With the invention of the X-ray image in 1895, diagnostic medicine for orthopaedic issues made a huge leap forward, eventually bringing us the MRI and CT scan. This technology has allowed us to capture clearer, more accurate pictures of orthopaedic problems. However, the actionable information we’ve been able to develop from that data has changed very little since the X-ray’s inception over a century ago. As digitization marches forward, new technologies are being developed to take our orthopaedic diagnoses to the next level.
Reducing Inconsistencies with Digital Radiology Imaging
Radiology provides us with a huge amount of useful information, but its storage in an unstructured format doesn’t deliver a lot of value for medical professionals dependent on EHRs to convey and spread that information. This is an opportunity to put radiologists in the spotlight, transforming how reporting massive amounts of imaging information is processed and delivered to the physicians, patients and caregivers who need it.
For clinicians, EHRs serve a dual purpose: getting patients into the proper care pathway and acting as a resource for inputting and retrieving information. Without context, raw data is virtually useless to clinicians. But when data is transformed into higher levels of information in the EHR, it allows clinicians to guide patients to the right treatments. Orthopaedic practices work especially closely with radiological data to deliver accurate care, so it’s important to recognize how the standardization of this information can play a key role in patient outcomes. For example, a recent study sent a 63-year-old woman to ten different imaging centers around the greater New York City area for MRIs to see what variations happened in the interpretation of her results, with two additional MRIs performed at the institute performing the study. It verifies why high-level spine surgeons don’t start treating a patient based solely on the radiology report without viewing the imagery themselves first. Why?
Not a single finding, of the 49 distinct findings reported, was unanimous across all ten reports. This was despite the fact that seven of the ten centers used closed 1.5T MRI systems, typically recognized for higher levels of clarity and readability. This level of inconsistency makes selecting the proper care pathways virtually impossible for a treating physician without having to spend precious time reassessing the radiologist’s conclusions for accuracy.
Artificial intelligence (AI) development in radiology helps reduce inconsistencies in image interpretation and quality, creating a true partnership between radiology and orthopaedics. In fact, just last year, the FDA cleared the way for one of the first AI algorithms intended to help with clinical decision support. The software uses imaging analytics to scrutinize wrist x-rays and detect distal radius fractures. Strong partnerships in the future between imaging software companies that incorporate artificial intelligence and EHR software developers would integrate the data and produce the next step in the care process at a fraction of the time and cost to do so today.
Imagine a system in which thousands of medical journal articles are digitized, human anatomy is quantified and algorithms are integrated to allow the EHR to become an intelligent decision support tool. We have a huge amount of data available today, but integrating it is a challenge. However, once it is integrated, care pathways become automatic and human error is dramatically decreased. AI in imaging and EHR data integration could bring actionable information more quickly to the orthopaedic surgeon’s fingertips – luckily, the seeds of the future have already been planted.