Register now After registration you will be able to apply for this opportunity online.
This opportunity is not published. No applications will be accepted.
Non-invasive AI-based neuroprosthetic system for diabetic peripheral neuropathy towards at-home use.
This project aims to develop and test a closed-loop neuroprosthetic for a non-invasive sensory system that employs transcutaneous electrical stimulation (TENS) for sensory feedback restoration in individuals with diabetic peripheral neuropathy (DPN). The system will encompass a AI-calibration algorithm for the calibration of neurostimulation parameters and the optimization during walking. This algorithm will be able to identify the optimal combination of electrodes and stimulation parameter to deliver a somatotopic and natural sensory feedback. The system will require a user friendly interface for calibration and for real time feedback and functional tasks. The system will be tested in a long term trial encompassing functional tasks and fMRI experiment to evaluated the motor response.
Sensory loss as a result of amputation or DPN breaks the sensory-motor loop resulting in reduced balance, higher gait asymmetry, and a higher risk of falling [1], to name a few prominent conditions. Existing neuroprostheses take advantage of both invasive [2] and non-invasive [3], [4] feedback technology in order to restore sensory function and complete the broken sensory-motor loop in amputees. Peripheral nerve stimulation in individuals with DPN has shown promise in its improvement of balance and reduction of falls[5], [6] suggesting a need for a portable system for daily use.
As shown in the figure above, the current system encodes information from a sensorized insole as well as knee angle information from IMUs as stimulation parameters. Then, TENS stimulation is applied to the peripheral nerves of the user. The overall goal of this project is to prepare this system for independent use at home.
The system is fully portable and reliable once properly set up and calibrated, and it has already shown promising results in diabetic neuropathy individuals. However, the choice of the stimulation parameters (location of electrodes, amplitude, charge and frequency) to obtain a somatotopic and natural sensation is a difficult step. Indeed, the calibration is time consuming and experimenter-dependent, and it is not ready for naïve use.
Sensory loss as a result of amputation or DPN breaks the sensory-motor loop resulting in reduced balance, higher gait asymmetry, and a higher risk of falling [1], to name a few prominent conditions. Existing neuroprostheses take advantage of both invasive [2] and non-invasive [3], [4] feedback technology in order to restore sensory function and complete the broken sensory-motor loop in amputees. Peripheral nerve stimulation in individuals with DPN has shown promise in its improvement of balance and reduction of falls[5], [6] suggesting a need for a portable system for daily use. As shown in the figure above, the current system encodes information from a sensorized insole as well as knee angle information from IMUs as stimulation parameters. Then, TENS stimulation is applied to the peripheral nerves of the user. The overall goal of this project is to prepare this system for independent use at home. The system is fully portable and reliable once properly set up and calibrated, and it has already shown promising results in diabetic neuropathy individuals. However, the choice of the stimulation parameters (location of electrodes, amplitude, charge and frequency) to obtain a somatotopic and natural sensation is a difficult step. Indeed, the calibration is time consuming and experimenter-dependent, and it is not ready for naïve use.
The student will be guided in understanding the principal cause of diabetic neuropathy, its effects and meaning in terms of reduction of quality of life. They will also be introduced to state of the art of neuroprostheses and AI with scientific literature readings, and a preliminary introduction about our developed sensory-feedback system.
The major goals for the student will be:
1. Test the calibration and the full system on healthy volunteers and diabetic patients.
2. Optimized a closed-loop AI-based calibration algorithm for optimal calibration. This algorithm will be able to identify the optimal combination of electrodes and stimulation parameter to deliver a somatotopic and natural sensory feedback in individuals with diabetic peripheral neuropathy (DPN). The AI algorithm will allow to have a calibration phase that is:
- Automatic
- Accurate
- Fast
- Adaptive
- Versatile
- Standard
3. Design an interactive GUI to perform the AI-calibration that can be easily used by naïve users.
The student will be guided in understanding the principal cause of diabetic neuropathy, its effects and meaning in terms of reduction of quality of life. They will also be introduced to state of the art of neuroprostheses and AI with scientific literature readings, and a preliminary introduction about our developed sensory-feedback system. The major goals for the student will be: 1. Test the calibration and the full system on healthy volunteers and diabetic patients. 2. Optimized a closed-loop AI-based calibration algorithm for optimal calibration. This algorithm will be able to identify the optimal combination of electrodes and stimulation parameter to deliver a somatotopic and natural sensory feedback in individuals with diabetic peripheral neuropathy (DPN). The AI algorithm will allow to have a calibration phase that is: - Automatic - Accurate - Fast - Adaptive - Versatile - Standard 3. Design an interactive GUI to perform the AI-calibration that can be easily used by naïve users.
Noemi Gozzi, PhD Student at the Neuroengineering laboratory, Email: noemi.gozzi@hest.ethz.ch
Lauren Chee, PhD Student at the Neuroengineering laboratory, Email: lauren.chee@hest.ethz.ch
Dr. Stanisa Raspopovic, Assistant Professor Neuroengineering laboratory, Head ETH Zurich, Switzerland Email: stanisa.raspopovic@hest.ethz.ch
Noemi Gozzi, PhD Student at the Neuroengineering laboratory, Email: noemi.gozzi@hest.ethz.ch Lauren Chee, PhD Student at the Neuroengineering laboratory, Email: lauren.chee@hest.ethz.ch
Dr. Stanisa Raspopovic, Assistant Professor Neuroengineering laboratory, Head ETH Zurich, Switzerland Email: stanisa.raspopovic@hest.ethz.ch