Sudden cardiac death (SCD) remains a pressing health issue, affecting hundreds of thousands each year globally. The heterogeneity among SCD victims, ranging from individuals with severe heart failure to seemingly healthy individuals, poses a significant challenge for effective risk assessment. Conventional risk stratification, which primarily relies on left ventricular ejection fraction, has resulted in only modest efficacy of implantable cardioverter-defibrillators for SCD prevention. In response, artificial intelligence (AI) holds promise for personalized SCD risk prediction and tailoring preventive strategies to the unique profiles of individual patients. Machine and deep learning algorithms have the capability to learn intricate nonlinear patterns between complex data and defined end points and leverage these to identify subtle indicators and predictors of SCD that may not be apparent through traditional statistical analysis. However, despite the potential of AI to improve SCD risk stratification, there are important limitations that need to be addressed. We aim to provide an overview of the current state-of-the-art of AI prediction models for SCD, highlight the opportunities for these models in clinical practice, and identify the key challenges hindering widespread adoption.

MZH Kolk, S Ruipérez-Campillo, AAM Wilde, RE Knops, SM Narayan, FVY Tjong

Heart Rhythm

06.09.2024

LinkDOI

Background: Segmenting computed tomography (CT) is crucial in various clinical applications, such as tailoring personalized cardiac ablation for managing cardiac arrhythmias. Automating segmentation through machine learning (ML) is hindered by the necessity for large, labeled training data, which can be challenging to obtain. This article proposes a novel approach for automated, robust labeling using domain knowledge to achieve high-performance segmentation by ML from a small training set. The approach, the domain knowledge-encoding (DOKEN) algorithm, reduces the reliance on large training datasets by encoding cardiac geometry while automatically labeling the training set. The method was validated in a hold-out dataset of CT results from an atrial fibrillation (AF) ablation study. Methods: The DOKEN algorithm parses left atrial (LA) structures, extracts “anatomical knowledge” by leveraging digital LA models (available publicly), and then applies this knowledge to achieve high ML segmentation performance with a small number of training samples. The DOKEN-labeled training set was used to train a nnU-Net deep neural network (DNN) model for segmenting cardiac CT in N = 20 patients. Subsequently, the method was tested in a hold-out set with N = 100 patients (five times larger than training set) who underwent AF ablation. Results: The DOKEN algorithm integrated with the nn-Unet model achieved high segmentation performance with few training samples, with a training to test ratio of 1:5. The Dice score of the DOKEN-enhanced model was 96.7% (IQR: 95.3% to 97.7%), with a median error in surface distance of boundaries of 1.51 mm (IQR: 0.72 to 3.12) and a mean centroid–boundary distance of 1.16 mm (95% CI: −4.57 to 6.89), similar to expert results (r = 0.99; p < 0.001). In digital hearts, the novel DOKEN approach segmented the LA structures with a mean difference for the centroid–boundary distances of −0.27 mm (95% CI: −3.87 to 3.33; r = 0.99; p < 0.0001). Conclusions: The proposed novel domain knowledge-encoding algorithm was able to perform the segmentation of six substructures of the LA, reducing the need for large training data sets. The combination of domain knowledge encoding and a machine learning approach could reduce the dependence of ML on large training datasets and could potentially be applied to AF ablation procedures and extended in the future to other imaging, 3D printing, and data science applications.

P Ganesan^*, R Feng^*, B Deb, FVY Tjong, AJ Rogers, S Ruipérez-Campillo, S Somani, Paul Clopton, T Baykaner, M Rodrigo, J Zou, F Haddad, M Zaharia, SM Narayan
^* denotes shared first authorship

Diagnostics

17.07.2024

LinkDOI

The efficacy of an implantable cardioverter-defibrillator (ICD) in patients with a non-ischaemic cardiomyopathy for primary prevention of sudden cardiac death is increasingly debated. We developed a multimodal deep learning model for arrhythmic risk prediction that integrated late gadolinium enhanced (LGE) cardiac magnetic resonance imaging (MRI), electrocardiography (ECG) and clinical data. Short-axis LGE-MRI scans and 12-lead ECGs were retrospectively collected from a cohort of 289 patients prior to ICD implantation, across two tertiary hospitals. A residual variational autoencoder was developed to extract physiological features from LGE-MRI and ECG, and used as inputs for a machine learning model (DEEP RISK) to predict malignant ventricular arrhythmia onset. In the validation cohort, the multimodal DEEP RISK model predicted malignant ventricular arrhythmias with an area under the receiver operating characteristic curve (AUROC) of 0.84 (95% confidence interval (CI) 0.71–0.96), a sensitivity of 0.98 (95% CI 0.75–1.00) and a specificity of 0.73 (95% CI 0.58–0.97). The models trained on individual modalities exhibited lower AUROC values compared to DEEP RISK [MRI branch: 0.80 (95% CI 0.65–0.94), ECG branch: 0.54 (95% CI 0.26–0.82), Clinical branch: 0.64 (95% CI 0.39–0.87)]. These results suggest that a multimodal model achieves high prognostic accuracy in predicting ventricular arrhythmias in a cohort of patients with non-ischaemic systolic heart failure, using data collected prior to ICD implantation.

