Identify ablation pattern for use in an ablation
Systems are provided for generating data representing electromagnetic states of a heart for medical, scientific, research, and/or engineering purposes. The systems generate the data based on source configurations such as dimensions of, and scar or fibrosis or pro-arrhythmic substrate location within, a heart and a computational model of the electromagnetic output of the heart. The systems may dynamically generate the source configurations to provide representative source configurations that may be found in a population. For each source configuration of the electromagnetic source, the systems run a simulation of the functioning of the heart to generate modeled electromagnetic output (e.g., an electromagnetic mesh for each simulation step with a voltage at each point of the electromagnetic mesh) for that source configuration. The systems may generate a cardiogram for each source configuration from the modeled electromagnetic output of that source configuration for use in predicting the source location of an arrhythmia.
1. A method performed by one or more computing systems for generating a mapping function to map patient information of a patient to an ablation pattern for treating an arrhythmia of the patient, the method comprising:
accessing simulated source configurations and simulated ablation patterns of ablation pattern simulations, each simulated source configuration including a simulated source location;
for each ablation pattern simulation,
generating a simulated feature vector that includes features based on the simulated source configuration of that ablation pattern simulation but not based on the simulated source location of that ablation pattern simulation and a feature based on a simulated cardiogram generated from a non-ablation pattern simulation based on the simulated source configuration but not based on a simulated ablation pattern; and
labeling the simulated feature vector with the simulated source location and the simulated ablation pattern of that ablation pattern simulation; and
training a classifier using the feature vectors and labels as training data.
2. The method of claim 1 further comprising:
generating a patient feature vector with features based on a patient source configuration of the patient but not based on a source location and a feature based on a patient cardiogram of the patient; and
applying the classifier to the patient feature vector to identify a source location and an ablation pattern for treating the patient.
3. The method of claim 2 further comprising outputting an indication of the identified source location and the identified ablation pattern for treating the patient.
4. The method of claim 3 wherein the patient is treated using stereotactic body radiation therapy.
5. A method of treating a heart of a patient, comprising:
identifying a source location within the heart of the patient and an ablation pattern, the identifying based on ablation pattern simulations, each ablation pattern simulation based on a source location and an ablation pattern wherein the identifying is based on a patient cardiogram of the patient;
positioning a neuromodulation device at a target site of the heart based on the identified source location; and
applying energy based on the identified ablation pattern to the target site of the heart using the neuromodulation device.
6. The method of claim 5 wherein the neuromodulation device is an ablation catheter and applying energy to the target site comprises applying radiofrequency energy that ablates neural or myocardial structures at the target site.
7. The method of claim 5 wherein the neuromodulation device is a cryoablation catheter and applying energy to the target site comprises applying cooling energy that cryoablates neural or myocardial structures at the target site.
8. The method of claim 5 wherein the neuromodulation device comprises an implantable pulse generator and applying energy to the target site comprises applying electrical energy to neural or myocardial structures at the target site.
9. The method of claim 5 wherein the neuromodulation device comprises a stereotactic body radiation therapy device and applying energy to the target site comprises applying radiation to neural or myocardial structures at the target site.
10. A method of treating a patient, comprising:
identifying a source location of an arrhythmia within the heart of the patient and an ablation pattern by directing a computing system to:
identify a simulated cardiogram that is similar to a patient cardiogram of the patient, the simulated cardiogram generated from a non-ablation pattern simulation of a model library of non-ablation pattern simulations, each non-ablation pattern simulation generated based on a computational model of the heart and a simulated heart configuration that includes a simulated source location, each heart configuration of a simulation mapped to a simulated cardiogram generated based on the simulation; and
identify a simulated source location and an ablation pattern based on the identified simulated cardiogram, the simulated source location and the ablation pattern used in generating an ablation pattern simulation, the ablation pattern simulation generated based on the computational model of the heart, based on the heart configuration that includes the simulated source location, and based on the ablation pattern;
positioning a neuromodulation device at a target site based on the identified simulated source location; and
applying energy based on the identified ablation pattern using the positioned neuromodulation device.
11. The method of claim 10 wherein the neuromodulation device is a stereotactic body radiation therapy device.
12. The method of claim 11 wherein the positioning includes a computing system sending instructions to the stereotactic body radiation therapy device.
13. The method of claim 11 further comprising identifying anatomical parameters of the patient from images collected by the stereotactic body radiation therapy device.
14. The method of claim 10 wherein the neuromodulation device is an ablation catheter and applying energy comprises applying radiofrequency energy that ablates neural or myocardial structures at the target site.
15. The method of claim 10 wherein the neuromodulation device is a cryoablation catheter and applying energy comprises applying cooling energy that cryoablates neural or myocardial structures at the target site.
16. The method of claim 10 wherein the neuromodulation device comprises an implantable pulse generator and applying energy comprises applying electrical energy to neural or myocardial structures at the target site.
