초록 ▼

Pattern classification techniques are provided for use with an implantable medical device for detecting cardiac ischemia substantially in real-time. Values representative of morphological features of electrical cardiac signals are detected by the implantable medical device. Then, a determination is

Pattern classification techniques are provided for use with an implantable medical device for detecting cardiac ischemia substantially in real-time. Values representative of morphological features of electrical cardiac signals are detected by the implantable medical device. Then, a determination is made as to whether the patient is subject to an on-going episode of cardiac ischemia by applying the values to a pattern classifier configured to identify patterns representative of cardiac ischemia. In one example, the determination is made substantially in real-time by the device itself based on the IEGM signals it detects. In other examples, the IEGM signals are relayed promptly to a bedside monitor or other external device, which analyzes the signals using the pattern classifier to detect ischemia. Therapy may be applied in response to cardiac ischemia. For example, if the implanted device is equipped with a drug pump, appropriate medications may be administered such as anti-thrombolytics.

대표청구항 ▼

What is claimed is: 1. A method for use with an implantable medical device for detecting cardiac ischemia in a patient in which the device is implanted, comprising: detecting a plurality of values representative of different morphological features of electrical cardiac signals, including ST segment

What is claimed is: 1. A method for use with an implantable medical device for detecting cardiac ischemia in a patient in which the device is implanted, comprising: detecting a plurality of values representative of different morphological features of electrical cardiac signals, including ST segment parameters and one or both of atrial depolarization morphological parameters and ventricular depolarization morphological parameters; and detecting whether the patient is subject to an on-going episode of cardiac ischemia by applying the plurality of values to a pattern classifier configured to identify patterns representative of cardiac ischemia. 2. The method of claim 1 wherein the atrial depolarization morphological parameters applied to the pattern classifier include one or more of atrial depolarization event duration, atrial depolarization event slope, and atrial depolarization event amplitude. 3. The method of claim 1 wherein the ventricular depolarization morphological parameters applied to the pattern classifier include one or more of ventricular depolarization event duration, ventricular depolarization event slope, and ventricular depolarization event amplitude. 4. The method of claim 1 wherein the pattern classifier is a linear discriminant-based pattern classifier and wherein detecting whether the patient is subject to an on-going episode of cardiac ischemia comprises: applying the values to the linear discriminant-based pattern classifier to generate a classification metric; comparing the classification metric to a threshold representative of cardiac ischemia; and generating a signal indicative of the detection of cardiac ischemia if the classification metric exceeds the threshold. 5. The method of claim 1 wherein the pattern classifier is a classification tree-based pattern classifier and wherein detecting whether the patient is subject to an on-going episode of cardiac ischemia comprises: applying the values to the classification tree to classify the values as either being representative of cardiac ischemia or not being representative of cardiac ischemia; and generating a signal indicative of the detection of cardiac ischemia if the classification tree classifies the values as being representative of cardiac ischemia. 6. The method of claim 1 wherein detecting whether the patient is subject to an on-going episode of cardiac ischemia is performed substantially in real-time by applying values representative of morphological features of a latest heartbeat of the patient to the pattern classifier. 7. The method of claim 1 wherein detecting whether the patient is subject to an on-going episode of cardiac ischemia is performed periodically by applying values representative of morphological features of a plurality of recent heartbeats of the patient to the pattern classifier. 8. The method of claim 1 further comprising training the pattern classifier to recognize cardiac ischemia based on morphological features of electrical cardiac signals associated with cardiac ischemia. 9. The method of claim 1 further comprising delivering therapy in response to the detection of an episode of cardiac ischemia. 10. The method of claim 9 wherein an implantable drug pump is provided and wherein the step of delivering therapy in response to the detection of an episode of cardiac ischemia includes the step of delivering anti-thrombolytic medications to the patient using the drug pump. 11. The method of claim 9 wherein an implantable cardiac stimulation device is provided and wherein the step of delivering therapy in response to the detection of an episode of cardiac ischemia includes the step of delivering pacing therapy at a reduced pacing rate. 12. The method of claim 1 further comprising generating a warning signal in response to detection of an episode of cardiac ischemia. 13. The method of claim 1 wherein the parameters detected and applied to the pattern classifier include an atrioventricular (AV) interval. 14. The method of claim 1 wherein the implantable medical device includes an implantable cardiac defibrillation device provided with defibrillation shock capacitors and wherein the capacitors are charged in response to the detection of an episode of cardiac ischemia. 15. A system for use with an implantable medical device for detecting cardiac ischemia in a patient in which the device is implanted, comprising: a detection unit operative to detect a plurality of values representative of different morphological features of electrical cardiac signals, including ST segment parameters and one or both of atrial depolarization morphological parameters and ventricular depolarization morphological parameters; and a pattern recognition-based cardiac ischemia detection unit operative to detect whether the patient is subject to an on-going episode of cardiac ischemia by applying the plurality of values to a pattern classifier configured to identify patterns representative of cardiac ischemia. 16. The system of claim 15 wherein the pattern recognition-based cardiac ischemia detection unit is operative to analyze one or more of: atrial depolarization event duration, atrial depolarization event slope, and atrial depolarization event amplitude. 17. The system of claim 15 wherein the pattern recognition-based cardiac ischemia detection unit is operative to analyze one or more of: ventricular depolarization event duration, ventricular depolarization event slope, and ventricular depolarization event amplitude. 18. A system for use with an implantable medical device for detecting cardiac ischemia in a patient in which the device is implanted, comprising: means for detecting a plurality of values representative of different morphological features of electrical cardiac signals, including ST segment parameters and one or both of atrial depolarization morphological parameters and ventricular depolarization morphological parameters; means for detecting whether the patient is subject to an on-going episode of cardiac ischemia by applying the plurality of values to a pattern classifier configured to identify patterns representative of cardiac ischemia; and means for delivering therapy in response to the detection of an episode of cardiac ischemia. 19. The system of claim 18 wherein the means for detecting whether the patient is subject to an on-going episode of cardiac ischemia comprises means for analyzing one or more of atrial depolarization event duration, atrial depolarization event slope, and atrial depolarization event amplitude in conjunction with ST segment parameters. 20. The system of claim 18 wherein the means for detecting whether the patient is subject to an on-going episode of cardiac ischemia comprises means for analyzing one or more of ventricular depolarization event duration, ventricular depolarization event slope, and ventricular depolarization event amplitude in conjunction with ST segment parameters.