Apparatus and method for the detection and treatment of arrhythmias using a neural network

• US 5,251,626 A
• Filed: 07/22/1992
• Issued: 10/12/1993
• Est. Priority Date: 07/03/1990
• Status: Expired due to Fees

• Current Assignee: Pacesetter Incorporated (Abbott Laboratories Incorporated)
• Original Assignee: Telectronics Pacing Systems, Inc (Abbott Laboratories Incorporated)
• Inventors: Dodd, Robert, Drane, Geoff, Flower, Barry, Murphy, Tony, Bassin, David, Lunsmann, Paul, Wickham, John, Nickolls, Peter
• Primary Examiner(s): Kamm, William E.
• Application Number:
  US07/918,934
• Time in Patent Office: 447 Days
• Field of Search: 128/419 D, 128/419 PG
• US Class Current: 607/14
• CPC Class Codes: A61N 1/36514 controlled by a physiological quantity other than heart potential, e.g. blood pressure controlled by two or more physical parameters A61N1/36585
• CPC Class Codes: Y10S 128/925 Neural network

Abstract

An apparatus and method for the detection and treatment of arrhythmias using a processor having a neural network with a hierarchical arrangement including a first lower level for classifying individual waveforms, a second higher level for diagnosing detected arrhythmias and a third higher level for the application of therapy in response to a diagnosed arrhythmia. The neural network may be a back propogation neural network or an associative memory type neural network. The arrhythmias detected may be at least one of bradycardia, tachycardia and fibrillation. The apparatus may include a cardioverting/defibrillating pacemaker. In general, the apparatus acquires physiological signals representative of heart activity in a patient. A neural network receives the physiological signals and determines if any arrhythmia is present, and if present, selects therapy to be applied to the heart. A therapy generator then applies the therapy selected by the neural network. The physiological signals may be processed or unprocessed ECG signal, signals indicative of the properties of the blood including the presence of gases, blood temperature, and blood flow signals or signals representative of ventricular wall impedance or ventricular volume.

Claims

1. An arrhythmia control apparatus, comprising:

means for acquiring physiological signals representative of heart activity of the patient;

means for processing said physiological signals, said processing means including a plurality of neural networks coupled to said acquiring means for receiving the physiological signals, classifying said signals into pathological and non-pathological categories, and selecting therapy to be applied to the heart if a pathological condition is determined to be present in the heart, said plurality of neural networks being arranged in series so that the output of one of said networks comprises the input of another of said networks, one of said networks selecting the therapy to be delivered to the heart in the event a pathological condition is determined to be present in the heart; and

,

therapy means coupled to said processing means for delivering to the heart the therapy selected by said processing means.

2. An arrhythmia control apparatus according to claim 1, wherein one of said networks extracts a series of features from said physiological signals.

3. An arrhythmia control apparatus according to claim 1, wherein one of said networks classifies said physiological signals into said pathological and non-pathological categories.

5. An arrhythmia control apparatus according to claim 1, wherein said apparatus further includes a power supply, sensing means for detecting QRS events of the heart, and a microprocessor coupled to said power supply, said sensing means and said plurality of neural networks for controlling power to said plurality of neural networks, said microprocessor controlling the flow of power to said neural networks such that power is supplied to less than all of said neural networks in the absence of a pathological condition of the heart and to all of said neural networks when a pathological condition is determined to be present in the heart.

4. An arrhythmia control apparatus comprising:

means for acquiring physiological signals representative of hearty activity of the patient;

means for processing said physiological signals, said processing means including a plurality of neural networks coupled to said acquiring means for receiving the physiological signals, classifying said signals into pathological and non-pathological categories, and selecting therapy to be applied to the heart fi a pathological condition is determined to be present in the heart, said plurality of neural networks being arranged in series so that the output of one of said networks comprises the input of another of said networks, a first one of said plurality of neural networks extracting a series of features from said physiological signals, a second one of said plurality of neural networks being serially connected to said first neural network to receive said extracted features and classify said physiological signals into pathological and non-pathological categories based on said features, and a third one of said plurality of neural networks being serially connected to said second neural network and, in the event a pathological condition is determined to be present by said second network, selecting a therapy to be delivered to the heart based on such pathological condition; and

,

therapy means coupled to said processing means for delivering to the heart the therapy selected by said processing means.

6. A method of controlling arrhythmias in a patient'"'"'s heart, comprising the steps of:

acquiring physiological signals representative of heart activity of the patient;

providing a processing means, including a plurality of neural networks, for receiving said physiological signals;

operating said neural networks in such a manner that the output of one of said networks comprises the input of another of said networks;

providing therapy means for delivering one of a plurality of different therapies to the heart;

inputting said physiological signals into one of said neural networks;

processing said physiological signals in said neural networks to detect whether an arrhythmia is present in the patient'"'"'s heart and to select a therapy to be applied to the heart if said processing means determines an arrhythmia to be present;

selecting, in one of said neural networks, the therapy to be delivered to the heart in the event a pathological condition is determined to be present in the heart; and

delivering to the heart the therapy selected by said processing means.

7. A method of controlling arrhythmia according to claim 6, further comprising extracting a series of features from individual senses physiological waveforms in one of said neural networks.

8. A method of controlling arrhythmias in a patient'"'"'s heart, comprising the steps of:

acquiring physiological signals representative of heart activity of the patient;

providing a processing means, including a plurality of neural networks, for receiving said physiological signals;

operating said neural networks in such a manner that the output of one of said networks comprises the input of another of said networks;

providing therapy means for delivering one of a plurality of different therapies to the heart;

inputting said physiological signals into one of said neural networks;

processing said physiological signals in said neural networks to detect whether an arrhythmia is present in the patient'"'"'s heart and to select a therapy to be applied to the heart if said processing means determines an arrhythmia to be present;

said processing step including the sub-steps of extracting a series of features from said physiological signals in a first one of said neural networks, receiving said extracted features and classifying said physiological signals into pathological and non-pathological categories based on said features in a second one of said neural networks, and in the event a pathological condition is determined to be present by said second neural network, selecting a therapy to be delivered to the heart based on such pathological condition; and

,

delivering to the heart the therapy selected by said processing means.

9. A method of controlling arrhythmia according to claim 8, further comprising the steps of providing a power supply, sensing means for detecting QRS events of the heart, and a microprocessor coupled to the power supply, the sensing means and the plurality of neural networks for controlling power to the plurality of neural networks, and

controlling the flow of power to said neural networks by said microprocessor in such a way that power is supplied to less than all of said neural networks in the absence of a pathological condition of the heart and to all of said neural networks when a pathological condition is determined to be present in the heart.

10. A method of controlling arrhythmia according to claim 8, further comprising the step of programming selected parameters utilized by said processing means in detecting arrhythmias.