MZH Kolk, S Ruipérez-Campillo, CP Allaart, AAM Wilde, RE Knops, SM Narayan, FVY Tjong

Nature Scientific Reports

27.06.2024

LinkDOI

Background and Objectives: The extensive collection of electrocardiogram (ECG) recordings stored in paper format has provided opportunities for numerous digitization studies. However, the traditional 10 s 12-lead ECG printout typically splits the ECG signals into four asynchronous sections of 3 leads and 2.5 s each. Since each lead corresponds to different time instants, developing a synchronization method becomes necessary for applications such as vectorcardiogram (VCG) reconstruction. Methods: A beat-level synchronization method has been developed and validated using a dataset of 21,674 signals. This method effectively addresses synchronization distortions caused by RR interval variations and preserves the time lags between R peaks across different leads for each beat. Results: The results demonstrate that the proposed method successfully synchronizes the ECG, allowing a VCG reconstruction with an average Pearson Correlation Coefficient of 0.9815±0.0426. The Normalized Root Mean Squared Error (NRMSE) and Mean Absolute Error (MAE) values for the reconstructed VCG are 0.0248±0.0214 mV and 0.0133±0.0123 mV, respectively. These metrics indicate the reliability of the VCG reconstruction achieved by means of the proposed synchronization method. Conclusions: The synchronization method has demonstrated its robustness and high performance compared to existing techniques in the field. Its effectiveness has been observed across a wide variety of signals, showcasing its applicability in real clinical environments. Moreover, its ability to handle a large number of signals makes it suitable for various applications, including retrospective studies and the development of machine learning methods.

E Ramírez, S Ruipérez-Campillo, F Castells, R Casado-Arroyo, J Millet

Biomedical Signal Processing and Control

01.05.2024

LinkDOI

In the field of cardiac electrophysiology (EP), effectively reducing noise in intra-cardiac signals is crucial for the accurate diagnosis and treatment of arrhythmias and cardiomyopathies. However, traditional noise reduction techniques fall short in addressing the diverse noise patterns from various sources, often non-linear and non-stationary, present in these signals. This work introduces a Variational Autoencoder (VAE) model, aimed at improving the quality of intra-ventricular monophasic action potential (MAP) signal recordings. By constructing representations of clean signals from a dataset of 5706 time series from 42 patients diagnosed with ischemic cardiomyopathy, our approach demonstrates superior denoising performance when compared to conventional filtering methods commonly employed in clinical settings. We assess the effectiveness of our VAE model using various metrics, indicating its superior capability to denoise signals across different noise types, including time-varying non-linear noise frequently found in clinical settings. These results reveal that VAEs can eliminate diverse sources of noise in single beats, outperforming state-of-the-art denoising techniques and potentially improving treatment efficacy in cardiac EP.

S Ruipérez-Campillo, A Ryser, TM Sutter, R Feng, P Ganesan, B Deb, KA Brennan, AJ Rogers, MZH Kolk, FVY Tjong, SM Narayan, JE Vogt

ICLR 2024 - Workshop on Time Series for Healthcare

28.03.2024

Link

High-density multielectrode catheters are becoming increasingly popular in cardiac electrophysiology for advanced characterisation of the cardiac tissue, dueto their potential to identify impaired sites. These are often characterised by abnormal electrical conduction, which may cause locally disorganised propagation wavefronts.To quantify it, a novel heterogeneity parameter based on vector field analysis is proposed, utilising finite differences to measure direction changes between adjacent cliques. The proposed Vector Field Heterogeneity metric has been evaluated on a set of simulations with controlled levels of organisation in vector maps, and a variety of grid sizes. Furthermore, it has been tested on animal experimental models of isolated Langendorff-perfused rabbit hearts. The proposed parameter exhibited superior capturing ability of heterogeneous propagation wavefronts compared to the classical Spatial Inhomogeneity Index, and simulations proved that the metric effectively captures gradual increments in disorganisation in propagation patterns. Notably, it yielded robust and consistent outcomes for 4 × 4 grid sizes, underscoring its suitability for the latest generation of orientation-independent cardiac catheters. Index Terms—Animal experimental models, cardiac signal processing, electrophysiology, high-density electrode catheters, vector field heterogeneity. Impact Statement—The authors introduce the Vector Field Heterogeneity (VFH) metric, which provides a precise evaluation of disorganisation in electrical propagation maps within cardiac tissue, potentially improving the diagnosis and characterisation of electrophysiological conditions.

L Pancorbo^*, S Ruipérez-Campillo^*, A Tormos, A Guill, R Cervigón, A Alberola, FJ Chorro, J Millet, F Castells
^* denotes shared first authorship

IEEE Open Journal of Engineering in Medicine and Biology

23.02.2024

LinkDOI

Background and Objectives: The extensive collection of electrocardiogram (ECG) recordings stored in paper format has provided opportunities for numerous digitization studies. However, the traditional 10 s 12-lead ECG printout typically splits the ECG signals into four asynchronous sections of 3 leads and 2.5 s each. Since each lead corresponds to different time instants, developing a synchronization method becomes necessary for applications such as vectorcardiogram (VCG) reconstruction. Methods: A beat-level synchronization method has been developed and validated using a dataset of 21,674 signals. This method effectively addresses synchronization distortions caused by RR interval variations and preserves the time lags between R peaks across different leads for each beat. Results: The results demonstrate that the proposed method successfully synchronizes the ECG, allowing a VCG reconstruction with an average Pearson Correlation Coefficient of 0.9815±0.0426. The Normalized Root Mean Squared Error (NRMSE) and Mean Absolute Error (MAE) values for the reconstructed VCG are 0.0248±0.0214 mV and 0.0133±0.0123 mV, respectively. These metrics indicate the reliability of the VCG reconstruction achieved by means of the proposed synchronization method. Conclusions: The synchronization method has demonstrated its robustness and high performance compared to existing techniques in the field. Its effectiveness has been observed across a wide variety of signals, showcasing its applicability in real clinical environments. Moreover, its ability to handle a large number of signals makes it suitable for various applications, including retrospective studies and the development of machine learning methods.

