Method and system for analyzing heart rhythms
AssigneeImplicity
InventorsEliot Crespin, Arnaud Rosier, David Perlmutter
Life of this patent
Estimating the term…
Abstract
A method and a system for analyzing electrocardiographic segments previously derived from a cardiac device so as to help to discriminate true positives episodes, including abnormal heart rhythms, from false positives episodes, including normal heart rhythms. Each episode received includes at least one segment of electrocardiographic signal, and each segment is segmented into sub-segments. Score vectors are obtained for each sub-segment to classify the episode so as to discriminate true positive episodes from false positive episodes, and the classification results, which include at least the true positive episodes, are output.
Claims20
A computer-implemented method for analyzing electrocardiographic episodes previously derived from a cardiac connected device so as to help to discriminate true positives episodes, comprising abnormal heart rhythms, from false positives episodes, comprising normal heart rhythms, said method comprising:
- receiving the episodes, each episode comprising at least one segment of electrocardiographic signal;
- for each segment in one episode:
- identifying R waves in the segment using at least one algorithm and using said R waves for calculating at least one feature of the segment;
- segmenting the segment into at least two sub-segments;
- for each sub-segment:
- using the R waves being comprised in said sub-segment for calculating at least one feature of the sub-segment;
- providing as input to a machine learning algorithm the at least one feature of the sub-segment and the at least one feature of the segment and obtaining as output a score vector, the machine learning algorithm being configured to output said score vector;
- using the score vectors obtained for each sub-segment in one episode to classify the episode so as to discriminate true positives episodes from false positives episodes;
- and
- outputting the classification results, comprising at least the true positives episodes.
A system for analyzing electrocardiographic episodes previously derived from a cardiac device so as to discriminate true positives episodes, comprising abnormal heart rhythms, from false positives episodes, comprising normal heart rhythms, said system comprising:
- at least one input adapted to receive the episodes, each episode comprising at least one segment of electrocardiographic signal;
- at least one processor configured to:
- for each segment in one episode:
- identifying R waves in the segment using at least one algorithm and using said R waves for calculating at least one feature of the segment;
- segmenting the segment into at least two sub-segments;
- for each sub-segment:
- using the R waves being comprised in the sub-segment for calculating at least one feature of the sub-segment;
- providing as input to a machine learning algorithm the at least one feature of the sub-segment and the at least one feature of the segment, wherein the machine learning algorithm is configured to output a score vector;
- using the score vectors obtained for each sub-segment in one episode to classify the episode so as to discriminate true positives episodes from false positives episodes;
- and
- at least one output adapted to provide the true positives episodes.
Dependencies are inferred from the claim text.