EP0433626A2
Expired / Lapsed
EP
⏱️ 0.0 years remaining
Physiologic waveform analysis
Assignee: Wayne State University
•
Filing: 1990-11-05
•
Granted: —
•
Expires: 2010-11-05
Abstract
Abstract A waveform analysis assembly (10) includes a sensor (12) for detecting physiological electrical and mechanical signals produced by the body. An extraction neural network (22, 22′) will learn a repetitive waveform of the electrical signal, store the waveform in memory (18), extract the waveform from the electrical signal, store the location times of occurrences of the waveform, and subtract the waveform from the electrical signal. Each significantly different waveform in the electrical signal is learned and extracted. A single or multilayer layer neural network (22) accomplishes the learning and extraction with either multiple passes over the electrical signal or accomplishes the learning and extraction of all waveforms in a single pass over the electrical signal. A reducer (20) receives the stored waveforms and times and reduces them into features characterizing the waveforms. A classifier neural network (36) analyzes the features by classifying them through nonlinear mapping techniques within the network representing diseased states and produces results of diseased states based on learned features of the normal and patient groups.
Inventors
- 👤 Robert Spitzer
- 👤 Mohamed Hassoun
Bibliographic Details
Application Number
—
Jurisdiction
EP
Data Source
google_patents
Last Synchronized
2026-10-31 22:13
CPC Classifications
A; A61; A61B; A61B5/00; A61B5/24; A61B5/316; A61B5/389; A61B5/397; A61B5/72; A61B5/7235; A61B5/7264; A61B5/7267; G; G06; G06N; G06N20/00; G06N3/00; G06N3/02; G06N3/04; G06N3/045; G06N3/0455; G06N3/0499; G06N3/08; G06N3/082; G06N3/0895; G06N3/09; G06N3/10; G16; G16H; G16H15/00; G16H50/00; G16H50/20; A61B5/7239; G06F; G06F2218/00; G06F2218/12; G16H40/00; G16H40/60; G16H40/63; Y; Y10; Y10S; Y10S128/00; Y10S128/92; Y10S128/925
Claims Analysis
Total Claims: 30
Claim 1
Independent Claim
137 words
A wave analysis assembly for discriminating physiological signals produced within the body and for producing output indicative of disease characteristics, said assembly comprising; sensing means (12) for sampling a physiological signal produced within the body and for producing an electrical signal, control means (20) for identifying the location of individual unknown waveforms occurring about a peak within said electrical signal and producing a learning set comprised of the individual unknown waveforms of a repetitive waveform, extraction network means (22, 22′) for receiving said learning set and for learning the repetitive unknown waveform within said electrical signal a mapping transformation of said individual waveforms of the learning set in a distributed network to produce a learned waveform from said electrical signal, and output means (42) for diagnosing and outputting information of disease diagnosis based on said extracted waveform.
Claim 2
Dependent Claim
137 words
A wave analysis assembly for discriminating physiological signals produced within the body and for producing output indicative of disease characteristics, said assembly comprising; sensing means (12) for sampling a physiological signal produced within the body and for producing an electrical signal, control means (20) for identifying the location of individual unknown waveforms occurring about a peak within said electrical signal and producing a learning set comprised of the individual unknown waveforms of a repetitive waveform, extraction network means (22, 22′) for receiving said learning set and for learning the repetitive unknown waveform within said electrical signal a mapping transformation of said individual waveforms of the learning set in a distributed network to produce a learned waveform from said electrical signal, and output means (42) for diagnosing and outputting information of disease diagnosis based on said extracted waveform.
Claim 3
Dependent Claim
31 words
An assembly as set forth in claim 1 further characterised said extraction network means (22, 22′) comprising a neural network for developing and storing an idealized representation of the unknown waveforms.
Claim 4
Dependent Claim
79 words
An assembly as set forth in claim 1 further characterised by said network means (22,22′) including weight matrix means for storing an idealized representation of the repetitive waveform by synaptic weights and updating the synaptic weights by an error signal, and learning means (24) for receiving one of said unknown waveforms and an output waveform produced by sensing said unknown waveform through said weight matrix, and comparing said unknown waveform and said output waveform to produce said error signal.
Claim 5
Dependent Claim
25 words
An assembly as set forth in claim 1 further characterised by including reducing means (34) for analyzing said waveform to produce features characterising said waveform.
