US5640966A
Expired / Lapsed
US
⏱️ 0.0 years remaining
Medical apparatus for analyzing electrical signals from a patient
Assignee: Siemens Elema AB
•
Filing: 1995-11-06
•
Granted: 1997-06-24
•
Expires: 2015-11-06
Abstract
Abstract An analysis apparatus such as an ECG apparatus has a control unit which includes an artificial neural network for discovering signal-recording electrodes which are erroneously attached to a patient. At least one artificial neural network is taught by being fed measurement signals from both correctly recorded measurements and from erroneously recorded measurements. The artificial neural network is then able to identify erroneous attachments with great accuracy from recorded measurement signals.
Inventors
- 👤 Bo Heden
- 👤 Mattias Ohlsson
- 👤 Lars Edenbrandt
- 👤 Ralf Rittner
- 👤 Olle Pahlm
- 👤 Carsten Peterson
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-11-09 13:44
CPC Classifications
A; A61; A61B; A61B5/00; A61B5/24; A61B5/25; A61B5/276; A61B5/316; A61B5/318; A61B5/346; A61B5/349; A61B5/72; A61B5/7221; A61B5/7235; A61B5/7264; G; G16; G16H; G16H50/00; G16H50/20
Claims Analysis
Total Claims: 23
Claim 1
Independent Claim
69 words
A medical apparatus for analyzing electrical signals from a patient comprising: a plurality of electrodes respectively adapted for placement and attachment at defined locations on a patient to pick up electrical signals from the patient; and artificial neural network means, connected to each of said electrodes, for determining whether any of said electrodes is placed at a location on the patient differing from the defined location for that electrode.
Claim 2
Dependent Claim
69 words
A medical apparatus for analyzing electrical signals from a patient comprising: a plurality of electrodes respectively adapted for placement and attachment at defined locations on a patient to pick up electrical signals from the patient; and artificial neural network means, connected to each of said electrodes, for determining whether any of said electrodes is placed at a location on the patient differing from the defined location for that electrode.
Claim 3
Dependent Claim
30 words
A medical apparatus as claimed in claim 1 wherein said artificial neural network means additionally comprises means for determining whether any of said electrodes is incorrectly attached to said patient.
Claim 4
Dependent Claim
41 words
A medical apparatus as claimed in claim 1 wherein said plurality of electrodes comprise ECG electrodes for sensing electrical signals arising from cardiac activity of the patient, and said apparatus further comprising means for generating an electrocardiogram from said electrical signals.
Claim 5
Dependent Claim
75 words
A medical apparatus as claimed in claim 3 wherein said artificial neural network means comprises an input layer having a plurality of input neurons, a hidden layer having a plurality of hidden neurons, and an output layer having an output neuron, and wherein said input layer, said hidden layer and said output layer, in combination, comprise means teachable using a back-propagation algorithm with previously recorded correct and erroneous electrical signals obtained from said ECG electrodes.
Claim 6
Dependent Claim
22 words
A medical apparatus as claimed in claim 4 wherein said input layer comprises an input layer containing up to 25 input neurons.
Claim 7
Dependent Claim
32 words
A medical apparatus as claimed in claim 4 wherein said hidden layer comprises a hidden layer having said plurality of hidden neurons in a range of from 4 to 6 hidden neurons.
Claim 8
Dependent Claim
20 words
A medical apparatus as claimed in claim 4 wherein said output layer comprises an output layer having one output neuron.
Claim 9
Dependent Claim
50 words
A medical apparatus as claimed in claim 4 wherein said input layer, said hidden layer and said output layer comprise, in combination, means teachable with harder punishment for designating correctly placed ECG electrodes as incorrectly placed ECG electrodes than for designating incorrectly placed ECG electrodes as correctly placed ECG electrodes.
Claim 10
Dependent Claim
71 words
A medical apparatus as claimed in claim 4 wherein said plurality of ECG electrodes include a first ECG electrode adapted for placement on the right arm of the patient and a second ECG electrode adapted for placement on the left arm of the patient, and wherein said artificial neural network means comprises means for determining whether the respective locations of said first and second ECG electrodes on said patient are reversed.
Claim 11
Dependent Claim
71 words
A medical apparatus as claimed in claim 4 wherein said plurality of ECG electrodes include a first ECG electrode adapted for placement on the left leg of the patient and a second ECG electrode adapted for placement on the left arm of the patient, and wherein said artificial neural network means comprises means for determining whether the respective locations of said first and second ECG electrodes on said patient are reversed.
