New research is exploring how artificial intelligence could help make deep brain stimulation (DBS) programming more precise and efficient. Researchers compared several machine-learning models to see how accurately they could predict which nerve fibers would be activated by different DBS electrode configurations, including standard and directional leads.
The goal is to make these predictions faster without sacrificing accuracy. The findings showed that artificial neural network and convolutional neural network models could accurately predict nerve fiber activation across the configurations studied, with CNNs performing particularly well for the more complex electrical fields created by directional leads.
This work could help advance more personalized approaches to DBS programming by making it easier to evaluate potential settings and understand how they may affect different brain circuits.
Researchers from the University of Florida, Case Western Reserve University and the Malcom Rand...