Claim 6
Dependent Claim
54 words
An assembly as set forth in claim 4 further characterised by including classifier network means (36) containing information comprising standard features and combinations thereof associated with disease states for comparing said features to said information to produce an output representative of diseased states based on the best match of said features with said information.
Claim 7
Dependent Claim
48 words
An assembly as set forth in claim 5 further characterised by said neural network (22,22′) comprising a single layer network for receiving one learning set for a waveform and learning the waveform in a first pass and extracting the waveform in a second pass over the electrical signal.
Claim 8
Dependent Claim
42 words
An assembly as set forth in claim 6 further characterised by said neural network (22′) comprising a multilayer network for receiving the learning sets of all waveforms and for learning and extracting the waveform in a first pass over the electrical signal.
Claim 9
Dependent Claim
88 words
A wave analysis assembly for discriminating physiological signals produced within the body and for producing output indicative of disease characteristics and including means for producing features characterising the physiological signal, said assembly comprising; input means for supplying a physiological signal comprising features, a classifier neural network (36) for storing information in a distributed network of known features and combinations thereof associated with disease and normal patient states and for transforming the features characterising the physiological signal through the distributed network to produce an output representative of diseased states.
Claim 10
Dependent Claim
88 words
A wave analysis assembly for discriminating physiological signals produced within the body and for producing output indicative of disease characteristics and including means for producing features characterising the physiological signal, said assembly comprising; input means for supplying a physiological signal comprising features, a classifier neural network (36) for storing information in a distributed network of known features and combinations thereof associated with disease and normal patient states and for transforming the features characterising the physiological signal through the distributed network to produce an output representative of diseased states.
Claim 11
Dependent Claim
46 words
An assembly as set forth in claim 8 further characterised by said classifier neural network (36) including reference memory means (40) for storing the features and combinations for diagnosis in a weight matrix, said features being nonlinearly mapped within said weight matrix to produce said output.
Claim 12
Dependent Claim
20 words
An assembly as set forth in claim 9 further characterised by including reducing means (36) for analyzing said physiological signal.
Claim 13
Dependent Claim
63 words
An assembly as set forth in claim 10 further characterised by including extraction network means (22,22′) for receiving said physiological signal to identify and learn a repetitive waveform comprising said electrical signal and for extracting said learned waveform, said extraction network (22,22′) including buffer means for storing said extracted waveform and associated times of occurrence of substantially similar waveforms within said electrical signal.
Claim 14
Dependent Claim
115 words
A wave analysis assembly for discriminating electrial signals having a repetitive waveform, said assembly comprising; input means (12) for sampling an electrical signal, control means (20) for identifying the location of individual unknown waveforms occurring about a peak within said electrical signal and producing a learning set comprised of the individual unknown waveforms of one repetitive waveform, extraction network means 922,22′) for receiving the learning set and for learning the repetitive unknown waveform by a mapping transformation of siad individual waveforms of the learning set in a distributed network to produce a learned waveform and for extracting the learned waveform from the electrical signal, amd output means (42) for outputting information of said extracted waveform.
Claim 15
Dependent Claim
115 words
A wave analysis assembly for discriminating electrial signals having a repetitive waveform, said assembly comprising; input means (12) for sampling an electrical signal, control means (20) for identifying the location of individual unknown waveforms occurring about a peak within said electrical signal and producing a learning set comprised of the individual unknown waveforms of one repetitive waveform, extraction network means 922,22′) for receiving the learning set and for learning the repetitive unknown waveform by a mapping transformation of siad individual waveforms of the learning set in a distributed network to produce a learned waveform and for extracting the learned waveform from the electrical signal, amd output means (42) for outputting information of said extracted waveform.
Claim 16
Dependent Claim
98 words
An assembly as set forth in claim 12 further characterised said extraction network means (22) comprising a neural network for developing and storing an idealized representation of the unknown waveforms, said network means (22,22′) including weight matrix means for storing an idealized representation of the repetitive waveform by synaptic weights and updating the synaptic weights by an error signal, and learning means (24) for receiving one of said unknown waveforms and an output waveform produced by sensing said unknown waveform through said weight matrix, and comparing said unknown waveform and said output waveform to produce said error signal.