Claim 12
Dependent Claim
63 words
A medical apparatus as claimed in claim 1 wherein said artificial neural network means comprises means for identifying a first specific incorrect location of any of said electrodes, and said medical apparatus further comprising further artificial neural network means for identifying a second specific incorrect location of any of said electrodes, said first and second specific incorrect locations of said electrodes being different.
Claim 13
Dependent Claim
46 words
A medical apparatus as claimed in claim 1 wherein said plurality of electrodes comprise a plurality of EEG electrodes adapted to sense electrical signals arising from brain activity of the patient, and said medical apparatus further comprising means for producing an electroencephalogram from said electrical signals.
Claim 14
Independent Claim
85 words
A method for analyzing electrical signals obtained from a patient comprising the steps of: respectively placing a plurality of electrodes at defined locations on a patient; picking up electrical signals from the patient via said plurality of electrodes; and supplying said electrical signals from said electrodes to an artificial neural network; and analyzing said electrical signals in said artificial neural network and thereby determining whether any of said electrodes is placed on said patient at a location other than the defined location for that electrode.
Claim 15
Independent Claim
85 words
A method for analyzing electrical signals obtained from a patient comprising the steps of: respectively placing a plurality of electrodes at defined locations on a patient; picking up electrical signals from the patient via said plurality of electrodes; and supplying said electrical signals from said electrodes to an artificial neural network; and analyzing said electrical signals in said artificial neural network and thereby determining whether any of said electrodes is placed on said patient at a location other than the defined location for that electrode.
Claim 16
Dependent Claim
34 words
A method as claimed in claim 13 comprising the additional step of: analyzing said electrical signals in said artificial neural network to determine whether any of said electrodes is incorrectly attached to said patient.
Claim 17
Dependent Claim
70 words
A method as claimed in claim 13 wherein the step of respectively placing said plurality of electrodes at defined locations on a patient comprises respectively placing said plurality of electrodes at defined locations on a patient for sensing electrical signals arising from cardiac activity of said patient and thereby generating ECG measurement signals, and wherein said method includes the additional step of producing an electrocardiogram from said ECG measurement signals.
Claim 18
Dependent Claim
30 words
A method as claimed in claim 15 comprising the additional step of teaching said artificial neural network with a back-propagation algorithm using previously recorded correct and erroneous ECG measurement signals.
Claim 19
Dependent Claim
55 words
A method as claimed in claim 16 wherein the step of teaching said artificial neural network comprises teaching said artificial neural network with a back-propagation algorithm having an asymmetrical error function for punishing a designation of correctly connected electrodes as being incorrectly connected harder than a designation of incorrectly connected electrodes as being correctly connected.
Claim 20
Dependent Claim
83 words
A method as claimed in claim 15 wherein the step of respectively placing said electrodes at defined locations on a patient includes placing a first of said electrodes on a right arm of the patient and placing a second of said electrodes on a left arm of the patient, and wherein the step of analyzing said electrical signals in said artificial neural network includes identifying whether the respective locations of said first of said electrodes and said second of said electrodes are reversed.
Claim 21
Dependent Claim
83 words
A method as claimed in claim 15 wherein the step of respectively placing said electrodes at defined locations on a patient includes placing a first of said electrodes on a left leg of the patient and placing a second of said electrodes on a left arm of the patient, and wherein the step of analyzing said electrical signals in said artificial neural network includes identifying whether the respective locations of said first of said electrodes and said second of said electrodes are reversed.
Claim 22
Dependent Claim
78 words
A method as claimed in claim 13 wherein the step of analyzing said electrical signals in said artificial neural network comprises analyzing said electrical signals in said artificial neural network to identify whether a first specific incorrect location of said electrodes exists, and said method comprising the additional steps of supplying said electrical signals to a further artificial neural network, and identifying, in said further artificial neural network, whether a second specific incorrect location of said electrodes exists.
Claim 23
Dependent Claim
63 words
A method as claimed in claim 13 wherein the step of respectively placing said plurality of electrodes at defined locations on a patient comprises respectively placing said plurality of electrodes at defined locations on a patient to sense electrical signals arising from brain activity of said patient, and wherein said method comprises the additional step of producing an electroencephalogram from said electrical signals.
References Cited (Backward Citations)
US5251626A
Cited
US5280792A
Cited
USPTO Annuity & Maintenance Windows
| Annuity Stage | Window Opens | Payment Deadline | Surcharge Period Until | Estimated Fee | Status |
|---|---|---|---|---|---|
| 3.5 Years (Year 4) | 2000-06-23 | 2000-12-23 | 2001-06-24 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 2004-06-23 | 2004-12-23 | 2005-06-24 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 2008-06-23 | 2008-12-23 | 2009-06-24 | $7700.00 | Window Closed / Lapsed |