an integral part of value-based care

Cardiovascular health care costs are projected to reach $1.1 trillion by 2035. These rising costs are partly due to the shortage of cardiologist specialists, high workload and burnout. AI that helps solve these problems can help improve patient outcomes and lower overall healthcare costs.

Addressing workflow inefficiencies and finding opportunities to automate time-consuming tasks is today’s challenge. AI will not replace clinicians, but it will improve the process of treating patients. For cardiac imaging, this means that tomorrow’s AI technology will significantly reduce misdiagnosis and expert disagreement, minimize healthcare costs by integrating it into workflows, and provide universal access to thousands of second-expert opinions. The end result is the support clinicians and researchers need to make decisions and draw conclusions that will improve patient outcomes and reduce healthcare costs.

Ronny Glide PhD

Ronny Shalev, Ph.D., is CEO and founder of Dyad Medical Inc., a company developing FDA-approved software that automatically analyzes the contents of heart and cardiovascular images using artificial intelligence. He has spent much of the past 25 years in leadership roles, including VP of Sales and Marketing at Orbotech, where he led teams of >100 people worldwide, and Director of the World-wide Program Management at Marvell Semiconductor. An expert in new project development, medical imaging, machine learning and visualization, he has a significant amount of entrepreneurial experience and is committed to using his skills to help physicians make accurate decisions to improve patient outcomes. He has a Ph.D. in electrical engineering and computer science from Case Western Reserve University.

Related AI Content:

Viz.ai launches two new AI-powered modules for pulmonary embolism and aortic disease

VIDEO: Trends in Artificial Intelligence in Medical Imaging — Interview with Sanjay Parekh, Ph.D.

Aidoc Releases First ‘OS’ Uniting AI in Enterprise Imaging to Address Technical Barriers to Using AI

GE Healthcare Receives FDA Approval for First X-ray AI to Help Assess Endotracheal Tube Placement

RSNA Announces Results of Brain Tumor AI Challenge

AI Gut Imaging Software Takes on US with FDA Approval and National Hospital Partnership

Riverain Technologies Expands Partnerships for Automated, Efficient End-to-End Lung Cancer Screening Capabilities

Nuance Launches AI-Powered Precision Imaging Network to Improve Clinical and Financial Outcomes Across the Continuum of Care

Philips integrates MedChat’s AI capabilities to optimize automated patient communication and hospital workflows

VIDA streamlines clinical respiratory studies with AI

Leave a Comment

Your email address will not be published. Required fields are marked *