Claim 17
Independent Claim
110 words
A method of discriminating physiological signals produced within the body and producing output indicative of disease characteristics, said method comprising the steps of: sensing the physiological signals produced within the body and producing an electrical signal, identifying the location of individual unknown waveforms occurring about a peak within the electrical signal, producing a learning set comprised of the individual unknown waveforms of one repetitive waveform within the electrical signal by a mapping transformation of the individual waveforms of the learning set in a distributed network to produce a learned waveform, extracting the learned waveform from the electrical signal, and outputting information of the disease diagnosis based on the extracted waveform.
Claim 18
Independent Claim
110 words
A method of discriminating physiological signals produced within the body and producing output indicative of disease characteristics, said method comprising the steps of: sensing the physiological signals produced within the body and producing an electrical signal, identifying the location of individual unknown waveforms occurring about a peak within the electrical signal, producing a learning set comprised of the individual unknown waveforms of one repetitive waveform within the electrical signal by a mapping transformation of the individual waveforms of the learning set in a distributed network to produce a learned waveform, extracting the learned waveform from the electrical signal, and outputting information of the disease diagnosis based on the extracted waveform.
Claim 19
Dependent Claim
22 words
A method as set forth in claim 14 further characterised by storing an idealized representation of the repetitive waveform by synaptic weights.
Claim 20
Dependent Claim
26 words
A method as set forth in claim 15 further characterised by receiving an unknown waveform and sending same through the weights to produce an output waveform.
Claim 21
Dependent Claim
33 words
A method as set forth in claim 16 further characterised by comparing the unknown waveform to the output waveform to produce an error signal, and updating the synaptic weights by the error signal.
Claim 22
Dependent Claim
18 words
A method as set forth in claim 17 further including reducing the waveform into features characterising the waveform.
Claim 23
Dependent Claim
43 words
A method as set forth in claim 18 further including analyzing the features with respect to information comprising standard features and combinations thereof associated with disease states, and outputting diseased states data based on the best match of the features with the information.
Claim 24
Independent Claim
91 words
A method of discriminating electrical signals having a repetitive waveform, the method including the steps of; sensing an electrical signal, identifying the location of individual unknown waveforms occurring about a peak within the electrical signal, producing a learning set comprised of the individual unknown waveforms assumed to be of one repetitive waveform, learning the unknown repetitive waveform by mapping the individual waveforms of the learning set in a distributed network to produce a learned waveform, extracting the learned waveform from the electrical signal based, and outputting information of the extracted waveform.
Claim 25
Independent Claim
91 words
A method of discriminating electrical signals having a repetitive waveform, the method including the steps of; sensing an electrical signal, identifying the location of individual unknown waveforms occurring about a peak within the electrical signal, producing a learning set comprised of the individual unknown waveforms assumed to be of one repetitive waveform, learning the unknown repetitive waveform by mapping the individual waveforms of the learning set in a distributed network to produce a learned waveform, extracting the learned waveform from the electrical signal based, and outputting information of the extracted waveform.
Claim 26
Dependent Claim
60 words
A method a s set forth in claim 20 further characterized by storing an idealized representation fo the repetitive waveform by synaptic weights, receibing an unknown waveform and sending same through the weights to produce an output waveform, comparing the unknown waveform to the output waveform to produce an error signal, and updating the synaptic weights by the error signal.
Claim 27
Independent Claim
64 words
A method for discriminating physiological signals produced within he body and for producing output indicative of disease characteristics, the method including the steps of; producing a physiological signal comprising features, storing information in a distributed network of known features and combinations thereof associated with disease and normal patient states, transforming the features through the distributed network to produce an output representative of diseased states.
Claim 28
Independent Claim
64 words
A method for discriminating physiological signals produced within he body and for producing output indicative of disease characteristics, the method including the steps of; producing a physiological signal comprising features, storing information in a distributed network of known features and combinations thereof associated with disease and normal patient states, transforming the features through the distributed network to produce an output representative of diseased states.
Claim 29
Dependent Claim
36 words
A method as set forth in claim 22 further characterised by storing the known features and combintions thereof in a weight matrix with the features being nonlinearly mapped within the weight matrix to produce the output.
Claim 30
Dependent Claim
34 words
A method a s set forth in claim 23 further characterised by training the distributed network by updating the weight matrix with addition features and combinations thereof asociated with disease and normal patient states.
References Cited (Backward Citations)
No citation records indexed.
USPTO Annuity & Maintenance Windows
Maintenance schedule is calculated for granted US utility patents.