Inflection point analysis: A machine learning approach for extraction of IEGM active intervals during atrial fibrillation

Delineation of intracavitary electrograms for the automatic quantification of decrement-evoked potentials in the coronary sinus with deep-learning techniques

Cardiac arrhythmias cause depolarization waves to conduct unevenly on the myocardial surface, potentially delaying local components with respect to a previous beat when stimulated at faster frequencies. Despite the diagnostic value of localizing the distinct local electrocardiogram (EGM) components for identifying regions with decrement-evoked potentials (DEEPs), current software solutions do not perform automatic signal quantification. Electrophysiologists must manually measure distances on the EGM signals to assess the existence of DEEPs during pacing or extra-stimuli protocols. In this work, we present a deep learning (DL)-based algorithm to identify decrement in atrial components (measured in the coronary sinus) with respect to their ventricular counterparts from EGM signals, for disambiguating between accessory pathways (APs) and atrioventricular re-entrant tachycardias (AVRTs). Several U-Net and W-Net neural networks with different configurations were trained on a private dataset of signals from the coronary sinus (312 EGM recordings from 77 patients who underwent AP or AVRT ablation). A second, separate dataset was annotated for clinical validation, with clinical labels associated to EGM fragments in which decremental conduction was elucidated. To alleviate data scarcity, a synthetic data augmentation method was developed for generating EGM recordings. Moreover, two novel loss functions were developed to minimize false negatives and delineation errors. Finally, the addition of self-attention mechanisms and their effect on model performance was explored. The best performing model was a W-Net model with 6 levels, optimized solely with the Dice loss. The model obtained precisions of 91.28%, 77.78% and of 100.0%, and recalls of 94.86%, 95.25% and 100.0% for localizing local field, far field activations, and extra-stimuli, respectively. The clinical validation model demonstrated good overall agreement with respect to the evaluation of decremental properties. When compared to the criteria of electrophysiologists, the automatic exclusion step reached a sensitivity of 87.06% and a specificity of 97.03%. Out of the non-excluded signals, a sensitivity of 96.77% and a specificity of 95.24% was obtained for classifying them into decremental and non-decremental potentials. Current results show great promise while being, to the best of our knowledge, the first tool in the literature allowing the delineation of all local components present in an EGM recording. This is of capital importance at advancing processing for cardiac electrophysiological procedures and reducing intervention times, as many diagnosis procedures are performed by comparing segments or late potentials in subsequent cardiac cycles.

Machine learning for atrial fibrillation risk prediction in patients with sleep apnea and coronary artery disease

Background

Patients with sleep apnea (SA) and coronary artery disease (CAD) are at higher risk of atrial fibrillation (AF) than the general population. Our objectives were: to evaluate the role of CAD and SA in determining AF risk through cluster and survival analysis, and to develop a risk model for predicting AF.

Methods

Electronic medical record (EMR) database from 22,302 individuals including 10,202 individuals with AF, CAD, and SA, and 12,100 individuals without these diseases were analyzed using K-means clustering technique; k-nearest neighbor (kNN) algorithm and survival analysis. Age, sex, and diseases developed for each individual during 9 years were used for cluster and survival analysis.

Results

The risk models for AF, CAD, and SA were identified with high accuracy and sensitivity (0.98). Cluster analysis showed that CAD and high blood pressure (HBP) are the most prevalent diseases in the AF group, HBP is the most prevalent disease in CAD; and HBP and CAD are the most prevalent diseases in the SA group. Survival analysis demonstrated that individuals with HBP, CAD, and SA had a 1.5-fold increased risk of developing AF [hazard ratio (HR): 1.49, 95% CI: 1.18-1.87, p = 0.0041; HR: 1.46, 95% CI: 1.09-1.96, p = 0.01; HR: 1.54, 95% CI: 1.22-1.94, p = 0.0039, respectively] and individuals with chronic kidney disease (CKD) developed AF approximately 50% earlier than patients without these comorbidities in a period of 7 years (HR: 3.36, 95% CI: 1.46-7.73, p = 0.0023). Comorbidities that contributed to develop AF earlier in females compared to males in the group of 50-64 years were HBP (HR: 3.75 95% CI: 1.08-13, p = 0.04) CAD and SA in the group of 60-75 years were (HR: 2.4 95% CI: 1.18-4.86, p = 0.02; HR: 2.51, 95% CI: 1.14-5.52, p = 0.02, respectively).