E Ramírez, S Ruipérez-Campillo, F Castells, R Casado-Arroyo, J Millet

Biomedical Signal Processing and Control

05.01.2024

LinkDOI

Background Risk stratification for ventricular arrhythmias currently relies on static measurements that fail to adequately capture dynamic interactions between arrhythmic substrate and triggers over time. We trained and internally validated a dynamic machine learning (ML) model and neural network that extracted features from longitudinally collected electrocardiograms (ECG), and used these to predict the risk of malignant ventricular arrhythmias. Methods A multicentre study in patients implanted with an implantable cardioverter-defibrillator (ICD) between 2007 and 2021 in two academic hospitals was performed. Variational autoencoders (VAEs), which combine neural networks with variational inference principles, and can learn patterns and structure in data without explicit labelling, were trained to encode the mean ECG waveforms from the limb leads into 16 variables. Supervised dynamic ML models using these latent ECG representations and clinical baseline information were trained to predict malignant ventricular arrhythmias treated by the ICD. Model performance was evaluated on a hold-out set, using time-dependent receiver operating characteristic (ROC) and calibration curves. Findings 2942 patients (61.7 ± 13.9 years, 25.5% female) were included, with a total of 32,129 ECG recordings during a mean follow-up of 43.9 ± 35.9 months. The mean time-varying area under the ROC curve for the dynamic model was 0.738 ± 0.07, compared to 0.639 ± 0.03 for a static (i.e. baseline-only model). Feature analyses indicated dynamic changes in latent ECG representations, particularly those affecting the T-wave morphology, were of highest importance for model predictions. Interpretation Dynamic ML models and neural networks effectively leverage routinely collected longitudinal ECG recordings for personalised and updated predictions of malignant ventricular arrhythmias, outperforming static models.

MZH Kolk, S Ruipérez-Campillo, L Alvarez-Florez, B Deb, EJ Bekkers, CP Allaart, ALCJ van der Lingen, P Clopton, I Isgum, AAM Wilde, RE Knops, SM Narayan, FVY Tjong

Lancet eBiomedicine

01.01.2024

LinkDOI

A precise method to measure spasticity is fundamental in improving the quality of life of spastic patients. The measurement methods that exist for spasticity have long been considered scarce and inadequate, which can partly be explained by a lack of consensus in the definition of spasticity. Spasticity quantification methods can be roughly classified according to whether they are based on neurophysiological or biomechanical mechanisms, clinical scales, or imaging techniques. This article reviews methods from all classes and further discusses instrumentation, dimensionality, and EMG onset detection methods. The objective of this article is to provide a review on spasticity measurement methods used to this day in an effort to contribute to the advancement of both the quantification and treatment of spasticity.

KO Kristinsdottir, S Ruipérez-Campillo, T Helgason

Chapter of the Book "Stroke-Management Pearls"

04.10.2023

LinkDOI

Background Segmentation of computed tomography (CT) is important for many clinical procedures including personalized cardiac ablation for the management of cardiac arrhythmias. While segmentation can be automated by machine learning (ML), it is limited by the need for large, labeled training data that may be difficult to obtain. We set out to combine ML of cardiac CT with domain knowledge, which reduces the need for large training datasets by encoding cardiac geometry, which we then tested in independent datasets and in a prospective study of atrial fibrillation (AF) ablation. Methods We mathematically represented atrial anatomy with simple geometric shapes and derived a model to parse cardiac structures in a small set of N = 6 digital hearts. The model, termed “virtual dissection,” was used to train ML to segment cardiac CT in N = 20 patients, then tested in independent datasets and in a prospective study. Results In independent test cohorts (N = 160) from 2 Institutions with different CT scanners, atrial structures were accurately segmented with Dice scores of 96.7% in internal (IQR: 95.3%–97.7%) and 93.5% in external (IQR: 91.9%–94.7%) test data, with good agreement with experts (r = 0.99; p < 0.0001). In a prospective study of 42 patients at ablation, this approach reduced segmentation time by 85% (2.3 ± 0.8 vs. 15.0 ± 6.9 min, p < 0.0001), yet provided similar Dice scores to experts (93.9% (IQR: 93.0%–94.6%) vs. 94.4% (IQR: 92.8%–95.7%), p = NS). Conclusions Encoding cardiac geometry using mathematical models greatly accelerated training of ML to segment CT, reducing the need for large training sets while retaining accuracy in independent test data. Combining ML with domain knowledge may have broad applications.

Ruibin Feng, Brototo Deb, Prasanth Ganesan, Fleur VY Tjong, Albert J Rogers, Samuel Ruipérez-Campillo, Sulaiman Somani, Paul Clopton, Tina Baykaner, Miguel Rodrigo, James Zou, Fracois Haddad, Matei Zahari, Sanjiv M. Narayan

Frontiers in cardiovascular medicine

02.10.2023

LinkDOI

This study presents a novel metric to evaluate the heterogeneity of cardiac substrate by using vector maps derived from omnipolar electrograms. This metric determines the level of disorganisation of electrical propagation having the potential to classify cardiac tissue under the catheter. We tested the methodology on propagation maps obtained from experimental recordings with and without electrical stimulation, under the assumption that the former exhibit greater heterogeneity. Results show the discriminatory behaviour of the parameter (p < 0.001), assigning higher values to non-stimulated maps and lower values in cases with stimulation. The clinical relevance of this paper lies in the introduction of a new metric defined on omnipolarderived vector maps, capable of identifying and quantifying areas of disorganised electrical propagation within the heart. This parameter has the potential to make orientation-independent catheterisation procedures more efficient providing electrophysiologists with valuable information for the management of arrhythmias.

