PROCESSING OF DATA RELATING TO MEASUREMENT OF
ELECTRICAL SIGNALS EMANATING FROM A SUBJECT'S BODY
Cross-Reference to Related Applications
The present application claims priority from Australian provisional patent application no. 2010903648 filed 13 August 2010, the contents of which are incorporated herein by reference in their entirety.
Field
This disclosure relates, generally, to the processing of data and, more particularly, to a system for, and a method of, processing of data relating to measurement of electrical signals emanating from a subject's body.
Background
Electrocardiography (ECG) is a technique which senses the electrical discharge of the heart's neuromuscular system by measuring the electrical potential induced by it at the surface of the body. The potential is conventionally sensed using electrodes which are attached at predetermined points on the body. For humans, the standard lead/electrode arrangements are either limb electrodes (so-called Left Arm, Left Leg, and Right Leg attachment points) or the 12-lead arrangement, in which an additional set of leads/electrodes are placed across the sternum and under the left arm. These electrodes are typically stuck on using an adhesive system of some kind and a conductive gel is used to improve the contact impedance of the electrode-to-skin connection.
Similarly, electroencephalography (EEG) is a technique for recording electrical activity resulting from the firing of neurons within the brain using multiple electrodes placed on the scalp.
Recent advances in electronics have made possible the effective use of dry electrodes, which are defined as solid conductive electrodes in which no conductive gel is used. The advantages of dry electrodes include reduced preparation of the contact surface and the ability to move the electrode locally without having to shave the skin, apply gel, re-glue, or to use a new disposable adhesive electrode.
Summary
In an aspect, there is provided a system for processing data relating to measurement of electrical signals emanating from a subject's body, the system including a plurality of sensors, arranged in a fixed relationship relative to one another, to be placed on, or in proximity to, the subject's body for obtaining a sample of data;
a signal conditioning module in communication with the sensors for conditioning data measured by the sensors and for producing output signals;
a filtering module for receiving and filtering the output signals from the signal conditioning module; and
a signal processing module in communication with the filtering module for processing the filtered signals to produce a two-component vector for further data analysis.
By "fixed" is meant a set of electrodes which always have the same geometric position relative to one another and, therefore, form a fixed pattern of contact on the surface of the body, in use. There may be a measure of compliant mounting to ensure a good contact with the body.
The sensors may be arranged in a fixed array.
The sensors may be electrodes. The electrodes may be arranged in groups, for example, differentially arranged pairs irrespective of the distance between the electrodes of each pair.
The signal conditioning module may include a sampling unit which samples the data from each sensor or groups of sensors simultaneously.
The signal conditioning module may include single-ended signal amplifiers connected to each electrode to enhance signal quality. This may be useful when the physical and electrical contact between the electrode and the subject is not guaranteed to remain constant, for example, when measurements are performed on a moving or non-compliant subject, either human or animal.
The filtering module may comprise at least one of a bandpass filter and a custom filter, the custom filter being selected based on underlying bioelectric data.
The signal processing module may be operative to process multiple channels simultaneously in order to calculate the two-component vector representative of that sample of data for subsequent data analysis.
In a second aspect, there is provided a method of processing data relating to measurement of electrical signals emanating from a subject's body, the method including
providing a plurality of sensors arranged in a fixed relationship relative to one another and placing the sensors on, or in proximity to, the subject's body for obtaining a sample of data;
conditioning the data measured by the sensors to produce output signals;
filtering the output signals; and processing the filtered signals to produce a two-component vector for further data analysis.
The method may include sampling the data from each sensor simultaneously. The method may include filtering the output signals using at least one of a bandpass filter and a custom filter, the custom filter being selected based on underlying bioelectric data.
The method may include processing multiple channels simultaneously in order to calculate the two-component vector representative of that sample of data for subsequent data analysis.
Brief Description of Drawings
In the drawings,
Fig. 1 shows a schematic front view of a sensor layout of an embodiment of a system for processing data relating to measurement of electrical signals emanating from a subject's body;
Fig. 2 shows a block diagram of an embodiment of a system for processing data relating to measurement of electrical signals emanating from a subject's body; and
Fig. 3 shows a flow chart of an embodiment of a method of processing data relating to measurement of electrical signals emanating from a subject's body.
Detailed Description of Exemplary Embodiment
In Fig. 2 of the drawings, reference numeral 10 generally designates an embodiment of a system for processing data relating to measurement of electrical signals emanating from a subject's body. These signals may be signals from a patient's heart which are monitored by electrocardiography (ECG) or from a subject's brain which is monitored by electroencephalography (EEG). In particular, with ECGs the system 10 can be used for human or animal subjects. For ease of explanation, the system 10 and the related method will be described with reference to its application to ECGs. It will, however, be appreciated that the system 10 is equally applicable for EEGs and any other form of system which measures electrical signals emanating from a subject's body.
The system 10 comprises a fixed array of sensors in a holder 12. The sensors are in the form of electrodes 14. As defined elsewhere, "fixed" means that the electrodes 14 have the same geometric position relative to one another and form a fixed pattern of contact on the surface of the subject's body, in use. The electrodes 14 may be mounted resiliently in the holder 12 so there is a measure of compliance out of the plane in which the electrodes 14 are arranged. A sample of an electrode array is shown at 16 in Fig. 1 of the drawings. As illustrated, the array 16 consists of four pairs of differentially-connected electrodes 14 labeled A+ - A.; B+ - B.; C+ - C; and D+ - D.. The electrodes 14 are arranged about a circle of a predetermined diameter to form four pairs of differentially-connected electrodes 16 with electrodes of each pair being arranged diametrically opposite each other. The fixed arrangements of the electrodes 16 in this configuration means that an electric field vector, illustrated at 18, can be calculated by summing vector component contributions from each electrode 14 or each pair of differentially-connected electrodes 14.
