From skin to bones: The benefits of a common labeling language for enterprise imaging

Every medical specialty uses imaging. A patient might have X-rays from a ten-year-old fracture along with MRI scans and surgical photos from a recent cancer. The wealth of information affords doctors a more detailed look at a patient.

Enterprise imaging brings all of the specialties together. A photograph of a melanoma by a dermatologist, say, may be of interest to a patient’s primary care physician when trying to recommend care or next steps.

Making all this imaging available to relevant clinicians in an easily accessible way can deliver more efficient and integrated care. How these images are presented and searchable, however, can get pretty complicated, says Dr. Alexander Towbin of Cincinnati Children’s Hospital.

“A picture is worth a thousand words, and that’s generally true but it’s because we understand the picture,” said Towbin, who will present at HIMSS21 in Las Vegas next month. “As a radiologist I can tell the story of an X-ray because I can understand the image.”

Dr. Towbin is among a group of imaging experts working to define a common ontology of body part labeling across enterprise imaging. This language will allow a variety of clinicians to access and understand a patient’s full relevant data, he says.

At HIMSS21, Towbin will discuss how the HIMSS-SIIM Enterprise Imaging Community plans to advocate for the ontology to be incorporated as part of existing standards such as DICOM and HL7 FHIR.

He’ll describe the three metadata elements needed to selection of a relevant comparison imaging examination; explain how experts were convened to select a standard body part ontology for use in enterprise imaging, and preview the plan to foster adoption of a standard body part ontology for use in enterprise imaging

The specificity and complexity of the medical landscape makes the task a challenging one. Towbin envisions a framework that works across specialties, practices, and even borders. How much detail is too much?

“Within radiology, when I image a chest that includes everything like the heart, throat, and bones,” he explained. “When a dermatologist images, it’s just skin. When a pathologist images, it’s down to cells. We need an imaging language that works across all of that, that everyone can relate to.”

A common language for imaging means different specialties can see relevant images for a patient from earlier on in their care process. Called a “hanging” or “downstream” protocol, Towbin says that these automated workflows can streamline the care process, putting the right information in front of the right clinician at the right time.

“Being able to see all of the images next to each other, each specialist can see … the patient more holistically,” he says.

The process of developing an industry-standard body part ontology, at least at the start, may be a slow one. Finding a common language that can suit both an oncologist and a physical therapist without being over- or under-specific means gathering input and buy-in across a range of specialties and organizations.

“What we’re trying to do is going to take vendor support, society support, support from standards bodies, informatics pros, physicians and nursing,” says Towbin. “We want representation from across the medical spectrum.”

Dr. Towbin, along with his co-presenter, Kim Garriott, chief innovation officer for Global Healthcare at NetApp, will offer more detail in their session “Importance of Body Part Labeling in Enterprise Imaging.” It’s scheduled for Tuesday, August 10 from 2:30-3:30 p.m. in Caesars Summit 211.

Source: Read Full Article