Real-Time processing of DVI signals on a FPGA as a telemedicine modular solution

Abstract

Nowadays, telemedicine networks are implemented with limited compatibility regarding data and signal transmission. This hinders their implementation as each medical-device manufacturer has its own set of transmission systems and methods, frequently unable to talk among themselves. On the other hand, medium tier and higher class medical devices have digital or/and analog video outputs (DVI,HDMI,VGA), which are commonly used for improving visibility of the device on medical institutions by using larger screens. Said devices usually lack of any other form of data output, and those who have it follow their own protocol of medical data transmission (only comprehensible by other devices of the same family, or manufacturer), thus making the use of video signals for the establishment of a telemedicine network a feasible option. An earlier implementation of a video-processing architecture using the Nios II softcore processor, along with custom dedicated processing modules for scaling, color conversion, and compression (M-JPEG) for a 22kLE FPGA has been developed to allow video signals from VSMs with DVI output to be transmitted by WWAN, taking into account the bandwidth limitations of each transmission channel.