3 AI-Based Technology Trends in Medical Imaging

Reading Time: 4 minutes read

Key takeaways from the 2023 RSNA Conference

The topic of artificial intelligence technology trends in medical imaging was once again infused throughout sessions and on display in the exhibitor hall at the 2023 Radiological Society of North America (RSNA) scientific assembly and annual meeting.

The need for imaging modalities to support an earlier, more accurate diagnosis continues to drive AI applications in the medical imaging market. The size of the market for AI in medical imaging is experiencing phenomenal growth from $1.12 billion in 2022 to $27.52 billion by 2029, according to Maximize Market Research, a global healthcare market research firm.

While more transformational change is coming, I would like to highlight three areas where AI is making a visible impact today.

Patient positioning

AI-based tools with varying degrees of automation and intelligence are enhancing medical imaging workflows including patient positioning. At RSNA, attendees showed strong interest in AI-driven systems that radiographers can use to help make patient positioning faster and more precise, and bring consistency to the process, all of which help improve image quality and reduce the need for retakes.

AI-based tools with varying degrees of automation and intelligence are available today to enhance medical imaging workflows.

Positioning is a time-intensive step in the image capture process. Even the most skilled radiographers can fail to get positioning just right. Then after getting the patient in the exact position, it is not unusual for a patient to move in the time it takes the radiographer to return to the console. Sometimes the slight change in position is unnoticed, resulting in the need for a retake, adding even more time to the image capture process.

Today sensors, cameras, and AI software all work together to automatically adjust equipment for each patient and exam, and alert the radiographer if the patient has moved so that corrective action can be taken. These smart capabilities enable the radiographer to correct positioning errors prior to imaging.

Image quality

AI and other algorithms also are improving image quality, which, in turn, helps enhance diagnosis and improve treatment planning. One notable AI outcome is advanced visualization techniques, also known as companion views, wherein images are processed for a specific interpretation task. Algorithms use sophisticated image analysis, multi-frequency enhancements, and grayscale transformations to accentuate features of interest, such as edges, lines, and fine textures. Examples of applications for advanced visualization include helping to locate tube and line tips or enhancing the appearance of a pneumothorax or collapsed lung, helping the physician see the area of interest more clearly.

Another image processing advancement that is rooted in AI helps balance noise and dose in images. These two factors are intertwined: the more an image is sharpened, the more noise is enhanced, which is not a good outcome. At RSNA, attendees saw AI technology that reduces image noise while retaining fine spatial detail. Notably clearer images are produced, and a better contrast-to-noise ratio is achieved at a broad range of exposures. This technological breakthrough is especially beneficial for the high-contrast detail necessary for musculoskeletal (MSK) imaging. The ability to lower radiation doses without a loss in image quality also has considerable benefits in neonatal and pediatric imaging where imaging at the lowest possible dose is critical.

Improving the patient experience

The AI advances in patient positioning have an additional important outcome: improving the patient experience. Patients undergoing a medical imaging exam are often worried, in pain, or both. Radiographers who spend less time on positioning the equipment are free to spend an added moment or two to reassure patients. Additionally, a speedier workflow gives patients added confidence that they are in the care of a competent radiographer.

The other side of the human experience is radiographers. Systems and technology displayed at RSNA demonstrated how AI-based tools give radiographers added confidence that they are capturing the best image possible and advancing the patient’s care. Using AI in the medical imaging process also frees up radiographers to focus on one of the more meaningful parts of their profession: interacting with patients.

Leaders in industries outside of healthcare regularly paint AI as being as significant as the industrial revolution in terms of economic impact, productivity, innovation, and overall impact on quality of life. In healthcare, it has the potential to improve diagnostic and treatment processes. While medical imaging providers continue to look toward the future, healthcare providers can embrace existing AI radiology solutions that are helping to improve clinical outcomes and enhance the imaging experience for patient and radiographers today.

About the Author

Vincent Chan is President and General Manager of Digital Radiography at Carestream.

A version of this article was originally published on AuntMinnie.com


This site uses Akismet to reduce spam. Learn how your comment data is processed.