Surgery – Diagnostic testing – Detecting nuclear – electromagnetic – or ultrasonic radiation
Reexamination Certificate
2001-07-18
2004-01-06
Paik, Sang Y. (Department: 3742)
Surgery
Diagnostic testing
Detecting nuclear, electromagnetic, or ultrasonic radiation
Reexamination Certificate
active
06675036
ABSTRACT:
BACKGROUND OF INVENTION
The present invention relates to noise cancellation, and more particularly, to techniques for canceling noise in a bio-potential signal caused by a patient's respiration.
A bio-potential signal is any electrical signal measured from the human body. An array of sensors (also commonly called transducers) are typically connected to the patient to acquire measurements of such signals. Typical examples of bio-potential signals include an electrocardiogram (ECG), an electroencephalogram (EEG), and an electromyogram (EMG). An ECG (also known as EKG) is a record of the electrical activity of the heart as recorded by an electrocardiograph. An EEG is a record of the electrical activity of the brain as recorded by an electroencepalograph. An EMG is a record of the electrical activity of muscle as recorded by an electromyograph.
When a bio-potential signal is acquired from a patient, the measured signal typically is representative of not only the desired electrical activity, but also additive electrical noise introduced by other systems. Bio-potential signals are intrinsically of low power and hence susceptible to electrical interference. A patient's respiratory system is one example of a system that may introduce additive electrical noise. Respiration causes variability in bio-potential signals primarily due to the changes in the geometry of the chest and tilting of the heart during lung or pulmonary function. Electrical noise associated with the respiration of a patient can overlap or combine with the electrical activity of a bio-potential signal. The resultant acquired signal contains bio-potential signal content that is indistinguishable from the noise content.
When an acquired bio-potential signal includes electrical noise introduced by other systems, the acquired signal is known as a noisy bio-potential signal. Typically, in order for the noisy bio-potential signal to have medical significance it must be filtered to remove the additive electrical noise. A bio-potential signal that has been filtered is known as a clean bio-potential signal. Clean bio-potential signals have varying degrees of cleanliness depending upon what type of filtering is utilized. The type of filtering utilized often depends upon the end use of the clean bio-potential signal and the degree of accuracy required by that end use.
SUMMARY OF INVENTION
Monitoring devices generally have lower accuracy requirements than diagnostic devices. Monitoring devices can therefore use filters that are less expensive and less complex than the filters required on devices with higher accuracy requirements. An ECG monitor is one example of a monitoring device that uses analog and digital filtering technology to output a “clean” bio-potential signal. Although the output signal is adequate for the purposes of monitoring, the signal is largely time delayed and the signal invariably has an appreciable amount of the desired frequency content removed. This type of output is unacceptable for many diagnostic devices.
The diagnostic quality of a bio-potential signal may be reduced depending upon the frequency content removed and the amount of time delay introduced during filtering. Diagnostic applications typically require very accurate, minimally time delayed data. A magnetic resonance imaging (MRI) device is one example of a device that requires very accurate, minimally time delayed bio-potential signals that contain an appreciable amount of the desired frequency content. If the bio-potential signal is largely time delayed, or has an appreciable amount of the desired frequency content removed, the bio-potential signal may not be as useful as a diagnostic tool. The desired frequency content is all, and only all, of the frequency content of the signal of interest. The measured signal includes additional frequency content (i.e., frequency content due to noise) that is removed according to the invention leaving only the desired frequency content. Generally, analog and digital filters remove the additional frequency content and frequency content of the signal of interest, resulting in a signal that has reduced quality as a diagnostic tool.
Cardiac and cardiovascular imaging using MRI techniques is frequently utilized because of the advantages MRI has over other imaging techniques that typically employ radiation, such as X-rays. However, for cardiac studies, the subject is often required to remain within the MRI device for a duration of up to sixty minutes. Frequently, certain bio-potential signals, most notably the ECG signal, are measured during an MRI scan. Measurement of bio-potential signals may be required not only to monitor and diagnose the status of critically ill patients using various types of monitoring and diagnostic medical equipment, but measurement may also be required to synchronize the acquisition of MRI data with certain physiological phenomena such as the beating of the heart when using medical imaging devices. This synchronization is known as triggering. When an ECG signal is utilized, a threshold detector is set to output an actuation signal when the peak of the QRS complex is detected. When a noisy ECG signal is utilized, it is difficult to set the threshold at a level that will output an actuation signal for all (but only all) peaks of the QRS complexes. If the threshold is set too high it may cause peaks of the QRS complexes to be missed, if the threshold is set too low other parts of the ECG signal will be detected by the threshold detector and a false trigger may result. Another problem associated with attempting to trigger off a noisy ECG signal is the introduction of time-wise inaccuracies, or jitter, in the detection of the peaks of the QRS complexes.
Accordingly, the invention provides a diagnostic medical imaging device including an adaptive filter noise canceler to reduce the electrical noise associated with the patient's respiration and the like in low powered bio-potential signals. The adaptive filter noise canceler includes an adaptive filter unit having a noise reference input for receiving a noise reference signal associated with the patient's respiration, a clean bio-potential input for receiving a clean bio-potential signal, and a filtered noise reference output for outputting a filtered noise reference signal. The adaptive filter noise canceler also includes a summing node having a noisy bio-potential input for receiving a noisy bio-potential signal, a filtered noise reference input for receiving the filtered noise reference signal, and a clean bio-potential output for outputting the clean bio-potential signal. The adaptive filter noise canceler also includes a feed-back loop that electrically couples the clean bio-potential output and the clean bio-potential input, and a feed-forward loop that electrically couples the filtered noise reference output and the filtered noise reference input.
In one embodiment of the invention, the noise reference signal is produced from the electrical signal generated by a bellows transducer connected to the patient's chest. The electrical signal corresponds to the extension or deflection of the patient's chest and therefore is associated with the patients respiration.
In another embodiment of the invention, the noise reference signal is produced from the electrical signal generated by an array of sensors connected to the patient using a number of generally known methods to measure the electrical noise caused by the patient's respiration (e.g., transthoracic impedance). The electrical noised is caused by the patient's respiration and is therefore inherently associated with the patients respiration.
It is an advantage of the invention to permit the continuous acquisition of low power bio-potential signals in the presence of varying values of electrical noise associated with the patient's respiration. The subtraction of a filtered noise reference signal from the noisy bio-potential signal effectively reduces the additive electrical noise associated with the patient's respiration and produces the clea
Kreger Kevin Scott
Pettersson Bo Johan
GE Medical Systems, Inc.
Michael & Best & Friedrich LLP
Paik Sang Y.
Robinson Daniel
LandOfFree
Diagnostic device including a method and apparatus for... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Diagnostic device including a method and apparatus for..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Diagnostic device including a method and apparatus for... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3214180