METHOD AND APPARATUS FOR PREDICTING POTENTIAL OCCURRENCE

OF ATRIAL FIBRILLATION EVENTS

CROSS-REFERENCE TO RELATED APPLICATION

This application claims priority from U.S. Provisional Patent Application No. 60/722,976 filed on October 4, 2005 by Page et al. and hereby incorporated by reference herein.

FIELD OF THE INVENTION

The present invention relates to heart monitoring and treatment, and, more particularly, to a method and an apparatus for predicting potential occurrence of atrial fibrillation events.

BACKGROUND

Cardiac arrhythmia is a condition wherein a heart experiences an irregular and abnormal cardiac rhythm that differs from its normal sinus rhythm. One type of cardiac arrhythmia is atrial fibrillation in which uncoordinated electrical impulses in a heart's atria cause the atria to fibrillate instead of contracting normally, thereby resulting in ineffective pumping of blood and an irregular and abnormal cardiac rhythm.

Atrial fibrillation is usually not in itself considered life-threatening. However, people with atrial fibrillation are at an increased risk for blood clots and strokes and, if left unattended to, the irregular and abnormal cardiac rhythm associated with atrial fibrillation can lead to heart attack and heart failure. Less severe symptoms of atrial fibrillation include weakness, shortness of breath, and heart palpitations.

Currently, once a patient is diagnosed with atrial fibrillation, cardioversion may be performed in an attempt to convert the irregular and abnormal cardiac rhythm to a normal sinus rhythm.

Cardioversion performed on the patient may include administration of medication, electrical stimulation, or a combination thereof. In some cases, a surgical intervention including implantation of a permanent pacemaker to regulate the cardiac rhythm combined with a drug therapy may be effected to treat atrial fibrillation.

For some individuals, atrial fibrillation results from an underlying condition such as high blood pressure or other heart disease and progressively develops over time with atrial fibrillation episodes occurring more frequently and having a longer duration as time passes.

For other individuals, atrial fibrillation may occur suddenly consequent to a certain event. In particular, after cardiac surgery performed on a patient, there is an increased risk that the patient may have atrial fibrillation. As a result, during a hospital stay of a patient having undergone cardiac surgery, a drug therapy directed to preventing occurrence of atrial fibrillation is performed on the patient. Even so, in some cases, the patient may still experience one or more atrial fibrillation episodes of various lengths, usually beginning two or three days after surgery. When atrial fibrillation does occur, the patient's hospital stay after surgery is normally prolonged by two to five days due to a need for additional medication including antiarrhythmic and anticoagulation drugs. Cardioversion is performed on the patient only if and when an atrial fibrillation episode is detected, which may take several minutes or hours. In addition, irrespective of whether or not the patient has experienced atrial fribrillation, after being released from the hospital, the patient is normally required to continue to take medication directed to preventing occurrence of atrial fibrillation for a period of time typically between four to twelve weeks. Therefore, the patient (1) must experience a hospital stay that may be unnecessarily long to take into account potential occurrence of atrial fibrillation which may never occur; (2) can only be treated for atrial fibrillation following its detection; and (3) must take medication directed to preventing occurrence of atrial fibrillation for a relatively long period of time following surgery.

Accordingly, there is a need for improvements directed to reliably assessing a likelihood of occurrence of atrial fibrillation so as to take appropriate preventive action.

SUMMARY OF THE INVENTION

In accordance with a first broad aspect, the invention provides a device comprising an input for receiving at least one signal indicative of activity of a heart and a processing unit coupled to the input. The processing unit is operative for deriving data regarding a plurality of cardiac parameters of the heart based at least in part on the at least one signal. The processing unit is further operative for determining a likelihood of occurrence of atrial fibrillation of the heart based at least in part on a correlation between the data regarding a first one of the cardiac parameters and the data regarding a second one of the cardiac parameters.

In one non-limiting example embodiment, a device as defined above is implantable in a subject. For example, in a non- limiting example of implementation, an implantable pacemaker comprises a device as defined above.

In accordance with a second broad aspect, the invention provides a computer-readable medium tangibly embodying a program element executable by a computing device to cause the computing device to (1) derive data regarding a plurality of cardiac parameters of a heart based at least in part on at least one signal indicative of activity of the heart; and (2) determine a likelihood of occurrence of atrial fibrillation of the heart based at least in part on a correlation between the data regarding a first one of the cardiac parameters and the data regarding a second one of the cardiac parameters.

In accordance with a third broad aspect, the invention provides a method for medical treatment. The method comprises: receiving at least one signal from at least one electrode contacting a heart of a subject; determining a likelihood of occurrence of atrial fibrillation of the heart prior to manifestation of atrial fibrillation at least in part on a basis of information derived from the at least one signal; and - conditioning medical treatment of the subject at least in part on a basis of the likelihood of occurrence of atrial fibrillation of the heart.

In accordance with a fourth broad aspect, the invention provides a method for recovery treatment of a subject that has undergone heart surgery. The method comprises: receiving at least one signal from at least one electrode positioned in contact with a heart of the subject during the heart surgery; - determining a likelihood of occurrence of atrial fibrillation of the heart prior to manifestation of atrial fibrillation of the heart at least in part on a basis of information derived from the at least one signal; and conditioning recovery treatment of the subject at least in part on a basis of the likelihood of occurrence of atrial fibrillation of the heart.

In accordance with a fifth broad aspect, the invention provides a method for recovery treatment of a subject that has undergone heart surgery. The method comprises: receiving at least one signal from at least one electrode positioned in contact with a heart of the subject during the heart surgery; - determining a likelihood of occurrence of atrial fibrillation of the heart prior to manifestation of atrial fibrillation of the heart at least in part on a basis of information derived from the at least one signal; and conditioning a length of a hospital stay of the subject at least in part on a basis of the likelihood of occurrence of atrial fibrillation of the heart.