L Pancorbo^*, S Ruipérez-Campillo^*, F Castells, J Millet
^* denotes shared first authorship

IEEE Computing in Cardiology (50th CinC, 2023)

01.10.2023

LinkDOI

Many patients remain in a comatose state after initially surviving a resuscitation following a cardiac arrest. The prognosis in this state carries the decision of life support withdrawal, thus needing an objective and deterministic guideline. The objective of this study, is to assist this decision by providing a model able to predict the cerebral performance category (CPC) of comatose patients following cardiac arrest from their electroencephalographic (EEG) signal. To achieve this, binary classifiers built with 3D Convolutional Neural Networks (CNNs) followed by Dense Neural Networks (DNN) are used in combination with a “divide and conquer” strategy, thus enabling the automatic extraction of features from the tensors of EEG signals, taking into consideration the spatial relation of the signals according to the electrodes’ distribution on the scalp. This work was submitted under the team name “BioITACA UPV” to “Predicting Neurological Recovery from Coma After Cardiac Arrest: The George B. Moody PhysioNet Challenge 2023”, and while the team did not score in the official phase, results obtained from a held-out subset of the training set demonstrate the capability of the model to classify by CPC from short segments of 5 seconds to long recordings of EEG data. Results show an average accuracy of 0.76 between the CPC classifiers and capability to discern between a good or bad outcome prognosis.

RT Ors-Quixal, E Ramírez-Candela, S Ruipérez-Campillo, F Castells, J Millet

IEEE Computing in Cardiology (50th CinC, 2023)

01.10.2023

LinkDOI

The aim of this study is to improve the prediction of long-term outcomes in patients with atrial fibrillation solely using electrogram (EGM) features. We developed three distinct models based on data from a cohort of N=561 patients, each targeting different aspects of EGM analysis: Principal Component Analysis (PCA): We applied PCA to analyze the variances of eigenvectors projecting more than a fixed threshold of the overall variance (15%). To identify common projection axes among these eigenvectors, we employed the k-means algorithm for clustering. Auto Regressive: This technique involves applying a bijective transformation to the coefficients, which are subsequently used as input for various machine learning classifiers, including Random Forest or Support Vector Classifier. Feature Engineering: We performed feature engineering by extracting voltage, rate, and shape similarity metrics from raw EGM (Electrogram) data.

M Pedron, P Ganesan, R Feng, B Deb, H Chang, S Ruipérez-Campillo, S Somani, Y Desai, AJ Rogers, P Clopton, SM Narayan

IEEE Computing in Cardiology (50th CinC, 2023)

01.10.2023

LinkDOI

The vectorcardiogram (VCG) provides a comprehensive representation of the heart's electrical activity in 3D aiding in the diagnosis and treatment of cardiovascular diseases. The conventional electrocardiogram (ECG) records twelve leads intermittently at intervals of 2.5 seconds, with lead II typically recorded continuously, which poses a challenge for reconstructing the VCG, as each lead's beats belong to different time instances. The purpose of this research is to propose and validate a methodology for accurately synchronizing the recording beats to reconstruct the VCG. To achieve this goal, a phantom was created to mimic the standard 12-lead ECG setup. The temporal offset of each beat from the first is calculated using cross-correlation utilizing the continuous lead and the same offset is applied to all leads, and finally reconstructing the VCG. The results demonstrate precise synchronization, as evidenced by Pearson correlation values of 0.9959±0.0034 , an MAE of 0.0077±0.0024 mV , and an RMSE of 0.0119±0.0038 mV in the VCG reconstruction. This technique is essential for the accurate diagnosis and treatment of cardiovascular diseases and can be applied to conventional ECG recordings taken on paper to obtain VCG.

E Ramítez, S Ruipérez-Campillo, F Castells, R Casado-Arroyo, J Millet

IEEE Computing in Cardiology (50th CinC, 2023)

01.10.2023

LinkDOI

Aims Left ventricular ejection fraction (LVEF) is suboptimal as a sole marker for predicting sudden cardiac death (SCD). Machine learning (ML) provides new opportunities for personalized predictions using complex, multimodal data. This study aimed to determine if risk stratification for implantable cardioverter-defibrillator (ICD) implantation can be improved by ML models that combine clinical variables with 12-lead electrocardiograms (ECG) time-series features. Methods and results A multicentre study of 1010 patients (64.9 ± 10.8 years, 26.8% female) with ischaemic, dilated, or non-ischaemic cardiomyopathy, and LVEF ≤ 35% implanted with an ICD between 2007 and 2021 for primary prevention of SCD in two academic hospitals was performed. For each patient, a raw 12-lead, 10-s ECG was obtained within 90 days before ICD implantation, and clinical details were collected. Supervised ML models were trained and validated on a development cohort (n = 550) from Hospital A to predict ICD non-arrhythmic mortality at three-year follow-up (i.e. mortality without prior appropriate ICD-therapy). Model performance was evaluated on an external patient cohort from Hospital B (n = 460). At three-year follow-up, 16.0% of patients had died, with 72.8% meeting criteria for non-arrhythmic mortality. Extreme gradient boosting models identified patients with non-arrhythmic mortality with an area under the receiver operating characteristic curve (AUROC) of 0.90 [95% confidence intervals (CI) 0.80-1.00] during internal validation. In the external cohort, the AUROC was 0.79 (95% CI 0.75-0.84). Conclusions ML models combining ECG time-series features and clinical variables were able to predict non-arrhythmic mortality within three years after device implantation in a primary prevention population, with robust performance in an independent cohort.

