Very few HIEs transfer medical images, but that data together with an exchange approach that aggregates patient data will reveal HIE’s potential, say experts at SIIM 2013.
GRAPEVINE, Texas - Only a tiny fraction of the health information exchanges in the U.S. today share medical images, but sharing that data holds incredible promise for patient care.
The obstacle for transmitting medical images across a health organization? Medical images are enormous and can slow down networks not equipped for the transfer.
The solution? New web-enabled technologies already available or being developed for healthcare.
“HIEs today have had a small impact compared to what they could have,” said James Philbin, PhD, of Johns Hopkins Medical Institutions, speaking Thursday on a panel at the SIIM annual meeting. HIEs have the potential to transform care, he added.
Put simply, HIEs are designed to transfer and share health information across organizations, allowing for a multi-directional flow of data between providers, Philbin explained. These electronic systems provide safe and secure access to patient information.
There’s a wealth of untapped potential in HIEs, he said.
For starters, of the roughly 300 HIEs today, less than 2 percent support medical images, he estimated. That’s because most of the data shared is small – demographics, prescriptions, allergies, payer information, for example. Medical images are extremely large, and current solutions don’t necessarily support that transfer. However, web technology - such as an architecture called RESTful (representational state transfer) - can be adopted to allow for transfer of medical images.
As image sizes and volumes grow, these technologies will become crucial, Philbin said.
Potential also rests in the model adopted by the HIE. Philbin outlined two models: exchange and archive, noting the latter should be the future of health information exchange.
The exchange model relies on a centralized shared database, with edge devices that infterface with the local IT systems. Data is discarded at the receiving end after a short period of time.
On the other hand, the archive model similarly relies on a centralized registry and repository, but the documents are kept for a much longer period of time. Herein lies the promise.
“This model has the opportunity for an aggregated patient record,” Philbin said. This could hold the key to making HIEs a sustainable business model, as they could manage medical data for providers. This could also pave the way for data analytics, he said, adding “That’s one of the ways the HIEs can actually add value.”
The technology for large scale HIEs that share medical images is not available today, but will be soon. Philbin said HIEs could see these capabilities in a few years, and taking an approach that aggregates and manages patient data will unlock HIE’s potential.
Deep Learning Detection of Mammography Abnormalities: What a New Study Reveals
June 19th 2023In multiple mammography datasets with the original radiologist-detected abnormality removed, deep learning detection of breast cancer had an average area under the curve (AUC) of 87 percent and an accuracy rate of 83 percent, according to research presented at the recent Society for Imaging Informatics in Medicine (SIIM) conference.
Detecting Intracranial Hemorrhages: Can an Emerging AI Advance Have an Impact?
June 9th 2022A 3D whole brain convolutional neural network could provide enhanced sensitivity and specificity for diagnosing intracranial hemorrhages on computed tomography, according to new research presented at the Society for Imaging Informatics in Medicine (SIIM) conference in Kissimmee, Fla.