MRTHOD FOR DECOMPOSING AN ELECTROMYOGRAM INTO INDIVIDUAL MOTOR UNIT ACTION POTENTIALS

Patent №

US 4,611,284

Granted

1986-09-09

Filed 1983

Owner

BOARD OF TRUSTEES OF THE LELAND STANFORD JUNIOR UNIVERSITY, THE, STANFORD, CA 94305 A CORP OF

AI components

2

vision · hardware

Assignment

Recorded

Dataset

AIPD

2023_r1 edition

Application

06531209

An automated method of decomposing an electromyogram for identifying and measuring individual motor-unit action potentials. An electromyogram interference pattern is sampled at its Nyquist rate and digitally filtered to accentuate the rapidly rising spike components. The spikes are classified by template matching, using their canonically registered discrete Fourier transforms to align and compare them. The classifications are verified by analysis of interspike intervals. The motor-unit action potentials are then averaged from the unfiltered interference pattern using the identified spikes as triggers.

VisionAI hardwareA61B 5/35A61B 5/7264A61B 5/7203A61B 5/7257G16H 50/20Y10S 128/92

AI classification

Vision1.00
AI hardware0.99
Machine learning0.34
Planning0.01
Natural language0.01
Speech0.01
Knowledge representation0.00
Evolutionary computation0.00

Ownership

BOARD OF TRUSTEES OF THE LELAND STANFORD JUNIOR UNIVERSITY, THE, STANFORD, CA 94305 A CORP OF

assignment · 42700856

Assignors

MCGILL, KEVIN C., DORFMAN, LESLIE J.

On an employer assignment, the assignors are typically the inventors.

© 2026 NYSGPT2525 LLC