Practical uses of fracture detection AI in the clinical setting

A neighbor took a tumble off a ladder while trimming trees. Your child fell while skateboarding with friends. An elderly relative slipped in the kitchen and is now experiencing pain in their right hip. What do all of these situations have in common? The potential for bone fractures or joint injuries, that if not addressed can lead to other health issues.

X-rays of the bones are a simple and easy way for radiologists to view and assess bone fractures, potential injuries, or joint abnormalities. Can artificial intelligence (AI) play a role in fracture detection? With the growing volume of imaging studies, busy emergency departments, and a shortage of radiologists, it appears AI can help by acting as an aid to radiologists, helping to speed and improve fracture diagnosis. This week’s Domain Knowledge highlights recent developments in fracture detection AI.

AI technology is a powerful tool for image recognition, which can be seen in this study where AI outperformed clinicians in detecting hip fractures. As discussed in this same study, however, the real concern lies with its usability in the clinical setting more than the performance. One common misinterpretation surrounding the use of AI by radiologists is looking at AI as a battle between human and machine. In reality, AI is developed with the objective of boosting radiologists. This study explains how the biggest potential in fracture detection AI comes when AI is used adjunctively with human readings, producing much better results than either one alone.

Gleamer’s BoneView AI companion is a good example of an AI tool that creates a smooth integration with clinical practice while supporting radiologists. BoneView acts as a second read that can detect about 30% more missed fractures and bone lesions. This hip fracture case study demonstrates how AI works in clinical practice.

AI and machine learnings moment in health care offers another example highlighting the practical use of AI in fracture detection. Mercy Radiology has seen an improvement in the quality of reporting and positive engagement from clinicians as well.

Nano-X has an FDA approved AI tool that supports clinicians in the evaluation of musculoskeletal disease, identifying compressing fractures in the spine. This AI algorithm is different than those mentioned above as it reads CT scans of the spine, however, the end goal of boosting clinicians’ efficiency is the same.

Wrist fractures are the most common type of fracture seen in emergency departments and account for 25% of fractures seen in children. That coupled with evidence that suggests wrist fractures are one of the most commonly missed fractures on x-rays, has clinicians looking to AI for a second read. A recently developed database of pediatric wrist X-ray images provides a boost to AI, as it allows for more reliable training of algorithms.

That’s a wrap for this week’s review of news and happenings in the healthcare AI space. In closing, I’ll leave you with an invitation to learn more about the benefits of using an AI Hub to manage multiple applications across your clinical needs and offer you a personal demo of Ferrum’s platform and growing AI catalog.

If you have AI tips, suggestions, or resources you’d like to share leave us a note below, and please feel free to suggest topics you would like to see covered in future posts.