Apparatus and methods for prediction of repeat ablation efficacy
An apparatus and method for prediction of pulmonary vein reconnection is disclosed. The apparatus includes an electrocardiogram device, at least a processor, and a memory communicatively connected to the at least a processor, wherein the memory contains instructions configuring the at least a processor to generate ablation evaluation training data, train an ablation evaluation machine-learning model using the ablation evaluation training data, receive, from the electrocardiogram device, the electrocardiogram data, and generate, using the ablation evaluation machine-learning model, ablation evaluation data of the patient, wherein generating the ablation evaluation data of the patient includes inputting, into the ablation evaluation machine-learning model, the electrocardiogram data and receiving as output, from the ablation evaluation machine-learning model, the ablation evaluation data of the patient.
1. An apparatus for prediction of repeat ablation efficacy, the apparatus comprising:
an electrocardiogram device, wherein the electrocardiogram device is configured to detect post-ablation arrhythmic electrocardiogram (ECG) data representative of a post-ablation arrhythmia of a patient who has previously undergone an ablation procedure at a primary ablation target;
at least a processor; and
a memory communicatively connected to the at least a processor, wherein the memory contains instructions configuring the at least a processor to:
receive, from the electrocardiogram device, the post-ablation arrhythmic ECG data;
generate, using a repeat-ablation efficacy machine-learning model, repeat-ablation efficacy data representing predicted efficacy of a repeat ablation procedure to resolve an arrhythmia of the patient, wherein generating the repeat-ablation efficacy data of the patient comprises:
iteratively training the repeat-ablation efficacy machine-learning model using training data, wherein ablation evaluation machine-learning model is trained and retrained until it surpasses an accuracy threshold, wherein new patient data is periodically integrated into the training data;
inputting, into the repeat-ablation efficacy machine-learning model, the post-ablation arrhythmic ECG data; and
receiving as output, from the repeat-ablation efficacy machine-learning model, the repeat-ablation efficacy data;
generate, using a treatment machine-learning model, a treatment recommendation based on the repeat-ablation efficacy data, the treatment recommendation comprising performing a repeat ablation procedure at a secondary ablation target on the patient; and
transmit, for display, the repeat-ablation efficacy data; and
an ablation device configured to perform the repeat ablation procedure at the secondary ablation target on the patient according to the generated treatment recommendation to increase a likelihood of success of the repeat ablation procedure.
2. The apparatus of claim 1 , wherein the repeat-ablation efficacy data comprises a predicted chance of pulmonary vein reconnection.
3. The apparatus of claim 1 , wherein generating the ablation training data from the subset of patient health records comprises de-identifying the subset of patient health records.
4. The apparatus of claim 1 , wherein the repeat-ablation efficacy data comprises a quantitative value or a classification.
5. The apparatus of claim 1 , wherein determining the treatment recommendation comprises generating, using the treatment machine-learning model, the treatment recommendation as a function of the post-ablation arrhythmic ECG electrocardiogram-data.
6. The apparatus of claim 5 , wherein determining a treatment recommendation comprises:
receiving treatment training data, wherein the treatment training data comprises historical ECG data correlated to historical treatment data;
training the treatment machine-learning model using the treatment training data; and
generating the treatment recommendation using the trained treatment machine-learning model.
7. The apparatus of claim 5 , wherein:
the apparatus further comprises a display device; and
the memory contains instructions further configuring the at least a processor to display the treatment recommendation using the display device.
8. The apparatus of claim 5 , wherein the memory contains instructions further configuring the at least a processor to:
receive a post-recommended treatment ECG and recommended treatment outcome data; and
retrain the treatment machine-learning model as a function of the post-recommended treatment ECG and the recommended treatment outcome data.
9. The apparatus of claim 1 , wherein:
the memory further comprises instructions configuring the at least a processor to receive prior-procedure intracardiac echocardiogram data; and
generating the repeat-ablation efficacy data of the patient further comprises inputting, into the repeat-ablation efficacy machine-learning model, the prior-procedure intracardiac echocardiogram data.
10. The apparatus of claim 1 , wherein:
the memory further comprises instructions configuring the at least a processor to receive prior-procedure ablation parameter data; and
generating the repeat-ablation efficacy data of the patient further comprises inputting, into the repeat-ablation efficacy machine-learning model, the prior-procedure ablation parameter data.
11. A method for prediction of repeat ablation efficacy, the method comprising:
detecting, using an electrocardiogram device, post-ablation arrhythmic electrocardiogram (ECG) data representative of a post-ablation arrhythmia of a patient who has previously undergone an ablation procedure at a primary ablation target;
receiving, by at least a processor, from the electrocardiogram device, the post-ablation arrhythmia ECG data;
generating, by the at least a processor and using a repeat-ablation efficacy machine-learning model, repeat-ablation efficacy data representing predicted efficacy of a repeat ablation procedure to resolve an arrhythmia of the patient, wherein generating the repeat-ablation efficacy data of the patient comprises:
iteratively training the repeat-ablation efficacy machine-learning model using the training data, wherein ablation evaluation machine-learning model is trained and retrained until it surpasses an accuracy threshold, wherein new patient data is periodically integrated into the training data;
inputting, into the repeat-ablation efficacy machine-learning model, the post-ablation arrhythmic ECG data; and
receiving as output, from the repeat-ablation efficacy machine-learning model, the repeat-ablation efficacy data;
transmitting, by the at least a processor, for display, the repeat-ablation efficacy data;
generating, using a treatment machine-learning model, a treatment recommendation based on the repeat-ablation efficacy data, the treatment recommendation comprising performing a repeat ablation procedure at a secondary ablation target on the patient; and
performing the repeat ablation procedure at the secondary ablation target on the patient according to the generated treatment recommendation to increase a likelihood of success of the repeat ablation procedure.
12. The method of claim 11 , wherein the repeat-ablation efficacy data comprises a predicted chance of pulmonary vein reconnection.
13. The method of claim 11 , wherein generating the ablation training data from the subset of patient health records comprises de-identifying the subset of patient health records.
14. The method of claim 11 , wherein the repeat-ablation efficacy data comprises a quantitative value or a classification.
15. The method of claim 11 , wherein determining the treatment recommendation comprises generating, using the treatment machine-learning model, the treatment recommendation as a function of the post-ablation arrhythmic ECG data.
16. The method of claim 15 , wherein determining a treatment recommendation comprises:
receiving treatment training data, wherein the treatment training data comprises historical ECG data correlated to historical treatment data;
training the treatment machine-learning model using the treatment training data; and
generating the treatment recommendation using the trained treatment machine-learning model.
17. The method of claim 15 further comprising displaying, using a display device, the treatment recommendation.
18. The method of claim 15 , further comprising:
receiving, by the at least a processor, a post-recommended treatment ECG and recommended treatment outcome data; and
retraining, by the at least a processor, the treatment machine-learning model as a function of the post-recommended treatment ECG and the recommended treatment outcome data.
19. The method of claim 1 , wherein:
the method further comprises receiving, by the at least a processor, prior-procedure intracardiac echocardiogram data; and
generating the repeat-ablation efficacy data of the patient further comprises inputting, into the repeat-ablation efficacy machine-learning model, the prior-procedure intracardiac echocardiogram data.
20. The method of claim 11 , wherein:
the method further comprises receiving, by the at least a processor, prior-procedure ablation parameter data; and
generating the repeat-ablation efficacy data of the patient further comprises inputting, into the repeat-ablation efficacy machine-learning model, the prior-procedure ablation parameter data.