MZH Kolk, S Ruipérez-Campillo, B Deb, E Bekkers, CP Allaart, AJ Rogers, ACJ Van Der Lingen, I Isgum, B De Vos, P Clopton, others

Europace

15.09.2023

LinkDOI

The purpose of the study is to better understand the complex nature of loneliness in older adults and the potential contributing factors that may impact their sense of connection and well-being. The study utilized a mixed-methods approach, combining quantitative measures such as heart rate monitoring with qualitative data collected through interviews and surveys. The findings suggest that loneliness in older adults may be influenced by multiple factors, including their level of education, resilience, and empathy and incidence in spontaneous heart rate variations. Results highlight the importance of empathy in promoting social connectedness and reducing feelings of loneliness in older adults, may have implications for developing targeted interventions aimed at reducing loneliness and improving the well-being of older adults.

R Cervigón, S Ruipérez-Campillo, J Millet, F Castells

IEEE Mediterranean Conference on Medical and Biological Engineering and Computing (2023)

14.09.2023

LinkDOI

In this study, a novel unsupervised classification framework for time series of medical nature is presented. This framework is based on the intersection of machine learning, Hilbert Spaces algebra, and signal theory. The methodology is illustrated through the resolution of three biomedical engineering problems: neuronal activity tracking, protein functional classification, and non-invasive diagnosis of atrial flutter (AFL). The results indicate that the proposed algorithms exhibit high proficiency in solving these tasks and demonstrate robustness in identifying damaged neuronal units while tracking healthy ones. Moreover, the application of the framework in protein functional classification provides a new perspective for the development of pharmaceutical products and personalised medicine. Additionally, the controlled environment of the framework in AFL simulation problem underscores the algorithm’s ability to encode information efficiently. These results offer valuable insights into the potential of this framework and lay the groundwork for future studies.Clinical relevance— The framework proposed in this study has the potential to yield novel insights into the effects of newly implanted electrodes in the brain. Furthermore, the categorization of proteins by function could facilitate the development of personalised and efficient medicines, ultimately reducing both time and cost. The simulation of atrial flutter also demonstrates the framework’s ability to encode information for arrhythmia diagnosis and treatment, which has the potential to lead to improved patient outcomes.

S Ruipérez-Campillo, F Castells, J Millet

IEEE Engineering in Medicine & Biology Society (45th EMBC, 2023)

24.07.2023

LinkDOI

The present study aims to design and fabricate a system capable of generating heterogeneities on the epicardial surface of an isolated rabbit heart perfused in a Langendorff system. The system consists of thermoelectric modules that can be independently controlled by the developed hardware, thereby allowing for the generation of temperature gradients on the epicardial surface, resulting in conduction slowing akin to heterogeneities of pathological origin. A comprehensive analysis of the system’s viability was performed through modeling and thermal simulation, and its practicality was validated through preliminary tests conducted at the experimental cardiac electrophysiology laboratory of the University of Valencia. The design process involved the use of Fusion 360 for 3D designs, MATLAB/Simulink for algorithms and block diagrams, LTSpice and Altium Designer for schematic captures and PCB design, and the integration of specialized equipment for animal experimentation. The objective of the study was to efficiently capture epicardial recordings under varying conditions. Clinical relevance— The proposed system aims to induce local epicardial heterogeneities to generate labeled correct signals that can serve as a golden standard for improving algorithms that identify and characterize fibrotic substrates. This improvement will enhance the efficacy of ablation processes and potentially reduce the ablated surface area.

I Segarra, A Cebrián, S Ruipérez-Campillo, A Tormos, FJ Chorro, F Castells, A Alberola, J Millet

IEEE Engineering in Medicine & Biology Society (45th EMBC, 2023)

24.07.2023

LinkDOI

High-density catheters combined with Orientation Independent Sensing (OIS) methods have emerged as a groundbreaking technology for cardiac substrate characterisation. In this study, we aim to assess the arrangements and constraints to reliably estimate the so-called omnipolar electrogram (oEGM). Performance was evaluated using an experimental animal model. Thirty-eight recordings from nine retrospective experiments on isolated perfused rabbit hearts with an epicardial HD multielectrode were used. We estimated oEGMs according to the classic triangular clique (4 possible orientations) and a novel cross-orientation clique arrangement. Furthermore, we tested the effects of interelectrode spacing from 1 to 4 mm. Performance was evaluated by means of several parameters that measured amplitude rejection ratios, electric field loop area, activation pulse width and morphology distortion. Most reliable oEGM estimations were obtained with cross-configurations and interelectrode spacings <=2 mm. Estimations from triangular cliques resulted in wider electric field loops and unreliable detection of the direction of the propagation wavefront. Moreover, increasing interelectrode distance resulted in increased pulse width and morphology distortion. The results prove that current oEGM estimation techniques are insufficiently accurate. This study opens a new standpoint for the design of new-generation HD catheters and mapping software.