In accordance with a sixth broad aspect, the invention provides a method for recovery treatment of a subject that has undergone heart surgery. The method comprises: receiving at least one signal from at least one electrode positioned in contact with a heart of the subject during the heart surgery; - determining a likelihood of occurrence of atrial fibrillation of the heart prior to manifestation of atrial fibrillation of the heart at least in part on a basis of information derived from the at least one signal; selecting between a plurality of treatment options at least in part on a basis of the likelihood of occurrence of atrial fibrillation of the heart, the plurality of treatment options including: dispensing medication to the subject in an attempt to prevent occurrence of atrial fibrillation of the heart;

precluding dispensing medication to the subject in an attempt to prevent occurrence of atrial fibrillation of the heart; implementing one of the treatment options on a basis of the selecting.

In accordance with a seventh broad aspect, the invention provides a method for recovery treatment of a subject that has undergone heart surgery. The method comprises: receiving at least one signal from at least one electrode positioned in contact with a heart of the subject during the heart surgery; determining a likelihood of occurrence of atrial fibrillation of the heart prior to manifestation of atrial fibrillation of the heart at least in part on a basis of information derived from the at least one signal; selecting between a plurality of treatment options at least in part on a basis of the likelihood of occurrence of atrial fibrillation of the heart, the plurality of treatment options including: - performing electrical stimulation of the heart of the subject in an attempt to prevent occurrence of atrial fibrillation of the heart; precluding performing electrical stimulation of the heart of the subject in an attempt to prevent occurrence of atrial fibrillation of the heart; implementing one of the treatment options on a basis of said selecting.

These and other aspects and features of the present invention will now become apparent to those of ordinary skill in the art upon review of the following description of specific embodiments of the invention in conjunction with the accompanying drawings.

BRIEF DESCRIPTION OF THE DRAWINGS

A detailed description of specific embodiments of the present invention is provided herein below, by way of example only, with reference to the accompanying drawings, in which:

Figure 1 is a diagrammatic view of an apparatus for performing a prediction regarding atrial fibrillation of a heart of a patient, in accordance with a first specific embodiment of the present

invention;

Figure 2 is a flow chart illustrating operation of a processing unit of the apparatus shown in Figure 1 , in a particular example of implementation;

Figure 3 is a block diagram illustrating interaction of the processing unit of the apparatus shown in Figure 1 with a communication device, in a particular example of implementation;

Figure 4 is a block diagram illustrating interaction of the processing unit of the apparatus shown in Figure 1 with a treatment unit, in a particular example of implementation; and

Figure 5 is a diagrammatic view of an implantable apparatus for performing a prediction regarding atrial fibrillation of a heart of a patient, in accordance with a second specific embodiment of the present invention.

In the drawings, the embodiments of the invention are illustrated by way of examples. It is to be expressly understood that the description and drawings are only for the purpose of illustration and are an aid for understanding. They are not intended to be a definition of the limits of the invention.

DETAILED DESCRIPTION OF EMBODIMENTS

Figure 1 shows an apparatus 10 for performing a prediction regarding atrial fibrillation of a heart 12 of a patient, in accordance with a first specific embodiment of the present invention. In this specific embodiment, the apparatus 10 is usable, for example, following cardiac surgery performed on the patient in order to predict whether the heart 12 may experience an atrial fibrillation episode.

In this particular embodiment, the apparatus 10 comprises three electrodes 14i-14 ₃ contacting the heart 12 and a processing unit 22 coupled to the electrodes 14i-14 ₃ via an input 20. As described below, the processing unit 22 interacts with the electrodes 14j-14 ₃ to monitor

electrophysiological activity of the heart 12 and perform a prediction regarding potential occurrence of an atrial fibrillation event based at least in part on this monitored electrophysiological activity. As also described below, based on the outcome of this prediction, various prophylactic actions such as electrical stimulation of the heart 12, administration of antiarrythmic drugs, or both, may be performed in an attempt to prevent occurrence of an atrial fibrillation episode of the heart 12.

In the specific embodiment shown in Figure 1, each of the electrodes 14pl4 ₃ contacts a respective portion of the heart 12 and is operative to sense electrophysiological activity of that portion of the heart 12 and transmit to the processing unit 22 a signal indicative of this electrophysiological activity. For example, the signal indicative of electrophysiological activity may be a signal indicative of an electric potential of the respective portion of the heart 12. In this particular case, the electrode 14i contacts the heart 12 at a right atrium/sinoatrial node (RA/SA node) region thereof, the electrode 14 ₂ contacts the heart 12 at a low right atrium (low RA) region thereof, and the electrode 14 ₃ contacts the heart 12 at a left atrium (LA) region thereof. Each of the electrodes 14j-14 ₃ may be a commercially-available temporary epicardial electrode that is sutured or otherwise attached to the epicardium of the heart 12 by a surgeon at the end of the cardiac surgery performed on the patient. Each of the electrodes 14 _r 143 extends through the body of the patient and is coupled to the processing unit 22 via the input 20.

Although in the embodiment shown in Figure 1 three electrodes 14i-14 ₃ are used, it is to be understood that one, two or any number of electrodes may be used without departing from the scope of the invention. Furthermore, while in the embodiment shown in Figure 1 a certain disposition of the electrodes 14]-14 ₃ is presented, it is to be understood that any disposition of the electrodes may be employed (e.g., one electrode in the SA node region and two electrodes in the LA region).

With continued reference to Figure 1, and as described below, the processing unit 22 is operative to process signals received from the electrodes 14i-14 ₃ so as to perform a prediction regarding atrial fibrillation of the heart 12. More specifically, the processing unit 22 uses signals received from the electrodes 14i-14 ₃ to monitor how a plurality of parameters

characterizing activity of the heart 12 correlate with each other and determines a likelihood of occurrence of an atrial fibrillation episode of the heart 12 based at least in part on this monitoring. In a specific non-limiting embodiment, the processing unit 22 includes a Holter monitoring module coupled to the electrodes 14i- 14 ₃ for receiving signals therefrom and a computer coupled to the Holter monitoring module for processing the received signals. Generally, the processing unit 22 may be implemented using hardware, firmware, software, control logic, or a combination thereof.

