A machine learning approach using Symmetric Projection Attractor Reconstruction combined with standard ECG features achieved 96% accuracy in detecting the Scn5a+/- mutation from short mouse ECG signals.
The novel Symmetric Projection Attractor Reconstruction method combined with machine learning can accurately detect Scn5a+/- mutations from short ECG signals in a mouse model.
BackgroundLife-threatening arrhythmias resulting from genetic mutations are often missed in current electrocardiogram (ECG) analysis. We combined a new method for ECG analysis that uses all the waveform data with machine learning to improve detection of such mutations from short ECG signals in a mouse model.ObjectiveWe sought to detect consequences of Na+ channel deficiencies known to compromise action potential conduction in comparisons of Scn5a+/- mutant and wild-type mice using short ECG signals, examining novel and standard features derived from lead I and II ECG recordings by machine learning algorithms.MethodsLead I and II ECG signals from anesthetized wild-type and Scn5a+/- mutant mice of length 130 seconds were analyzed by extracting various groups of features, which were used by machine learning to classify the mice as wild-type or mutant. The features used were standard ECG intervals and amplitudes, as well as features derived from attractors generated using the novel Symmetric Projection Attractor Reconstruction method, which reformulates the whole signal as a bounded, symmetric 2-dimensional attractor. All the features were also combined as a single feature group.ResultsClassification of genotype using the attractor features gave higher accuracy than using either the ECG intervals or the intervals and amplitudes. However, the highest accuracy (96%) was obtained using all the features. Accuracies for different subgroups of the data were obtained and compared.ConclusionDetection of the Scn5a+/- mutation from short mouse ECG signals with high accuracy is possible using our Symmetric Projection Attractor Reconstruction method. Life-threatening arrhythmias resulting from genetic mutations are often missed in current electrocardiogram (ECG) analysis. We combined a new method for ECG analysis that uses all the waveform data with machine learning to improve detection of such mutations from short ECG signals in a mouse model. We sought to detect consequences of Na+ channel deficiencies known to compromise action potential conduction in comparisons of Scn5a+/- mutant and wild-type mice using short ECG signals, examining novel and standard features derived from lead I and II ECG recordings by machine learning algorithms. Lead I and II ECG signals from anesthetized wild-type and Scn5a+/- mutant mice of length 130 seconds were analyzed by extracting various groups of features, which were used by machine learning to classify the mice as wild-type or mutant. The features used were standard ECG intervals and amplitudes, as well as features derived from attractors generated using the novel Symmetric Projection Attractor Reconstruction method, which reformulates the whole signal as a bounded, symmetric 2-dimensional attractor. All the features were also combined as a single feature group. Classification of genotype using the attractor features gave higher accuracy than using either the ECG intervals or the intervals and amplitudes. However, the highest accuracy (96%) was obtained using all the features. Accuracies for different subgroups of the data were obtained and compared. Detection of the Scn5a+/- mutation from short mouse ECG signals with high accuracy is possible using our Symmetric Projection Attractor Reconstruction method.
Bonet-Luz et al. (Thu,) conducted a other in Scn5a+/- genetic mutation (Brugada syndrome). Symmetric Projection Attractor Reconstruction method with machine learning vs. Standard ECG intervals and amplitudes was evaluated on Classification of genotype (wild-type vs mutant). A machine learning approach using Symmetric Projection Attractor Reconstruction combined with standard ECG features achieved 96% accuracy in detecting the Scn5a+/- mutation from short mouse ECG signals.