We have developed a client/server application for real time intraoperative neurophysiological monitoring of brain and spinal cord function. The following PVM services are used:
When an operative procedure is scheduled for monitoring, an intrumentation cart including Unix workstation is wheeled into the operating room, attached to the Ethernet running through the hospitals, and booted. There is a single instance of a cshell polling process which enables dependable inclusion in the virtual machine of hosts whose connections to the network are transient. This process utilizes approximately 3% of the CPU cycles of the machine on which it runs. It performs the following functions:
For each procedure, data is collected continuously, processed, and saved to a file. Any observer process (client) connected to the virtual machine can access the data through the server process running on the machine on which the data is being acquired. The virtual machine includes 50 hosts located at the University of Pittsburgh, George Washington University (Washington, D.C.), and Mount Sinai Hospital (New York City). The version of this application built on top of PVM required 2 man months for development, has proved to be more fault tolerant than previous versions, and has been used to monitor 500 surgical procedures to date.
Krieger D, G Burk, RJ Sclabassi. "Neuronet: A distributed real-time system for monitoring neurophysiologic function in the medical environment." IEEE Computer 24(3): 45-55, 3/91.
Simon, R., Krieger, D., Znati, T., Lofink, R., and Sclabassi, R.J., ``MultiMedia MedNet: A medical collaboration and consultation system", IEEE Computer 28(5): 65-73, 1995.