17. A method of treating a patient, comprising:
identifying a source location of an arrhythmia within the heart of the patient by directing a computing system to:
identify a simulated cardiogram that is similar to a patient cardiogram of the patient, the simulated cardiogram generated from a simulation of a model library of simulations, each simulation generated based on a computational model of the heart and a simulated heart configuration that includes a simulated source location, each heart configuration of a simulation mapped to a simulated cardiogram generated based on that simulation; and
identify a simulated source location of a simulated heart configuration used in generating the simulation from which the identified simulated cardiogram was generated;
positioning a neuromodulation device at a target site based on the identified simulated source location; and
applying energy to the target site using the positioned neuromodulation device.
18. A method performed by one or more computing systems for generating a mapping function to map patient information of a patient to an ablation pattern for treating an arrhythmia of the patient, the method comprising:
accessing mappings of derived electromagnetic (EM) data that is derived from EM output of a heart to a source location and an ablation pattern that is identified to be successful in stopping an arrhythmia associated with that source location of a heart that generates that EM output;
for each mapping,
generating a feature vector with one or more features, a feature being based on the derived EM data of the mapping; and
generating a label for the feature vector based on the source location and the ablation pattern of the mapping; and
training the mapping function using the feature vectors and labels as training data.
19. The method of claim 18 wherein the mapping function is selected from the group consisting of a fully-connected neural network, a convolutional neural network, a recurrent neural network, autoencoder neural network, a restricted Boltzmann machine, a support vector machine, and a Bayesian classifier.
20. The method of claim 18 wherein the arrhythmia is selected from the group consisting of sinus tachycardia, ectopic atrial rhythm, junctional rhythm, ventricular escape rhythm, atrial fibrillation, ventricular fibrillation, focal atrial tachycardia, atrial microreentry, ventricular tachycardia, atrial flutter, premature ventricular complexes, premature atrial complexes, atrioventricular nodal reentrant tachycardia, atrioventricular reentrant tachycardia, permanent junctional reciprocating tachycardia, junctional tachycardia.
21. The method of claim 18 further comprising:
generating a patient feature vector with one or more features based on a patient derived EM data of a patient; and
applying the mapping function to the patient feature vector to identify a source location and an ablation pattern for treating the patient.
22. The method of claim 21 further comprising outputting an indication of the identified source location and the identified ablation pattern for treating the patient.
23. The method of claim 22 wherein the patient is treated using stereotactic body radiation therapy.
24. The method of claim 22 wherein the indication is output to an ablation device.
25. The method of claim 18 wherein the derived EM data is further mapped to a source configuration of a heart.
26. The method of claim 18 wherein the mappings are generated based on ablation pattern simulations in which a simulated ablation pattern is successful at treating an arrhythmia associated with a simulated source location.
27. The method of claim 18 wherein the mappings are generated based on non-ablation pattern simulations the simulate an arrhythmia based on a simulated source location but not based on a simulated ablation pattern associated with that simulated source location.
28. One or more computing systems for generating a mapping function to map patient information of a patient to an ablation pattern for treating an arrhythmia of the patient, the one or more computing systems comprising:
one or more computer-readable storage mediums that store:
mappings of a source configuration of a heart to a source location and an ablation pattern that is identified as being successful for stopping an arrhythmia associated a heart with that source configuration; and
computer-executable instructions for controlling the one or more computing systems to:
for each mapping,
generate a feature vector with one or more features, a feature being based on the source configuration of that the mapping; and
generate a label for the feature vector based on the source location and the ablation pattern of the mapping; and
train the mapping function using the feature vectors and labels as training data; and
one or more processors for controlling the one or more computing systems to execute the one or more computer-executable instructions.
29. The one or more computing systems of claim 28 wherein the mapping function is selected from the group consisting of a fully-connected neural network, a convolutional neural network, a recurrent neural network, autoencoder neural network, a restricted Boltzmann machine, a support vector machine, and a Bayesian classifier.
30. The one or more computing systems of claim 28 wherein at least one of the computing systems is cloud-based computing system.
31. The one or more computing systems of claim 28 wherein the computer-executable instructions include additional instructions to:
generate a patient feature vector with one or more features based on a source configuration of a heart of a patient; and
apply the mapping function to the patient feature vector to identify a source location and an ablation pattern for treating the patient.
32. The one or more computing systems of claim 31 wherein at least one of the computing systems is a cloud-based a computing system.
33. The one or more computing systems of claim 31 wherein the computer-executable instructions further control the one or more computing systems to output an indication of the identified source location and the identified ablation pattern for treating the patient.
34. The one or more computing systems of claim 33 wherein the indication is output to an ablation device.
35. The one or more computing systems of claim 28 wherein a source configuration is mapped to a cardiogram.
36. The one or more computing systems of claim 35 wherein the computer-executable instructions include additional instructions to:
access a patient cardiogram of a patient;
identify a source configuration to which a cardiogram that is similar to the patient cardiogram is mapped;
generate a patient feature vector with one or more features based on the identified source configuration of a heart of a patient; and
apply the mapping function to the patient feature vector to identify a source location and an ablation pattern for treating the patient.
37. The one or more computing systems of claim 28 wherein mappings are generated based on ablation pattern simulations.
38. The one or more computing systems of claim 37 wherein the mappings are further generated based on non-ablation pattern simulations.