S Ruipérez-Campillo, M Crespo, A Tormos, A Guill, A Cebrián, A Alberola, J Heimer, FJ Chorro, J Millet, F Castells

Physical and Engineering Sciences in Medicine

26.06.2023

LinkDOI

Aims There is a clinical spectrum for atrial tachyarrhythmias wherein most patients with atrial tachycardia (AT) and some with atrial fibrillation (AF) respond to ablation, while others do not. It is undefined if this clinical spectrum has pathophysiological signatures. This study aims to test the hypothesis that the size of spatial regions showing repetitive synchronized electrogram (EGM) shapes over time reveals a spectrum from AT, to AF patients who respond acutely to ablation, to AF patients without acute response. Methods and results We studied n = 160 patients (35% women, 65.0 ± 10.4 years) of whom (i) n = 75 had AF terminated by ablation propensity matched to (ii) n = 75 without AF termination and (iii) n = 10 with AT. All patients had mapping by 64-pole baskets to identify areas of repetitive activity (REACT) to correlate unipolar EGMs in shape over time. Synchronized regions (REACT) were largest in AT, smaller in AF termination, and smallest in non-termination cohorts (0.63 ± 0.15, 0.37 ± 0.22, and 0.22 ± 0.18, P < 0.001). Area under the curve for predicting AF termination in hold-out cohorts was 0.72 ± 0.03. Simulations showed that lower REACT represented greater variability in clinical EGM timing and shape. Unsupervised machine learning of REACT and extensive (50) clinical variables yielded four clusters of increasing risk for AF termination (P < 0.01, χ2), which were more predictive than clinical profiles alone (P < 0.001). Conclusion The area of synchronized EGMs within the atrium reveals a spectrum of clinical response in atrial tachyarrhythmias. These fundamental EGM properties, which do not reflect any predetermined mechanism or mapping technology, predict outcome and offer a platform to compare mapping tools and mechanisms between AF patient groups.

P Ganesan, B Deb, F Feng, M Rodrigo, S Ruipérez-Campillo, AJ Rogers, P Clopton, JJ Wang, S Zeemering, U Schotten, WJ Rappel, SM Narayan

Europace

18.03.2023

LinkDOI

Objective: The aim of this study is to propose a method to reduce the sensitivity of the estimated omnipolar electrogram (oEGM) with respect to the angle of the propagation wavefront. Methods: A novel configuration of cliques taking into account all four electrodes of a squared cell is proposed. To test this approach, simulations of HD grids of cardiac activations at different propagation angles, conduction velocities, interelectrode distance and electrogram waveforms are considered. Results: The proposed approach successfully provided narrower loops (essentially a straight line) of the electrical field described by the bipole pair with respect to the conventional approach. Estimation of the direction of propagation was improved. Additionally, estimated oEGMs presented larger amplitude, and estimations of the local activation times were more accurate. Conclusions: A novel method to improve the estimation of oEGMs in HD grid of electrodes is proposed. This approach is superior to the existing methods and avoids pitfalls not yet resolved. Relevance: Robust tools for quantifying the cardiac substrate are crucial to determine with accuracy target ablation sites during an electrophysiological procedure.

F Castells^*, S Ruipérez-Campillo^*, I Segarra, R Cervigón, R Casado-Arroyo, JL Merino, J Millet
^* denotes shared first authorship

Computers in Biology and Medicine

01.03.2023

LinkDOI

Background Ventricular arrhythmia (VA) precipitating sudden cardiac arrest (SCD) is among the most frequent causes of death and pose a high burden on public health systems worldwide. The increasing availability of electrophysiological signals collected through conventional methods (e.g. electrocardiography (ECG)) and digital health technologies (e.g. wearable devices) in combination with novel predictive analytics using machine learning (ML) and deep learning (DL) hold potential for personalised predictions of arrhythmic events. Methods This systematic review and exploratory meta-analysis assesses the state-of-the-art of ML/DL models of electrophysiological signals for personalised prediction of malignant VA or SCD, and studies potential causes of bias (PROSPERO, reference: CRD42021283464). Five electronic databases were searched to identify eligible studies. Pooled estimates of the diagnostic odds ratio (DOR) and summary area under the curve (AUROC) were calculated. Meta-analyses were performed separately for studies using publicly available, ad-hoc datasets, versus targeted clinical data acquisition. Studies were scored on risk of bias by the PROBAST tool. Findings 2194 studies were identified of which 46 were included in the systematic review and 32 in the meta-analysis. Pooling of individual models demonstrated a summary AUROC of 0.856 (95% CI 0.755–0.909) for short-term (time-to-event up to 72 h) prediction and AUROC of 0.876 (95% CI 0.642–0.980) for long-term prediction (time-to-event up to years). While models developed on ad-hoc sets had higher pooled performance (AUROC 0.919, 95% CI 0.867–0.952), they had a high risk of bias related to the re-use and overlap of small ad-hoc datasets, choices of ML tool and a lack of external model validation. Interpretation ML and DL models appear to accurately predict malignant VA and SCD. However, wide heterogeneity between studies, in part due to small ad-hoc datasets and choice of ML model, may reduce the ability to generalise and should be addressed in future studies.

