Adaptive logic networks in rehabilitation of persons with incomplete spinal cord injury
- Univ. of Alberta, Edmonton (Canada)
Persons with incomplete spinal cord injury are generally at least partially paralyzed and are often unable to walk. Manually-controlled electrical stimulation has been used to act upon nerves or muscles to cause leg movement so such persons can achieve functional walking. They use crutches or a mobile walker for support, and initiate each stimulus by pressing a button. Artificial intelligence and machine learning techniques are now making it possible to automate the process of stimulus-initiation. Supervised training of an automatic system can be based on samples of correct stimulation given by the patient or by a therapist, accompanied by data from sensors indicating the state of the person`s body and its relation to the ground during walking. A major issue is generalization, i.e. whether the result of training can be used for control at a later time or in somewhat different circumstances. As the possibilities grow for increasing the number and variety of sensors on a patient, and for easily implanting more numerous stimulation channels, the need is increasing for powerful learning systems which can automatically develop effective and safe control algorithms. This paper explains the foundations of adaptive logic networks, and illustrates how they have been used to develop an experimental walking prosthesis used in a laboratory setting. Successful generalization has been observed using parameters from training which took place minutes to days earlier.
- Research Organization:
- Pacific Northwest National Lab. (PNNL), Richland, WA (United States)
- OSTI ID:
- 377061
- Report Number(s):
- PNL-SA-26375; CONF-9503142-; ON: DE96009360; TRN: 96:003982-0018
- Resource Relation:
- Conference: Workshop on environmental and energy applications of neural networks conference, Richland, WA (United States), 30-31 Mar 1995; Other Information: PBD: [1995]; Related Information: Is Part Of Applications of neural networks in environmental and energy sciences and engineering. Proceedings of the 1995 workshop on environmental and energy applications of neural networks; Hashem, S.; Keller, P.E.; Kouzes, R.T.; Kangas, L.J.; PB: 193 p.
- Country of Publication:
- United States
- Language:
- English
Similar Records
Effectiveness of radiation therapy without surgery in metastatic spinal cord compression: Final results from a prospective trial
Spinal reirradiation after short-course RT for metastatic spinal cord compression