Estimating the prevalence of activation patterns in data segments during electrophysiology mapping

1. A method for diagnosing and treating pathologies in a heart, the method comprising:

using a system including a mapping probe and a processing system for:

sensing activation signals of physiological activity in the heart with the mapping probe including a plurality of mapping electrodes disposed in the heart, each of the plurality of mapping electrodes having an electrode location;

identifying patterns among the sensed activation signals representing activation propagation;

generating a similarity measure between each unique pair of identified patterns;

classifying the patterns into groups based on the similarity measure;

determining a plurality of characteristic representations, each of the plurality of characteristic representations corresponding to a unique one of the groups and comprising a single numerical representation that summarizes the patterns in the corresponding group; and

displaying at least one of the plurality of characteristic representations to aid in visualization;

identifying at least one site in the heart having a pathology based on the at least one of the plurality of characteristic representations displayed; and

treating myocardial tissue at or near the at least one site in the heart to treat the pathology.

2. The method according to claim 1 , wherein:

displaying the at least one of the plurality of characteristic representations comprises displaying, for each group, a characteristic pattern corresponding to the group and prevalence information associated with the characteristic pattern.

3. The method according to claim 2 , wherein the characteristic representation includes at least one of a mean, variance, covariance, standard deviation, median, and prevalence.

4. The method according to claim 1 , wherein identifying patterns further includes generating a pattern map for each sensed activation signal, each pattern map having at least one of a vector field map that represents a direction and magnitude of activation signal propagation, a voltage propagation map that represents a direction and magnitude of voltage propagation, a phase propagation map that represents a direction and magnitude of phase propagation, and an action potential duration map that represents a duration of an action potential.

5. The method according to claim 1 , wherein the patterns classified into groups are compared with at least one pattern template for each of the groups.

6. The method of claim 1 , wherein identifying patterns further includes:

identifying unclassifiable patterns that are not classifiable into any groups of similar patterns; and

determining a measure of randomness based on the unclassifiable patterns.

7. The method according to claim 1 , wherein generating the similarity measure further includes generating a similarity matrix including the patterns, each entry of the similarity matrix representing the similarity measure for each unique pair of identified patterns generated based on a correlation of the corresponding patterns.

8. The method according to claim 1 , wherein classifying the patterns further includes:

determining a correlation coefficient for each unique pair of patterns; and

classifying the patterns into distinct groups based on a percentage of patterns among each group having a particular correlation coefficient.

9. A method for diagnosing and treating pathologies in a heart, comprising:

using a system including a mapping probe and a processing system for:

sensing activation signals of cardiac activity with the mapping probe including a plurality of mapping electrodes disposed in the heart, each of the plurality of mapping electrodes having an electrode location;

identifying patterns among the sensed activation signals;

generating a similarity measure between each of unique pairs of identified patterns;

classifying the patterns into groups based on the similarity measure;

determining a characteristic representation for each group of the groups, wherein each characteristic representation comprises a single numerical representation that summarizes the patterns in the corresponding group; and

displaying at least one characteristic representation determined for the groups to aid in visualization;

identifying at least one site in the heart having a pathology based on the at least one of the plurality of characteristic representations displayed; and

treating myocardial tissue at or near the at least one site in the heart to treat the pathology.

10. The method according to claim 9 , wherein the characteristic representation includes at least one of a mean, variance, covariance, standard deviation, median, and a prevalence of the pattern.

11. The method according to claim 9 , further comprising generating a plurality of pattern maps for each activation signal, each pattern map having at least one of a vector field map which represents a direction and a magnitude of an activation signal propagation, a voltage propagation map which representation a direction and a magnitude of voltage propagation, a phase propagation map which represents a direction and a magnitude of phase propagation, and an action potential duration map which represents a duration of an action potential.

12. The method according to claim 11 , wherein generating the plurality of pattern maps further includes:

identifying unclassifiable pattern maps that are not classifiable into any groups of similar patterns; and

determining a measure of randomness based on the unclassifiable pattern maps.

13. The method according to claim 9 , wherein generating the similarity measure further comprises generating a similarity matrix including the patterns, each entry of the similarity matrix representing the similarity measure for each unique pair of identified patterns generated based on a correlation of the corresponding patterns.

14. The method according to claim 9 , wherein classifying the patterns further comprises:

determining a correlation coefficient for each unique pair of patterns; and

classifying the patterns into distinct groups based on a percentage of patterns among each group having a particular correlation coefficient.