Figure 2 illustrates operation of the processing unit 22 in a particular non-limiting example of implementation. In this particular example, the processing unit 22 generates data based on signals received from the electrodes 14)-14 ₃ over a period of time T conceptually divided into a number of successive time intervals N _1n , _enah each having a duration At = T / N _1n , _enah . For instance, the period of time T may be two hours and each time interval At may be of two minutes for a number of time intervals N _lnU , _nαλ of 60.

Step 200:

The first time interval At ₁ is considered with index / = 1 .

Step 202: The processing unit 22 derives data regarding a number of normal sinus beats (νSB) N _NSB , and a number of premature atrial contractions (PAC) N _{PAC l} in the time interval At ₁ based at least in part on signals received from the electrodes 14]-14 ₃ . More particularly, the processing unit 22 detects each beat B _k , in the time interval At ₁ based on the magnitude of the signals received from the electrodes 14i-14 ₃ . The processing unit 22 also derives for each detected beat B _{k l} data regarding a number of parameters characterizing that beat, based on the signals received from the electrodes 14j-14 ₃ . For instance, the processing unit 22 may derive values of the following time parameters characterizing each beat B _k , based on activation times of the electrodes 14 _r 14 ₃ :

_/ itit ra-ult ial , c ami , Parameter indicative of an intra-atrial conduction time, which may ^J be the intra-atrial conduction time itself. Values of two such conduction times may be derived in the non-limiting example of implementation shown in Figure 1, namely: (1) the right atrial conduction time corresponding to the time interval between the activation time of the electrode and the activation time of the electrode 14 ₂ , and (2) the total atrial conduction time corresponding to the time interval between the activation time of the electrode 14i and the activation time of the electrode 14 ₃ .

t atrioventricul .ar cona , Parameter indicative of an atrioventricular conduction time, which may ^J be the atrioventricular conduction time itself. In the non-limiting example of implementation shown in Figure 1 , ventricular activity is captured and represented in signals received from the electrodes 14]-143. A value of an atrioventricular conduction time, for example, for the right atrium/ventricle, may thus be derived on a basis of signals received from the electrodes 14i-14 ₃ by computing a time interval between signal components associated with atrial activity and signal components associated with ventricular activity.

dv Parameter indicative of a time derivative of the signal of one of the electrodes 14]- 14 ₃ , which may be the time derivative itself. Values of three such time derivatives may be derived in the non-limiting example of implementation shown in Figure 1, namely: (1) the time derivative of the signal received from the electrode 14i; (2) the time derivative of the signal received from the electrode 14 ₂ ; and (3) the time derivative of the signal received from the electrode 14 ₃ .

Based on the derived data regarding the parameters characterizing each beat B _k , , the processing unit 22 determines whether each beat B _{k l} is a NSB or a PAC so as to obtain the number of NSBs N _{NSB l} and the number of PACs N _{FAC l} in the time interval At ₁ .

Step 204:

For each NSB NSB , , where j e (1,..., N _NSH , ) , the processing unit 22 derives data regarding a plurality of cardiac parameters X _r ..X _u based at least in part on signals received from the electrodes 14i-14 ₃ . Each of the cardiac parameters X _r ..X _u is a parameter characterizing activity of the heart 12. For example, each of the cardiac parameters X _v ..X _ιU may be a

-— — ' ^• ' ^■ ~ - f — ^• in which case the processing unit 22 has already derived the data regarding these parameters for each NSB NSB _{1 1} . A given one of the cardiac parameters X _r ..X _Xf may also be a parameter indicative of a time integral of the signal of one of the electrodes 14i-14j during each NSB NSB ₁ , . These and other examples of parameters for each of the cardiac parameters

X _r ..X _υ are presented below: however, it should be noted that parameters other than the ones listed below are possible for each of the cardiac parameters X _x ...X _hl without departing from the scope of the invention.

t _{mt ιa} _ _alru ,ι _uml Parameter indicative of an intra-atrial conduction time, which may be the intra- atrial conduction time itself. Values of two such conduction times may be derived in the non-limiting example of implementation shown in Figure 1, namely: (1) the right atrial conduction time corresponding to the time interval between the activation time of the electrode 14i and the activation time of the electrode 14 ₂ , and (2) the total atrial conduction time corresponding to the time interval between the activation time of the electrode 14] and the activation time of the electrode \4^.

t all nn aili ic iiU ,it , Parameter indicative of an atrioventricular conduction time, ' which may ^J be the atrioventricular conduction time itself. In the non-limiting example of implementation shown in Figure 1 , ventricular activity is captured and represented in signals received from the electrodes 14 ₁ - 14 ₃ . A value of an atrioventricular conduction time, for example, for the right atrium/ventricle, may thus be derived on a basis of signals received from the electrodes 14[-H ₃ by computing a time interval between signal components associated with atrial activity and signal components associated with ventricular activity.

dv Parameter indicative of a time derivative of the signal from one of the

" ^r electrodes 14 _r 14 ₃ , which may be the time derivative itself. Values of three such time derivatives may be derived in the non-limiting example of implementation shown in Figure 1, namely: (1) the time derivative of the signal received from the electrode 14i; (2) the time derivative of the signal received from the electrode 14 ₂ ; and (3) the time derivative of the signal received from the electrode 14 ₃ .

f _v fa Parameter indicative of a time integral of the signal of one of the electrodes 14i-14 ₃ , which may be the time integral itself. Values of three such time integrals may be derived in the non-limiting example of implementation shown in Figure 1, namely: (1) a time integral of the signal of the electrode 14j; (2) a time integral of the signal of the electrode 14 ₂ ; and (3) a time integral of the signal of the electrode 14 ₃ .

1 _{44 mu} ,, _uι , Parameter indicative of heart period based on atrial activation, which may be the heart period itself. In the non-limiting example of implementation shown in Figure 1 , a value of this parameter may be derived, for example, on a basis of the signal received from the electrode 14] by computing a time interval between successive activation of the right atrium.

