Surgery – Diagnostic testing – Cardiovascular
Reexamination Certificate
2001-07-26
2004-05-04
Schaetzle, Kennedy (Department: 3762)
Surgery
Diagnostic testing
Cardiovascular
C600S300000, C600S519000, C600S521000, C128S920000
Reexamination Certificate
active
06731972
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The invention relates to an evaluation method for determining the time stationarity of measured physiological, in particular cardiologic, signals, which can be implemented preferably on implantable cardiologic devices, such as defibrillators, heart pacemakers or comparable devices.
2. Background Art
For comprehensibility of the present invention, the background must be explained in detail. The most important method of obtaining information on the cardiovascular system resides in recording the ECG by electrodes that are attached to a patient's body surface. Alternatively, a suitable measuring arrangement that includes an implant connected to an electrode may help record an intracardiac ECG or a signal that reflects the course of contraction in an entirely different way, such as intracardiac impedance. As a rule, the morphology of the signal course is evaluated for pathologic processes to be diagnosed or suitable therapeutic measures to be taken.
If any discrete parameters are extracted from the continuous ECG signal, such as the distance of the R wave of the heartbeat at an instant from the R wave of the preceding heartbeat (RR intervals), and if the sequence of values thus produced is outlined in dependence on the number of consecutive heartbeats measured, one will obtain a time series of this cardiac parameter.
It is assumed that in addition to accidental influences, the time series substantially reflects the global regulatory behavior of the cardiovascular system, i.e. the time series, in an abstract manner, gives information on internal transfer functions between components, on latencies of the information transfer between sensors and actuators, or on the working points of the internal transfer functions.
Attempts are made to characterize the status of the cardiovascular system by modeling from time series. This serves various purposes, such as risk stratification or prediction of potentially lethal cardiac events by a defibrillator, or gauging an adaptive frequency cardiac pacemaker. For correct modeling to be ensured, it is necessary to eliminate time series that are influenced by accidental events or during which a patient's condition of physical or psychic stress changes during the period of observation. In the first case, a deterministic model produced from the measurement does not represent the factual condition of the cardiovascular system. In the second case, this condition is not defined at all, owing to a drift or a sudden rise of a parameter of the cardiovascular system. These considerations show that, with a view to obtaining a criterion of stationarity, a selection of certain segments of a time series is a fundamental requirement for modeling and thus for characterizing certain conditions.
So as to be employed in electronic implants, a test for stationarity is required to exhibit, as further decisive criteria, the ability of being automated and the greatest possible simplicity, since the computing capacity is restricted in such implants—at least according to the current state of the art. This is fundamentally due to the fact that as compact as possible computing components must he used which can operate only at an energy supply of limited capacity. As will be explained below, the invention proceeds from these problems.
Regarding the problems posed by the conditions of stationarity upon measurement of any time series, standard tests are known, which are based on a complicated statistical evaluation of the time series and which, as such, do not appear suitable to implementation in implanted cardiologic devices of the current state of processor engineering. However, they will be explained in connection with the invention in order to illustrate the influence of disturbances of stationarity in a time series segment on the evaluation itself. Disturbances of stationarity of a segment can be divided into random and dynamic disturbances. Random disturbances comprise events produced in the heart that are not subject to regulation by the cardiovascular system, such as a temporary atrioventricular block or extrasystoles, i.e. contractions of the ventricle that are not due to sinus node excitation. Random disturbances result from the reception of stimuli from the surroundings, such as sudden noises, but also from inner processes, such as motivations, changes of attention or dreams, Dynamic disturbances result for instance from varying physical and mental stresses over the day that are due to a candidate's activities and go along with varying metabolic demand.
Dynamic disturbances can largely be avoided in physiological experiments. Conventionally, this is achieved by the test conditions being correspondingly prepared, i.e. the candidate is examined in an environment of poor excitation and simultaneously invited to show a passive behavior. However, the examination as such causes multiple psychic processes which can be mastered only by massive drugging. This will interfere with the examination of the regulatory behavior of the cardiovascular system that is affected by the drugging.
Therefore, strictly speaking, preparing test conditions is not possible, and a stationary segment cannot be defined “a priori”, which is all the more true for an implant of autonomic operation. In clinical tests, stationary measuring intervals are mostly determined “by inspection” and “a posteriori”, i.e., for further evaluation, the experimenter selects, from a given time series, a segment as being stationary during which, in his individual view, the signal seems to have a more uniform course than in the remaining segments. The disadvantages of this way of proceeding reside in the subjectivity of this evaluation method, which is primarily based on the lack of quantitative criteria. Moreover, a range may be defined as stationary that is in fact not stationary, which will render any subsequent evaluations faulty. Finally, any subjective definition of measuring intervals that are assumed to be stationary will make various patients comparable only to a limited degree. Also, by nature, subjective evaluation methods of this type cannot be automated.
For identification of the stationary conditions that are necessary for modeling and thus for general diagnostics, and for elimination of nonstationary measuring segments, a method is required that enables the stationarity to be determined “a posteriori” and that simultaneously obeys to the following criteria:
The method must be objective, it must make available a criterion for the fundamental existence of a stationary condition, and it must be quantifiable, comparable and capable of being automated.
In addition, the method is desired not to be based on pure empirism, but to produce a reference to physiological modifications that accompany nonstationary cardiovascular behavior, for the specificity of the test to be augmented.
The simplest type of a statistic test for stationarity is based on the comparison of empiric distributions in two successive segments. The statistic moments of the distributions are determined and compared by means of so-called “two sample tests”. If a significant difference results between the moments of the two distributions, the assumption of stationarity is rejected. This may be a very simple method, but it interprets the cardiovascular system as a purely stochastic system, i.e. as a system having an unlimited number of internal degrees of freedom. Moreover, varying conditions of the cardiovascular system are represented by identical distributions.
Assuming that, apart from external and internal random disturbances, the cardiovascular system obeys to deterministic conditions of development, i.e that the subsequent heartbeat is precisely determined by all the preceding heartbeats, the nonlinear dynamics provide various stationarity tests. Since, however, they are based on the so-called correlation integral, they are very complicated, requiring a well-founded knowledge on a number of unknown parameters and the existence of comparatively long stationary intervals. Mor
Meyer Wolfgang
Schaldach Max
Schaldach, Jr. Max
Biotronik Mess- und Therapiegeräte GmbH & Co. Ingenieurbüro Berl
Browdy and Neimark , P.L.L.C.
Schaetzle Kennedy
Schaldach, Jr. Max
LandOfFree
Evaluation method and apparatus for determining the time... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Evaluation method and apparatus for determining the time..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Evaluation method and apparatus for determining the time... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3269606