MZH Kolk, B Deb, S Ruipérez-Campillo, NK Bhatia, P Clopton, AAM Wilde, SM Narayan, RE Knops, FVY Tjong

The Lancet eBiomedicine

01.03.2023

LinkDOI

CD8+ T cells underpin effective anti-tumor immune responses in melanoma; however, their functions are attenuated due to various immunosuppressive factors in the tumor microenvironment (TME), resulting in disease progression. T cell function is elicited by the T cell receptor (TCR), which recognizes antigen peptide-major histocompatibility complex (pMHC) expressed on tumor cells via direct physical contact, i.e., two-dimensional (2D) interaction. TCR–pMHC 2D affinity plays a central role in antigen recognition and discrimination, and is sensitive to both the conditions of the T cell and the microenvironment in which it resides. Herein, we demonstrate that CD8+ T cells residing in TME have lower 2D TCR–pMHC bimolecular affinity and TCR–pMHC–CD8 trimolecular avidity, pull fewer TCR–pMHC bonds by endogenous forces, flux lower level of intracellular calcium in response to antigen stimulation, exhibit impaired in vivo activation, and show diminished anti-tumor effector function. These detrimental effects are localized in the tumor and tumor draining lymph node (TdLN), and affect both antigen-inexperienced and antigen-experienced CD8+ T cells irrespective of their TCR specificities. These findings implicate impaired antigen recognition as a mechanism of T cell dysfunction in the TME.

Z Yuan, MJ O’Melia, K Li, J Lyu, F Zhou, P Jothikumar, NA Rohner, MP Manspeaker, DM Francis, K Bai, C Ge, MN Rushdi, L Chingozha, S Ruipérez-Campillo, H Lu, SN Thomas, C Zhu

bioRxiv

13.09.2022

LinkDOI

The diagnosis and treatment of cardiac arrhythmias relies on catheter recordings, that may be inefficient because of the continued use of the bipolar processing and analysis techniques of traditional catheters, missing the potential of the novel matrix catheters. This results in the need of more processing of the signals and longer cardiac scans to obtain accurate information about the state of the tissue being analysed. This study proposes a new clique configuration to compute omnipolar EGM (oEGM) in multi-electrode array catheters to obtain parameters of interest in a more robust and accurate manner. Numerous simulations with varying input parameters are designed to emulate the propagation of electrical activity on the cardiac tissue surface captured by the catheter and characterise the differences between the classic method of omnipolar analysis (triangular clique) and our proposed new method (cross clique). The results show that the cross clique is more robust to variations in the direction of wave propagation, and more accurate in the estimation of the local activation time (LAT).

ISegarra, S Ruipérez-Campillo, FCastells, J Millet

IEEE Computing in Cardiology (49th CinC, 2022)

04.09.2022

LinkDOI

Automated segmentation of myocardial fibrosis in late gadolinium enhancement (LGE) cardiac MRI (CMR) has the potential to improve efficiency and precision of diagnosis and treatment of cardiomyopathies. However, state-of-the-art Deep Learning approaches require manual pixel-level annotations. Using weaker labels can greatly reduce manual annotation time and expedite dataset curation, which is why we propose fibrosis segmentation methods using either slice-level or stack-level fibrosis labels. 5759 short-axis LGE CMR image slices were retrospectively obtained from 482 patients. U-Nets with slice-level and stack-level supervision are trained with 446 weakly-labeled patients by making use of a myocardium segmentation U-Net and fibrosis classification Dilated Residual Networks (DRN). For comparison, a U-Net is trained with pixel-level supervision using a training set of 81 patients. On the proprietary test set of 24 patients, pixel-level, slice-level and stack-level supervision reach Dice scores of 0.74, 0.70 and 0.70, while on the external Emidec dataset of 100 patients Dice scores of 0.55, 0.61 and 0.52 were obtained. Results indicate that using larger weakly-annotated datasets can approach the performance of methods using pixel-level annotated datasets and potentially improve generalization to external datasets.

RC Klein, RE van Lieshout, MZH Kolk, K Geijtenbeek, R Vos, S Ruipérez-Campillo, R Feng, B Deb, P Ganesan, RE Knops, I Isgum, SM Narayan, E Bekkers, B Vos FVY Tong

IEEE Computing in Cardiology (49th CinC, 2022)

04.09.2022

LinkDOI

Many patients at high risk of life-threatening ventricular arrhythmias (VA) and sudden cardiac death (SCD) who received an implantable cardioverter defibrillator (ICD), never receive appropriate device therapy. The presence of fibrosis on LGE CMR imaging is shown to be associated with increased risk of VA. Therefore, there is a strong need for both automatic segmentation and quantification of cardiac fibrosis as well as better risk stratification for SCD. This study first presents a novel two-stage deep learning network for the segmentation of left ventricle myocardium and fibrosis on LGE CMR images. Secondly it aims to effectively predict device therapy in ICD patients by using a graph neural network approach which incorporates both myocardium and fibrosis features as well as the left ventricle geometry. Our segmentation network outperforms previous state-of-the-art methods on 2D CMR data, reaching a Dice score of 0.82 and 0.77 on myocardium and fibrosis segmentation, respectively. The ICD therapy prediction network reaches an AUC of 0.60 while using only CMR data and outperforms baseline methods based on current guideline markers for ICD implantation. This work lays a strong basis for future research on improved risk stratification for VA and SCD.

FE van Lieshout, RC Klein, MZH Kolk, K van Geijtenbeek, R Vos, S Ruipérez-Campillo, R Feng, B Deb, P Ganesan, RE Knops, I Isgum, SM Narayan, E Bekkers, B Vos, FVY Tjong

IEEE Computing in Cardiology (49th CinC, 2022)

04.09.2022

LinkDOI

Accurate non-invasive diagnoses in the context of cardiac diseases are problems that hitherto remain unresolved. We propose an unsupervised classification of atrial flutter (AFL) using dimensional transforms of ECG signals in high dimensional vector spaces. A mathematical model is used to generate synthetic signals based on clinical AFL signals, and hierarchical clustering analysis and novel machine learning (ML) methods are designed for the un-supervised classification. Metrics and accuracy parameters are created to assess the performance of the model, proving the power of this novel approach for the diagnosis of AFL from ECG using innovative AI algorithms.