For instance, in a non-limiting example of implementation, the processing unit 22 may derive values often cardiac parameters X _x ...X ₁₀ based at least in part on signals received from the electrodes Hi-H ₃ , where: ^i = ',m ra-amai umci ,1-2 ^{is the ri} g ^ht atrial conduction time; ^X 2 = Wamα/ cK/ ,i-3 ^{is the total atrial} conduction time; ^X ₃ = t _a , _n<ne»wcu i _a , _to . _u i ^{is the} atrioventricular conduction time; dv is the time derivative of the signal received from the electrode 14j

⁴ dt 1 dv

X, = is the time derivative of the signal received from the electrode 14 ₂ ; dt

X - * is the time derivative of the signal received from the electrode 14 ₃ ; ^{6 "} dt X ₁ = Jv dt is the time integral of the signal received from the electrode H ₁ ;

X _s = Jv dt is the time integral of the signal received from the electrode H ₂ ;

X ₉ = Jv dt is the time integral of the signal received from the electrode H ₃ ; and

X _w = t _{λA mU} , _mιl is the heart period based on atrial activation.

Again, it is to be expressly understood that the above list of parameters and number of parameters represents only a specific example and that various other cardiac parameters may be derived and used in other examples of implementation.

It will thus be appreciated that, for the time interval At ₁ , the processing unit 22 derives a series of data elements (x, ,...,x _v ),„ , such as values for each cardiac parameter X _1n , where m e (L..., M) .

Step 206:

For each pair of cardiac parameters (X _p ;X _q ) , where (p,q) e (l,...,M)and p ≠ q , the processing unit 22 derives an indication of a level of correlation C(X _p ',X ₁₁ ), between the data regarding the cardiac parameter X _p and the data regarding the cardiac parameter X _q . More particularly, the processing unit 22 derives an indication of a level of correlation C(X _p ; X ₁₁ ), between the series of data elements (x, ,...,x _λ ) _{p ι} for the cardiac parameter X _p and the series of data elements (x, ,...,x _{V β} ) _{q ι} for the cardiac parameter X _q . For example, the indication of the level of correlation C(X _p \X _q ), may be the coefficient of correlation or covariance between the series of data elements (x, ,...,x _λ ) _/; , and the series of data elements (x, ,...,x _N ) _(/ , , or any other conceivable indication of the level of correlation between these two series of data elements. In a non-limiting example of implementation, the indication of the level of correlation may be the correlation coefficient expressed as:

where x _p is the mean of the series of data elements (x, ,...,x _λ ) _{p l} and x _q is the mean of the series of data elements (x, ,...,x _{N SB} ) _q , .

Thus, for the time interval At ₁ , the processing unit 22 obtains a set of correlation indicators

(C(X, -X ₂ ) _I ,...,C(X _KI _ _[ - X _M ) _I ) = (C _υ ,...,C _{R l} ) , where R = ^M{~M ^ ,

Although in the above example each pair of cardiac parameters (X _p ; X _q ) formable from the

M cardiac parameters produces the set of R correlation indicators, in other examples of implementation, only a subset of these pairs (i.e., not all of these pairs) may be taken into consideration, thereby producing a set of correlation indicators containing less than R correlation indicators.

Step 208:

If the index i is equal to N _{int em//s} . the last considered time interval At ₁ was the last time interval in the period of time T and step 210 is performed; otherwise, the value of the index i is incremented by one and the processing unit 22 returns to step 202.

Step 210:

For each indication of a level of correlation C ₁ . , where r e (\,...,R) , the processing unit 22 derives an indication of a probability of stability over time of that indication C ₁ . , P(C ₁ . ) . More specifically, the processing unit 22 derives the indication of the probability of stability over time P(C ₁ .) based on the values of C ₁ . ,...., C ₁ . _λ . That is, the processing unit 22 derives an indication of a probability of stability over time of the indication of the level of correlation C ₁ , f(C, ) , based on the values of C i ,,...,C _{1 v} ; an indication of a probability of stability over time of the indication of the level of correlation C ₂ , P(C ₂ ) , based on the values of C _{2 1} ,..., C _{2 N} ; and so on.

In one non-limiting example of implementation, the value of each P(C _r ) is obtained by applying a non-parametric statistical method on the values of C _{r ]} ,...,C _{r X a ι} . The non- parametric statistical method, which is based on the technique of surrogate data, computes a weighted sum of the values of C,. ,,..., C,. _{V ( (} using a set of weights reflecting a temporal ordering of the values of C ₁ . ,,..., C ₁ . _κ , for instance, the index / of the time intervals.

Surrogate time series are generated by random permutations of the values of C ₁ . ,,...,C,. _{λ ι} and, for each generated surrogate time series, a respective weighted sum of the randomly permutated values of C _{r ]} ,...,C _{r N} is computed, thereby resulting in a distribution of the weighted sum under consideration. The value of each P(C ₁ .) is obtained

by comparing the actual weighted sum of the values of C ₁ . ,,..., C ₁ . _v to the distribution of the weighted sums for the surrogate time series.

Thus, for the period of time T , the processing unit 22 obtains a probability of stability vector P = [P(C ₁ ),..., P(C _R )) . This probability of stability vector P is indicative of the degree of temporal stability of the respective level of correlation between each pair of cardiac parameters

(X _p , ' X ₁₁ ) . In effect, a given entry P(C ₁ . ) in the probability of stability vector P is indicative of how stable the level of correlation C ₁ . between the pair of cardiac parameters (X _p ;X _q ) ϊs over time. The less stable is C ₁ . over time, i.e. the greater is the variation in C ₁ . over time, the lower is P(C ₁ . ) , and vice versa.

Step 212:

The processing unit 22 determines a likelihood of occurrence of atrial fibrillation of the heart

12 based at least in part on the respective level of correlation between each pair of cardiac parameters (X _p ; X ₁₁ ) . More specifically, in this particular embodiment, the processing unit

22 determines a likelihood of occurrence of atrial fibrillation of the heart 12 based at least in part on the probabilities of stability P(C ₁ ),..., P(C _R ) of the levels of correlation.

