Abstract
The smartphone client-server teleradiology system appears promising and may have the potential to allow urgent management decisions in acute stroke. However, this study was retrospective, involved relatively few patient studies, and only two readers. Generalizing conclusions about its clinical utility, especially in other diagnostic use cases, should not be made until additional studies are performed.
Introduction
In acute stroke management, "time is brain." Every minute of ischemia results in the loss of approximately 1.9 million neurons. Rapid diagnosis via Computed Tomography (CT) is essential to determine the appropriate intervention, such as thrombolytic therapy or surgical thrombectomy. However, specialists are often away from high-end medical workstations when emergencies occur.
Traditional mobile teleradiology faces three significant hurdles: high bandwidth requirements for large DICOM datasets (often hundreds of megabytes), security risks associated with storing sensitive patient data on portable devices, and the limited processing power of smartphones for advanced 3D reconstructions.
System Architecture
To address these challenges, researchers implemented a client-server architecture using ResolutionMD Enterprise software. Unlike traditional systems that download raw data to the device, this approach offloads heavy image rendering to a remote server.
The server processes the DICOM images and transmits only the rendered frames as compressed JPEGs to the smartphone. User interactions, such as scrolling or zooming, are sent back to the server, which then renders and returns the updated view. This "zero-footprint" approach ensures no patient data is permanently stored on the mobile device, mitigating HIPAA and privacy risks.
Clinical Evaluation
The system's efficacy was tested in a retrospective study using 112 Noncontrast Computed Tomography (NCCT) brain scans and 65 Computed Tomography Angiogram (CTA) head scans from the Calgary Stroke Program database. Two neuroradiologists compared interpretations made on a standard medical workstation against those made on an iPhone or iPod Touch. The study focused on three critical markers: intraparenchymal hemorrhage (ICH), early signs of ischemia, and vessel occlusion.
Results and Discussion
The smartphone system demonstrated near-perfect agreement with the gold-standard workstation. Hemorrhage detection achieved 100% sensitivity and specificity. Accuracy for identifying ischemic changes exceeded 98%, while vessel occlusion detection showed 94.4% sensitivity and 100% specificity. Notably, there was no significant difference in interpretation time between the workstation and the smartphone, indicating that radiologists could work just as efficiently on the mobile platform.
The study also highlighted "perceptual compensation," where radiologists naturally held the phone closer to their eyes to offset the smaller screen size. Combined with the device's high luminance and pixel density, the diagnostic quality remained high. While Wi-Fi provided fluid interaction (up to 14 fps), 3G speeds at the time of the study were insufficient for fluid 3D interaction, though the architecture is designed to scale with improving cellular networks.
Conclusion
The client-server teleradiology system offers a secure, high-performance solution for mobile stroke diagnosis. While limited by its retrospective nature and small reader pool, the results suggest that expert interpretation is not compromised by the mobile medium. Future prospective studies in real-world clinical environments are needed to fully validate its utility in time-critical emergency medicine.
Summary updated