Conclusion

Machine learning based algorithms demonstrated that CAD, SA, HBP, and CKD are significant risk factors for developing AF in a Latin-American population.

Domain Heterogeneity in Radiofrequency Therapies for Pain Relief: A Computational Study with Coupled Models

The objective of the current research work is to study the differences between the predicted ablation volume in homogeneous and heterogeneous models of typical radiofrequency (RF) procedures for pain relief. A three-dimensional computational domain comprising of the realistic anatomy of the target tissue was considered in the present study. A comparative analysis was conducted for three different scenarios: (a) a completely homogeneous domain comprising of only muscle tissue, (b) a heterogeneous domain comprising of nerve and muscle tissues, and (c) a heterogeneous domain comprising of bone, nerve and muscle tissues. Finite-element-based simulations were performed to compute the temperature and electrical field distribution during conventional RF procedures for treating pain, and exemplified here for the continuous case. The predicted results reveal that the consideration of heterogeneity within the computational domain results in distorted electric field distribution and leads to a significant reduction in the attained ablation volume during the continuous RF application for pain relief. The findings of this study could provide first-hand quantitative information to clinical practitioners about the impact of such heterogeneities on the efficacy of RF procedures, thereby assisting them in developing standardized optimal protocols for different cases of interest.

Thermal ablation of biological tissues in disease treatment: A review of computational models and future directions

Percutaneous thermal ablation has proven to be an effective modality for treating both benign and malignant tumours in various tissues. Among these modalities, radiofrequency ablation (RFA) is the most promising and widely adopted approach that has been extensively studied in the past decades. Microwave ablation (MWA) is a newly emerging modality that is gaining rapid momentum due to its capability of inducing rapid heating and attaining larger ablation volumes, and its lesser susceptibility to the heat sink effects as compared to RFA. Although the goal of both these therapies is to attain cell death in the target tissue by virtue of heating above 50°C, their underlying mechanism of action and principles greatly differs. Computational modelling is a powerful tool for studying the effect of electromagnetic interactions within the biological tissues and predicting the treatment outcomes during thermal ablative therapies. Such a priori estimation can assist the clinical practitioners during treatment planning with the goal of attaining successful tumour destruction and preservation of the surrounding healthy tissue and critical structures. This review provides current state-of-the-art developments and associated challenges in the computational modelling of thermal ablative techniques, viz., RFA and MWA, as well as touch upon several promising avenues in the modelling of laser ablation, nanoparticles assisted magnetic hyperthermia and non-invasive RFA. The application of RFA in pain relief has been extensively reviewed from modelling point of view. Additionally, future directions have also been provided to improve these models for their successful translation and integration into the hospital work flow.

Temporal irregularity quantification and mapping of optical action potentials using wave morphology similarity

Background

Cardiac optical mapping enables direct and high spatio-temporal resolution recording of action potential (AP) morphology. Temporal alterations in AP morphology are both predictive and consequent of arrhythmia. Here we sought to test if methods that quantify regularity of recorded waveforms could be applied to detect and quantify periods of temporal instability in optical mapping datasets in a semi-automated, user-unbiased manner.

Methods and results

We developed, tested and applied algorithms to quantify optical wave similarity (OWS) to study morphological temporal similarity of optically recorded APs. Unlike other measures (e.g. alternans ratio, beat-to-beat variability, arrhythmia scoring), the quantification of OWS is achieved without a restrictive definition of specific signal points/features and is instead derived by analysing the complete morphology from the entire AP waveform.

Using model datasets, we validated the ability of OWS to measure changes in AP morphology, and tested OWS mapping in guinea pig hearts and mouse atria. OWS successfully detected and measured alterations in temporal regularity in response to several proarrhythmic stimuli, including alterations in pacing frequency, premature contractions, alternans and ventricular fibrillation.

Conclusion

OWS mapping provides an effective measure of temporal regularity that can be applied to optical datasets to detect and quantify temporal alterations in action potential morphology. This methodology provides a new metric for arrhythmia inducibility and scoring in optical mapping datasets.