In one non-limiting example embodiment, determining a likelihood of occurrence of atrial fibrillation of the heart 12 comprises deriving a quantitative indication of a likelihood of occurrence of atrial fibrillation of the heart 12, based at least in part on the data regarding the levels of correlation. For example, this quantitative indication may be a probability of occurrence of atrial fibrillation of the heart 12 or any other indicator quantifying the risk that atrial fibrillation of the heart 12 may occur. In a particular case, the quantitative indication of a likelihood of occurrence of atrial fibrillation of the heart 12 is the result of a function of the probabilities of stability /(P(C, ),..., P(C _R )) .

In another non-limiting example embodiment, determining a likelihood of occurrence of atrial fibrillation of the heart 12 comprises determining whether a condition related to atrial fibrillation of the heart 12 is satisfied, based at least in part on the data regarding the levels of

correlation. When the condition is determined to be satisfied, an atrial fibrillation episode of the heart 12 is deemed likely to occur. When the condition is determined to be not satisfied, an atrial fibrillation episode of the heart 12 is deemed unlikely to occur.

Satisfaction of the condition related to atrial fibrillation of the heart 12 may take on various forms depending on how this condition is defined. For example, the condition may be deemed to be satisfied if the value of each probability of stability P(C ₁ ) is less than a respective predetermined threshold P ₁ (C ₁ ) , i.e. P(C ₁ ) < P ₁ (C ₁ ) , P(C ₂ ) < P ₁ (C ₂ ) , and so on. As another example, the condition may be deemed to be satisfied if the result ;μ of a function of the probabilities of stability g(P(C^ ),..., P(C _R )) is greater than a predetermined threshold y _t (or less than the predetermined threshold y, , depending on how the function is defined). In a particular case, the function g(P(C _x ),...,P(C _R )) may be the function /(P(C ₁ ),..., P(C _n )) mentioned above which is used to derive a quantitative indication of a likelihood of occurrence of atrial fibrillation of the heart 12, in which case the result y is the quantitative indication of a likelihood of occurrence of atrial fibrillation (e.g., a probability of occurrence of atrial fibrillation of the heart 12). It is to be understood that the condition may be defined in various other ways without departing from the scope of the invention. For instance, the condition may be defined such that it is deemed to be satisfied when each of a plurality of sub- conditions is satisfied.

In one non-limiting example embodiment, in addition to using the data on the levels of correlation between the cardiac parameters X ₁ ...X _hl in determining a likelihood of occurrence of atrial fibrillation of the heart 12, the processing unit 22 determines this likelihood also based at least in part on data regarding at least one cardiac parameter Z . The data regarding each cardiac parameter Z may be derived, for instance, at step 202 or step 204. In a particular case, the cardiac parameter Z is a function of the number of PACs N _PAi , in each time interval

At ₁ for the period of time T such as the total number of PACs for the period of time T ,

λ

N _PAi , = ^T N _PAC , , or the density of PACs for the period of time T , N _P4i , IT . hi such a case

1=1 and in an embodiment in which determining a likelihood of occurrence of atrial fibrillation

comprises determining whether a condition is satisfied, the condition may be deemed to be satisfied, for example, if the value of each probability of stability P(C ₁ ) is less than a respective predetermined threshold P ₁ (C, ) (or if the result y of a function of the probabilities of stability g(P(C _] ),..., P(C _R )) is greater (or less) than a predetermined threshold y, ), and the total number of P ACs N _P4( , is greater than a certain value.

Thus, upon completing step 212, the processing unit 22 has determined a likelihood of occurrence of atrial fibrillation of the heart 12.

Use of correlation between cardiac parameters to assess a likelihood of occurrence of atrial fibrillation as described above has been exemplified by a study conducted on thirty patients having undergone isolated coronary artery bypass surgery. Information regarding this study is presented herein only for illustrative purposes and should not be interpreted as limiting in any respect. During the study, characteristics of atrial unipolar electrograms were assessed using continuous recordings during four post-operative days (starting twenty- four hours after surgery). Electrogram integrals, time derivative of activation complexes, AA intervals, interatrial and atrioventricular conduction times were extracted from signals obtained via three epicardial temporary electrodes sutured on the right atrium free wall and the posterior left atrium. Atrial fibrillation occurred at 2.8 ± 1.2 days postoperatively in fifteen patients referred to as Group I patients (mean age of 67 years old). Measurements in Group I patients data sets were made on a beat-to-beat basis over two-hour intervals preceding atrial fibrillation episodes and compared to randomly selected two-hour intervals in fifteen patients without postoperative atrial fibrillation, referred to as Group II patients (mean age of 63 years old). The parameters were grouped in pairs to assess their correlation throughout the study intervals, thereby forming a set of 27 variables. The temporal stability of each correlation variable, expressed as a probability, was tested using a non-parametric method with reference to time series consisting of random permutations. A cluster analysis of the correlation variables showed significantly less stability in Group I than in Group II time series (100% discrimination power). Thus, this study exemplified that a correlation stability test of atrial electrophysiological variables has the potential to discriminate atrial behavioral patterns prone to atrial fibrillation. Once again, it is to be expressly understood that the above information

regarding this study is presented herein only for illustrative purposes and should not be interpreted as limiting the present invention in any respect.

Hence, as mentioned above, upon completing step 212, the processing unit 22 has determined a likelihood of occurrence of atrial fibrillation of the heart 12. Based on this determined likelihood, the processing unit 22 may perform various actions, examples of which will now be described with reference to Figures 3 and 4.

Figure 3 illustrates a first example of an action that may be performed by the processing unit 22 based on the determined likelihood of occurrence of atrial fibrillation of the heart 12. hi this example, the processing unit 22 is operative to generate a signal that is released via an output 30 to a notification device 32 so as to cause the notification device 32 to provide an indication of the likelihood of occurrence of atrial fibrillation of the heart 12.