For the illustrated array 16, the projection of any electrical field, as a result of electric signals generated by the subject's body imposed on the array 16 can be calculated in terms of in-plane axes x and y with origin at the centre of the array 16 as: -
Where
ex = (C* - C. ) + - 0_)cosQ + +(CL - D.) os (~)
and
The above equations apply to the illustrated array 16. A similar formula may be derived for other shapes or fixed arrays such as electrodes placed at the intersections of a regular grid or electrodes randomly placed but fixed in position relative to one another.
The benefit of an array such as that illustrated at 16 of Fig. 1 of the drawings is that it can be applied at any place on the subject's body and a vector projection 18 of the body's internal electrical field arising from the body's electric signals at that point can be calculated without requiring prior knowledge of the location of the electrodes 14 on the body.
This also has the effect that a given array 16 can be used in different areas of the body, for example, to measure the maternal ECG of a pregnant female, be it human or animal, by placing the electrode array 16 in close proximity to the maternal heart and then to measure the foetal ECG by placing the electrode array in close proximity to the womb.
Output data from the holder 12 is fed to a signal conditioning module 20 of the system 10. As shown at step 22 in Fig. 3 of the drawings, the system 10 samples the data output from the holder 12 using differential amplification of signals from the fixed electrode array 16. The channels are sampled simultaneously at a fixed sample rate of an adequate resolution and Nyquist frequency. For example, for most ECG systems, a 16-bit sampling at 1000 samples/s is adequate.
An output signal from the signal conditioning module 20 is fed to a filtering module 24. As shown at step 26 in Fig. 3 of the drawings, in the filtering module, the output signal is amplified and filtered for each channel. The filtering module 24 either comprises a bandpass filter which uses standard methods within a band from approximately 2Hz to 70Hz. In addition, or instead, the filtering module 24 can make use of a custom filter deemed to provide the best band-limited representation of the underlying bioelectrical signals from the subject's body. The custom filter increases the lower limit of the band to filter out signals generated where there is a lot of background noise or bioelectric noise such as muscle movement. The band for the custom filter is therefore about 16Hz to 70 Hz. In addition, or instead, the custom filter notches out a band in which bioelectric signals occur as a result of "skin twitch" which occurs in some animals. The notched out band is about 27Hz to 35Hz.
The data output from the filtering module 24 is in the form of multiple simultaneous channels, each representing one of the differential signals, for example, the A+ - A. pair of electrodes 14 in Equation 1 above.
The data are then forwarded to a signal processing module 28 where, as shown at step 30 in Fig. 3, a vector projection from the channel data is determined. This is done by applying Equation 1 above. Equation 1 is applied to the channel data on a simultaneous sample-by-sample basis in order to calculate the vector components of a particular sample of data. This has the effect of reducing multiple channel data points to two components being a vector magnitude and phase or x and y components. In the case of the illustrated array 16, the four channel data points are reduced to two vector components. For a greater number of pairs of electrodes the dimensionality reduction is even greater as the end result is always a two component vector.
The vector component is subjected to diagnostic analysis at step 32 of the flowchart of Fig. 3 of the drawings. The advantage of reducing the channel data to a two component vector is that the amount of data to be analysed is significantly reduced. It is reasonable to assume that there is no loss of information given that it is assumed that the underlying bioelectrical signals, particularly in the case of an ECG, are generated by a single source vector, i.e. the heart's electrical excitation wave. Hence the signal projection as represented by the two component vector 18 is sufficient to represent it. It may also be assumed that all non-coherent noise in the signals such as, for example, may arise from electrical noise in the circuits and amplifiers or localised bioelectric artifacts under the electrodes will be reduced and in some cases will sum to zero, hence reducing the overall noise content of the signal.
After reduction to the single vector projection at step 30, the data are analysed, as indicated above at step 32, using standard methods. These methods include simple peak detection and thresholding, matched filtering, autocorrelation, or frequency filtering, according to the diagnostic purpose of the ECG measurement and the preference of the operator. In many cases the final diagnostic decision depends on a software program which uses several of these methods and algorithms and combines the results in a so-called artificial intelligence system in order to provide a high-level diagnostic conclusion equivalent to that of a well-trained human operator. It is well understood in the field that ECG signals are a projection of an internal electrical vector and hence the representation lends itself to intuitive and systematic analytical methods.
It is an advantage of the disclosed embodiments that, firstly, the use of a fixed array of electrodes can be applied at any place on the subject's body and the vector projection of the body's internal electric field can be calculated without requiring any prior knowledge of the location on the body of the electrodes. This also has the advantage that the same instrument can be used at different areas of the body to monitor different signals. Also, as describe above, by reducing the channel data to a two-component vector, the amount of data that must be analysed is significantly reduced. Thus, lower signal processing capabilities are sufficient and this renders the system suitable for use as a portable, handheld system.
The system allows for rapid application to a subject's body that may not be stationary. This is particularly useful for non-compliant subjects, such as uncooperative human subjects or animal subjects.
It will be appreciated by persons skilled in the art that numerous variations and/or modifications may be made to the disclosure as shown in the specific embodiments without departing from the scope of the disclosure as broadly described.
The present embodiments are, therefore, to be considered in all respects as illustrative and not restrictive.