Methods for Lowering the Power Consumption of OS-Based Adaptive Deep Brain Stimulation Controllers.

Area of Science:

• Neuroscience
• Biomedical Engineering
• Computer Science

Background:

• Adaptive deep brain stimulation (aDBS) requires sophisticated hardware-software systems for clinical application.
• Current operating systems (OS) for aDBS controllers introduce significant power overhead, limiting their use in implantable devices.

Purpose of the Study:

• To develop and evaluate energy-saving techniques for an OS-based adaptive deep brain stimulation (aDBS) controller.
• To demonstrate the feasibility of an energy-efficient OS for implantable aDBS systems.

Main Methods:

• Implemented an energy-efficient OS (YetiOS) on an STM32L476RE microcontroller.
• Developed a dual threshold aDBS algorithm for suppressing pathological neural activity.
• Applied four energy-saving techniques: tick-less idle mode, dynamic sampling, buffered read, and duty cycling.
• Tested the system using a simulated parkinsonian basal ganglia model and local field potentials (LFPs).

Main Results:

• The OS-based controller alone consumed 10.03 mW at 1 kHz sampling rate.
• The proposed energy-saving techniques reduced power consumption to 12 μW.
• The energy-efficient OS facilitated rapid and flexible testing of control methods.

Conclusions:

• The developed energy-saving techniques significantly reduce power consumption for OS-based aDBS controllers.
• This approach enables the development of energy-efficient implantable aDBS devices.
• The OS-based controller can support a wide range of control algorithms for various neurological conditions.