For instance, the notification device 32 may be a bedside computer terminal adjacent to the patient and the indication of the likelihood of occurrence of atrial fibrillation of the heart 12 may be provided by displaying information on a display of the bedside computer terminal. In an embodiment in which the processing unit 22 determines a quantitative indication of a likelihood of occurrence of atrial fibrillation of the heart 12, the displayed information can include this quantitative indication itself, e.g.. a probability of occurrence of an atrial fibrillation episode of the heart 12. In an embodiment in which atrial fibrillation of the heart 12 is deemed likely or unlikely to occur based on whether a condition is or is not satisfied, the displayed information may include a message characterizing a risk that atrial fibrillation may occur such as "High risk of atrial fibrillation" if the condition is satisfied or "Low risk of atrial fibrillation" if the condition is not satisfied. Alternatively, the processing unit 22 may generate and transmit a signal to the notification device 32 only responsive to a determination that the condition is satisfied. In such a case, the signal sent to the notification device 32 acts as an alarm signal to indicate likely occurrence of an atrial fibrillation episode of the heart 12.

Although in the above example the notification device 32 is a bedside computer terminal, it is to be understood that the notification device 32 can take on various forms. Generally, the notification device 32 can be a static notification device such as a computer terminal or a

mobile notification device (including a mobile communication device) such as a laptop computer, a tablet personal computer (PC), a portable digital assistant (PDA), a wide local area network (WLAN) phone, a cellular phone, or any other conceivable device capable of providing a perceivable indication of the likelihood of occurrence of atrial fibrillation of the heart. Accordingly, the signal generated by the processing unit 22 may be transmitted to the notification device 32 via a wired link, a wireless link, or a combination thereof, depending on whether the notification device 32 is static or mobile. Furthermore, while in the above example visual means are used to indicate potential occurrence of an atrial fibrillation episode, it will be appreciated that audible alarm signals may also be used to indicate potential occurrence of an atrial fibrillation episode.

The notification device 32 providing an indication of a likelihood of occurrence of atrial fibrillation of the heart 12 enables medical personnel responsible for the patient to provide adequate treatment to the patient so as to attempt to prevent an atrial fibrillation episode of the heart 12. In effect, when the notification device 32 indicates that an atrial fibrillation episode of the heart 12 is likely to occur, the medical personnel may give prophylactic treatment to the patient in an attempt to prevent occurrence of an atrial fibrillation episode.

The prophylactic treatment may include electrical stimulation (e.g. rapid atrial pacing) of the heart 12 using, for example, an electrical cardioversion unit or temporary pacemaker which may be available in proximity of the patient. The electrical stimulation may be a so-called "overdrive" pacing of the heart 12, which imposes a pacing faster than the normal sinus rhythm of the heart 12 but slower than an atrial fibrillation rhythm. Alternatively, the prophylactic treatment may include administration of antiarrhythmic drugs such as ibutilide, procainamide, flecainide, propafenone, sotalol, amidodarone, or any other drug useful in preventing or treating arrhythmia. As yet another alternative, the prophylactic treatment may include a combination of both electrical stimulation of the heart 12 and administration of drugs to the patient.

Advantageously, giving prophylactic treatment to the patient based on the determined likelihood of occurrence of atrial fibrillation may prevent an actual atrial fibrillation episode of the heart 12 from occurring. In turn, this reduces the risk of negative effects associated with

atrial fibrillation such as uncomfortable symptoms of and complications resulting from atrial fibrillation, irregular cardiac rhythm, strokes, necessity for anticoagulation drugs, prolongation of hospital stay, etc.

Furthermore, in some cases, rather than leading to administration of prophylactic treatment to the patient, information regarding the determined likelihood of occurrence of atrial fibrillation of the heart 12 may be relied on to preclude administration of certain types of treatment to the patient (e.g., preclude administration of certain medication or preclude electrical stimulation treatments). This may avoid provision of unnecessary treatment in some cases.

Moreover, it may be recognized that information regarding the determined likelihood of occurrence of atrial fibrillation of the heart 12 may also be used for adjusting a length of the hospital stay of the patient. For instance, the length of the hospital stay of the patient when this determined likelihood indicates that atrial fibrillation of the heart 12 is unlikely to occur may be less than when this determined likelihood indicates that atrial fibrillation is likely to occur. This can avoid patients experiencing unnecessary prolonged hospital stays when the risk of occurrence of atrial fibrillation is relatively low.

Referring now to Figure 4, there is shown a second example of an action that may be performed by the processing unit 22 based on the determined likelihood of occurrence of atrial fibrillation of the heart 12. In this example, the processing unit 22 is operative to generate a signal that is released via an output 34 to a treatment unit 40 proximate to the patient so as to cause the treatment unit 40 to provide treatment to the patient directed to preventing an atrial fibrillation episode of the heart 12. For instance, the treatment unit 40 may be implemented as a bedside treatment unit or a unit worn by the patient.

The treatment unit 40 is adapted to automatically provide treatment to the patient directed to preventing an atrial fibrillation episode of the heart 12 in response to receiving a signal from the processing unit 22. For example, in certain non-limiting example embodiments, the treatment unit 40 is capable of operating in a plurality of operative conditions and, responsive to receiving a signal from the processing unit 22, the treatment unit 40 is caused to acquire one

of these operative conditions.

In one non-limiting embodiment, the treatment unit 40 may include an electrical stimulation unit for electrically stimulating the heart 12. The electrical stimulation unit may comprise a pulse generator adapted to generate electric pulses for electrically stimulating the heart 12, the pulses being transmitted to the heart 12 via one or more of the electrodes 14i-14 ₃ or via another electrode (not shown) such as a permanently implanted heart lead. In another non- limiting embodiment, the treatment unit 40 may include a drug administration unit for automatically administrating drugs to the patient. The drug administration unit may comprise a container containing one or more pharmacological agents, a controllable pump connected to the container, and a conduit linking the pump to the patient's bloodstream so as to controllably release the pharmacological agents in the patient's bloodstream. In yet another non-limiting embodiment, the treatment unit 40 may include both an electrical stimulation unit and a drug administration unit for automatically electrically stimulating the heart 12 and administrating medication to the patient.