S Ruipérez-Campillo, J Millet, FCastells

IEEE Computing in Cardiology (49th CinC, 2022)

04.09.2022

LinkDOI

Persistent atrial fibrillation ablation has a high recurrence rate. In this work, we performed an analysis of bipolar intracavitary signals obtained with a conventional 24-pole diagnostic catheter (Woven Orbiter) placed in the right atrium and coronary sinus in a cohort of patients with persistent atrial fibrillation undergoing ablation to detect features predictive of acute procedural success (conversion to sinus rhythm during ablation) and the occurrence of recurrences. The goal is to arrive at a quantitative description of the degree of randomness of the atrial response in atrial fibrillation and to demonstrate the presence of hidden periodic components. This was done by the determination of the autocorrelation function. Results showed that higher correlation in relative maximum peaks, and a lower dominant atrial frequency (greater distance between relative amplitude maxima) may be associated with a greater likelihood of achieving reversion to sinus rhythm and lower probability of recurrences. A larger study is needed to draw conclusions.

R Cervigón, E Franco, S Ruipérez-Campillo, C Lozano, F Castells, J Moreno

IEEE Computing in Cardiology (49th CinC, 2022)

04.09.2022

LinkDOI

Loneliness in older adults is associated with functional decline, depression and even death. Given the prevalence of loneliness, the aim of this study was to examine the association between loneliness and cardiac biomarkers in older people that attend to cardiology consultation. The results showed that loneliness was more prevalent in women than in men, and it was associated with marital status too. ECG recording were analyzed and QT interval and T-wave length showed higher values in people suffering from loneliness, as well as higher cardiac frequency, where the presence of meaning in life be a protective factor. Studies with a larger sample size are needed, but these results appear to show a relationship between biomarkers and mental state.

ML Cardo, A Chulián, S Ruipérez-Campillo, J Millet, F Castells, R Cervigón

IEEE Computing in Cardiology (49th CinC, 2022)

04.09.2022

LinkDOI

Background and objectives Macroreentrant atrial tachyarrhythmias (MRATs) can be caused by different reentrant circuits. The treatment for each MRAT type may require ablation at different sites, either at the right or left atria. Unfortunately, the reentrant circuit that drives the arrhythmia cannot be ascertained previous to the electrophysiological intervention. Methods A noninvasive approach based on the comparison of atrial vectorcardiogram (VCG) loops is proposed. An archetype for each group was created, which served as a reference to measure the similarity between loops. Methods were tested in a variety of simulations and real data obtained from the most common right (peritricuspid) and left (perimitral) macroreentrant circuits, each divided into clockwise and counterclockwise subgroups. Adenosine was administered to patients to induce transient AV block, allowing the recording of the atrial signal without the interference of ventricular signals. From the vectorcardiogram, we measured intrapatient loop consistence, similarity of the pathway to archetypes, characterisation of slow velocity regions and pathway complexity. Results Results show a considerably higher similarity with the loop of its corresponding archetype, in both simulations and real data. We found the capacity of the vectorcardiogram to reflect a slow velocity region, consistent with the mechanisms of MRAT, and the role that it plays in the characterisation of the reentrant circuit. The intra-patient loop consistence was over 0.85 for all clinical cases while the similarity of the pathway to archetypes was found to be 0.85 ± 0.03, 0.95 ± 0.03, 0.87 ± 0.04 and 0.91 ± 0.02 for the different MRAT types (and p<0.02 for 3 of the 4 groups), and pathway complexity also allowed to discriminate among cases (with p<0.05). Conclusions We conclude that the presented methodology allows us to differentiate between the most common forms of right and left MRATs and predict the existence and location of a slow conduction zone. This approach may be useful in planning ablation procedures in advance.

S Ruipérez-Campillo, S Castrejón, M Martínez, R Cervigón, OMeste, JL Merino, J Millet, FCastells

Computer methods and programs in biomedicine

01.02.2021

LinkDOI

The objective of this study is to non-invasively characterise a variety of atrial flutter (AFL) types, defined by a maroreentrant circuit. A vectorcardiographic approach is proposed to compare atrial macroreentrant circuits. Vectorcardiogram (VCG) arechetypes are computed so that parameters such as similarity among loops can be calculated. The methodology was employed in a set of artificial VCGs created from a computational simulation based on a mathematical model and in signals from real patients. Adenosine was used to block the ventricular contribution to the ECG signal, later transformed to a VCG analysed from different perspectives. Results demonstrate a high similarity for cases belonging to a group with its archetype in synthetic and real cases. Slow conduction velocity regions were found to be very well represented in VCGs, in accordance with AFL mechanisms and its importance when characterising atrial macroreentries. The conclusion is that our methodology allows differentiation between the most recurrent types of AFL through the analysis of its VCG representation, predicting the presence of slow conduction regions along the macroreentry. This can be very useful when planning in advance the ablation procedure.

S Ruipérez-Campillo, S Castrejón, M Martínez, R Cervigón, O Meste, JL Merino, J Millet, F Castells

IEEE Computing in Cardiology (47th CinC, 2020)

13.09.2020

LinkDOI