Radiology mobile apps vendors take diverse approaches to developing solutions, depending on the intensions for the app, device and platform, and end user.
ORLANDO, FL - In many ways, the developers at MIM Software Inc. pioneered radiology mobile apps, clearing the way for other companies who have since entered the space.
Among the greatest challenges MIM Software faced when seeking FDA approval for a diagnostic radiology iPhone app was the device hardware. The FDA wanted the company to account for any hardware issues encountered when using the app in various environments. But MIM developers were creating with software in mind.
“The FDA was thinking in terms of hardware, but we were thinking software,” said Mark Cain, CTO of MIM Software, speaking at a SIIM 2012 session Thursday. “So we had to approach hardware issues. We had to handle the technical differences, such as the hardware not being manufactured for diagnostic use and [that people] could use the device in various environments.”
The main problem? Calibration. Because they couldn’t make changes to Apple’s hardware, they had to account for that in the software. They developed a way for users to visually calibrate their display by viewing a series of circles. They also created what Cain called the “tap test,” to address any environmental changes such as too-bright sunlight that might affect accuracy. Users are asked to tap on a dark square on the screen. If they don’t see it, you have to adjust the screen’s brightness or your environment, Cain explained.
These were just a few of the technical hurdles Cain and his team faced when seeking FDA approval. Their process took a few years, but Cain said he thinks the path will be different for developers seeking approval today.
“The FDA has come a long way and we got to watch it happen first hand,” he said.
Cain joined a panel of vendors to discuss mobile applications for radiology. Each vendor has taken a different approach, depending on the intensions of the app (MIM Software’s is approved for use as a diagnostic tool that can be used in absence of a work station.) and the end user. Some build apps only for Apple’s operating system; others included Andriod’s platform.
For example, Carestream’s app is built in HTML 5, which means it can run on any operating system, but the catch is that it might not take advantage of the capabilities of a specific device. Carestream’s motive wasn’t for a replacement to the diagnostic work station, but something the broader community of physicians and patients could access.
“We have been historically coming from position of radiology production - PACS, RIS, etc. - but understood that who you’re communicating this information to is continuing to broaden,” said Diana Nole, president of digital medical solutions for Carestream, speaking on the panel. “We decided to go with an agnostic approach from mobility standpoint.”
David Hirschorn, MD, of Staten Island University Hospital, who moderated the panel, noted that mobile devices - and how they are used - aren’t controlled like the computing environment of a hospital. You can’t dictate which operating system, device, or even environment, radiologists will use to access the app. “You can’t control how people will be accessing mobile devices and why and where, so the vendors are taking different approaches to meet these needs.”
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.