Based on the determined likelihood of occurrence of atrial fibrillation of the heart 12, the processing unit 22 may generate a signal that is sent to the treatment unit 40. For instance, in an embodiment in which atrial fibrillation of the heart 12 is deemed likely or unlikely to occur based on whether a condition is or is not satisfied, the processing unit 22 may generate a signal that is transmitted to the treatment unit 40 in response to a determination that the condition is satisfied. Responsive to receiving the signal from the processing unit 22, the treatment unit 40 acquires an operative condition in which it automatically provides treatment to the patient directed to preventing an atrial fibrillation episode of the heart 12. For example, in an embodiment in which the treatment unit 40 includes an electrical stimulation unit, in response to receiving the signal from the processing unit 22, the electrical stimulation unit provides an electrical stimulation treatment of the heart 12 in an attempt to prevent atrial fibrillation of the heart 12. As mentioned previously, the electrical stimulation treatment may be in the form of an overdrive pacing treatment. As another example, in an embodiment in which the treatment unit 40 includes a drug administration unit, in response to receiving the signal from the processing unit 22, the drug administration unit administers drugs to the patient in an attempt to prevent atrial fibrillation of the heart 12. The administered drugs may be any of the drugs

previously mentioned.

The characteristics of the treatment provided to the patient may depend on the determined likelihood of occurrence of atrial fibrillation of the heart 12. For instance, in an embodiment in which the processing unit 22 determines a probability of occurrence of an atrial fibrillation episode of the heart 12, the characteristics of the treatment may be related to this probability. For example, a first probability of occurrence of an atrial fibrillation episode may result in a treatment having a first set of characteristics while a second, different probability of occurrence of an atrial fibrillation episode may result in a treatment having a second, different set of characteristics. The characteristics of the treatment may be specified in terms of, e.g., electrical stimulation parameters (e.g. voltage, frequency, etc.) and/or medication administration parameters (e.g. type of medication, dosage, etc.). Thus, as the processing unit 22 continuously monitors activity of the heart 12 and updates the assessment of the likelihood of occurrence of atrial fibrillation of the heart 12, the treatment provided to the patient is automatically regulated to ensure that the patient is provided with the optimal treatment at any given time.

Figure 5 shows an implantable apparatus 110 for performing a prediction regarding atrial fibrillation of a heart 112 of a patient, in accordance with a second non-limiting embodiment of the present invention. Once implanted into the patient, the apparatus 110 is operative to predict whether the heart 112 may experience an atrial fibrillation episode.

In this particular embodiment, the implantable apparatus 110 comprises three electrodes 114]- 114 ₃ contacting the heart 112 and an enclosure 120 within which is disposed a processing unit 122 coupled to the electrodes 114i-l 14 ₃ , a treatment unit 140 coupled to the processing unit 122, and a signaling unit 142 coupled to the processing unit 122. A power source 134 such as a battery is also disposed within the enclosure 120 for powering the processing unit 122, the treatment unit 140, and the signaling unit 142. The enclosure 120 is adapted to be implanted, for example, in an upper left chest portion of the patient within a subcutaneous pocket and may be made of a biocompatible material such as those conventionally used in manufacturing permanent pacemakers.

As described below, the processing unit 122 interacts with the electrodes 1 14i-l 14 ₃ to monitor electrophysiological activity of the heart 112 and determine a likelihood of occurrence of atrial fibrillation of the heart 112 based at least in part on this monitored electrophysiological activity. As also described below, based on the determined likelihood, the processing unit 122 may interact with the treatment unit 140 and the signaling unit 142 so as to cause electrical stimulation of the heart 112. administration of antiarrythmic drugs, or both, in order to attempt to prevent an atrial fibrillation episode of the heart 1 12.

In the specific embodiment shown in Figure 5, each of the electrodes 1 14i-l 14 ₃ is operative to sense electrophysiological activity of a respective portion of the heart 112 and to transmit to the processing unit 122 a signal indicative of this sensed electrophysiological activity. For example, the signal indicative of electrophysiological activity may be a signal indicative of an electric potential of the respective portion of the heart 1 12. In this particular case, the electrode 114i is disposed in a right atrium/sinoatrial node (RA/SA) region of the heart 112, the electrode 1 14 ₂ is disposed in a low right atrium (low RA) region of the heart 112, and the electrode 114 ₃ is disposed in a left atrium (LA) region of the heart 112. Each of the electrodes 114i-l 14 ₃ may be a commercially available electrode that is transvenously implanted into the heart 1 12 and fixed to the myocardium thereof by known means.

Although in the embodiment shown in Figure 5 three electrodes 1 14i-l H ₃ are used, it is to be understood that one, two or any number of electrodes may be used without departing from the scope of the invention. Furthermore, while in the embodiment shown in Figure 5 a certain disposition of the electrodes 114 _r l 14 ₃ is presented, it is to be understood that any disposition of the electrodes may be employed (e.g., one electrode in the SA node region and two electrodes in the LA region).

The processing unit 122 is coupled to the electrodes 114i-l 14 ₃ for receiving the signal from each of the electrodes 114(-114 ₃ . In this particular non-limiting example of implementation, the processing unit 122 operates as the processing unit 22 described previously in connection with Figures 1 and 2. Specifically, the processing unit 122 is operative to process signals received from the electrodes 114i-l 14 ₃ so as to determine a likelihood of occurrence of atrial fibrillation of the heart 112. Based on this determined likelihood, the processing unit 122 may

perform various actions, examples of which will now be described.

With continued reference to Figure 5, the treatment unit 140 is adapted to automatically provide treatment to the patient directed to preventing an atrial fibrillation episode of the heart 1 12. In one embodiment, the treatment unit 140 may include an electrical stimulation unit for electrically stimulating the heart 112. The electrical stimulation unit may comprise a pulse generator adapted to generate electric pulses for electrically stimulating the heart 1 12, the pulses being transmitted to the heart 1 12 via one or more of the electrodes 114i-l 14 ₃ or via another electrode (not shown) disposed within the heart 1 12. In another embodiment, the treatment unit 140 may include a drug administration unit for automatically administrating drugs to the patient. The drug administration unit may comprise a container containing one or more pharmacological agents, a controllable pump connected to the container, and a conduit linking the pump to the patient's bloodstream so as to controllably release the pharmacological agents in the patient's bloodstream. In yet another embodiment, the treatment unit 140 may include both an electrical stimulation unit and a drug administration unit for automatically electrically stimulating the heart 112 and administrating medication to the patient.

