Diagnostic Accuracy of X-rays Interpreted Using Cell Phone App

Researchers found that diagnostic accuracy via cell phone app was on par with traditional digital imaging and internet-based workstation viewing.
Researchers found that diagnostic accuracy via cell phone app was on par with traditional digital imaging and internet-based workstation viewing.

A chest X-ray sent to a smartphone via the WhatsApp® instant messaging service has a diagnostic accuracy that is on par with a chest X-ray viewed at a traditional digital-imaging, Internet-based workstation, according to results released at the 2017 Annual European Congress of Radiology in Vienna, Austria.

The study, by a team from Ireland, found no differences between the two viewing platforms in their ability to identify or exclude acute pathology.

“Our study shows that WhatsApp® is a reliable tool to communicate radiological images without any significant loss of diagnostic information,” principal investigator Guy Handelman, with the Royal College of Surgeons in Ireland Education and Research Centre at Beaumont Hospital, in Dublin, said.

For their study, the investigators recruited 6 physicians who were representative of a conventional non-radiologically–trained on-call team.

The physicians, who were residents from medical and surgical subspecialties, compared the diagnostic accuracy of the two viewing platforms for interpreting a series of 100 chest X-ray images. The chest X-ray images were specifically selected to represent standard emergency chest X-ray findings.

“The use of instant messaging applications is ubiquitous in modern hospitals, with clinical teams using it to communicate quickly with other clinicians,” Handelman pointed out. “While unwell patients who have imaging out of normal working hours benefit from quick second opinions and advice from senior on-call staff, there is some inherent degradation in quality when images are captured on a smartphone camera, and thus the safety of such a practice is in doubt.”

Prior assessment of the image quality of chest radiographs sent via a multimedia messaging service found the image quality to be too poor to be clinically useful, he added. Since then, however, screen and camera quality have increased dramatically, resulting in increased use of the technology by on-call teams.

There were no statistically significant differences between the two modalities in sensitivity (44.99% for WhatsApp® vs 53.98% for the workstation, P=.2581), specificity (87.02% vs 92.36%, P=.1371), positive predictive value (46.35% vs 61.81%, P=.1837), or negative predictive value (87.82% vs 90.37%, P=.1972).

“The principal findings show that the use of WhatsApp® to communicate chest X-rays is reproducible with respect to viewing the same images on a workstation,” Handelman said.

He emphasized, however, that “there are technical considerations to this practice which may have medico-legal ramifications.” In transmission of images, WhatsApp® compresses the image to conserve the amount of data that need to be transmitted, which may lead to a significant loss of image resolution and the possibility of artifact.

He was quick to point out that “technical aspects of fidelity loss are probably secondary to the operator-dependent quality of photography.” Of note, his team has designed an image-acquisition construct that holds the phone still at a set distance and angle to the screen, which allows the physician to take photos using voice commands.

Reference

Handleman G, Rogers A, Babiker Z, et al. WhatsApp® with media messaging in diagnosis of acute CXR pathology — an inter-observer study among residents. AbstractC-2675. Presented at: 2017 Annual European Congress of Radiology; March 1-5, 2017; Vienna, Austria.

You must be a registered member of Endocrinology Advisor to post a comment.

Sign Up for Free e-Newsletters