Based on the determined likelihood of occurrence of atrial fibrillation of the heart 112, the processing unit 122 may generate a signal that is sent to the treatment unit 140. For instance, in an embodiment in which atrial fibrillation of the heart 112 is deemed likely or unlikely to occur based on whether a condition is or is not satisfied, the processing unit 122 may generate a signal that is sent to the treatment unit 140 in response to a determination that the condition is satisfied. Responsive to receiving the signal from the processing unit 122, the treatment unit 140 automatically provides treatment to the patient directed to preventing an atrial fibrillation episode of the heart 112. For example, in an embodiment in which the treatment unit 140 includes an electrical stimulation unit, in response to receiving the signal from the processing unit 122, the electrical stimulation unit provides an electrical stimulation treatment of the heart 112 in an attempt to prevent atrial fibrillation of the heart 112. As mentioned previously, the electrical stimulation treatment may be in the form of an overdrive pacing treatment. As another example, in an embodiment in which the treatment unit 140 includes a drug administration unit, in response to receiving the signal from the processing unit 122, the drug administration unit administers drugs to the patient in an attempt to prevent atrial fibrillation

of the heart 112. The administered drugs may be any of the drugs previously mentioned.

The characteristics of the treatment provided to the patient may depend on the determined likelihood of occurrence of atrial fibrillation of the heart 112. For instance, in an embodiment in which the processing unit 122 determines a probability of occurrence of an atrial fibrillation episode of the heart 112, the characteristics of the treatment may be related to this probability. For example, a first probability of occurrence of an atrial fibrillation episode may result in a treatment having a first set of characteristics while a second, different probability of occurrence of an atrial fibrillation episode may result in a treatment having a second, different set of characteristics. The characteristics of the treatment may be specified in terms of, e.g., electrical stimulation parameters (e.g. voltage, frequency, etc.) and/or medication administration parameters (e.g. type of medication, dosage, etc.). Thus, as the processing unit 122 continuously monitors activity of the heart 1 12 and updates the assessment of the likelihood of occurrence of atrial fibrillation of the heart 112, the treatment provided to the patient is automatically regulated to ensure that the patient is provided with the optimal treatment at any given time.

Continuing with Figure 5, the signaling unit 142 is adapted to provide a perceivable indication of the determined likelihood of occurrence of atrial fibrillation of the heart 112. More particularly, based on the determined likelihood of occurrence of atrial fibrillation of the heart 112, the processing unit 122 may generate a signal that is sent to the signaling unit 142. For instance, in an embodiment in which atrial fibrillation of the heart 112 is deemed likely or unlikely to occur based on whether a condition is or is not satisfied, the processing unit 122 may generate a signal that is sent to the signaling unit 142 in response to a determination that the condition is satisfied. Responsive to receiving the signal from the processing unit 122, the signaling unit 142 provides a perceivable indication that an atrial fibrillation episode of the heart 112 is likely to occur.

For example, the signaling unit 142 may include an emitter adapted to emit an audible signal indicating that an atrial fibrillation episode of the heart 112 is likely to occur. Since in the embodiment shown in Figure 5, the signaling unit 142 is disposed within the enclosure 120 which is implanted in the patient, the emitted audible signal is sufficiently loud to be heard by

the patient or any individual proximate to the patient. In other embodiments, the signaling unit 142 may be external to the patient. For instance, the signaling unit 142 may be implemented as a unit worn by the patient (e.g. a bracelet or wrist watch) or as a static or mobile notification device (including a mobile communication device). In such embodiments, the processing unit 122 may be operative to wirelessly send a signal to the signaling unit 142 using, for example, radio frequency (RF) technology. In addition, in embodiments in which the signaling unit 142 is external to the patient, the signaling unit 142 may include a display or other visual indicator for providing a visual indication regarding the likelihood of occurrence of an atrial fibrillation episode.

Advantageously, the perceivable indication of likely occurrence of atrial fibrillation of the heart 112 enables the patient or any individual perceiving the indication to become aware of the risk of atrial fibrillation and to take precautionary actions. For instance, the patient may alter his or her ongoing activities (e.g. stop driving or stop exercising) until the indication of likely occurrence of atrial fibrillation of the heart 1 12 is no longer present. Furthermore, the patient may proceed to take prescribed drugs in an attempt to prevent occurrence of atrial fibrillation. The prescribed drugs may be any of the drugs previously mentioned. Thus, the signaling unit 142 prompting the patient to take the necessary drugs is such that a drug administration unit may be omitted from the treatment unit 140. In fact, in some embodiments, the implantable apparatus 1 10 may comprise the signaling unit 142 and not include the treatment unit 140, in which case treatment provided to the patient is based solely on detection of the signal provided by the signaling unit 142.

Those skilled in the art will appreciate that in some embodiments, certain functionality of the processing unit 22 or 122 may be implemented as pre-programmed hardware or firmware elements (e.g., application specific integrated circuits (ASICs), electrically erasable programmable read-only memories (EEPROMs), etc.), or other related components. In other embodiments, the processing unit 22 or 122 may comprise an arithmetic and logic unit (ALU) having access to a code memory (not shown) which stores program instructions for the operation of the ALU in order to implement the functionality and execute the various processes and functions described above. The program instructions may be stored on a medium which is fixed, tangible and readable directly by the processing unit 22 or 122, (e.g.,

removable diskette, CD-ROM, ROM, or fixed disk), or the program instructions may be stored remotely but transmittable to the processing unit 22 or 122 via a modem or other interface device (e.g., a communications adapter) connected to a network over a transmission medium. The transmission medium may be either a tangible medium (e.g.. optical or analog communications lines) or a medium implemented using wireless techniques (e.g., microwave, infrared or other transmission schemes).

Although various embodiments have been illustrated, this was for the purpose of describing, but not limiting, the invention. Various modifications will become apparent to those skilled in the art and are within the scope of the present invention, which is defined more particularly by the attached claims.