This paper introduces a method based on Denoising AutoEncoder (DAE) for missing data imputation.This method, called mDAE hereafter, results from a modification of the loss function and a straightforward procedure for choosing the hyper-parameters.An ablation study shows on several UCI Machine Learning Repository datasets, the benefit of using this modified loss function and an overcomplete structure, in terms of Root Mean Squared Error (RMSE) of reconstruction.This numerical study is complemented by a comparison between mDAE and eight alternative approaches (four classical and four more recent), using the Mean Distance to the Best (MDB) criterion, which quantifies the overall performance of each method across all the datasets.

Electrocardiogram (ECG) signals provide essential information about the heart's condition and are widely used for diagnosing cardiovascular diseases. The morphology of a single heartbeat over the available leads is a primary biosignal for monitoring cardiac conditions. However, analyzing heartbeat morphology can be challenging due to noise and artifacts, missing leads, and a lack of annotated data. Generative models, such as denoising diffusion generative models (DDMs), have proven successful in generating complex data. We introduce $\texttt{BeatDiff}$, a light-weight DDM tailored for the morphology of multiple leads heartbeats. We then show that many important ECG downstream tasks can be formulated as conditional generation methods in a Bayesian inverse problem framework using $\texttt{BeatDiff}$ as priors. We propose $\texttt{EM-BeatDiff}$, an Expectation-Maximization algorithm, to solve this conditional generation tasks without fine-tuning. We illustrate our results with several tasks, such as removal of ECG noise and artifacts (baseline wander, electrode motion), reconstruction of a 12-lead ECG from a single lead (useful for ECG reconstruction of smartwatch experiments), and unsupervised explainable anomaly detection. Numerical experiments show that the combination of $\texttt{BeatDiff}$ and $\texttt{EM-BeatDiff}$ outperforms SOTA methods for the problems considered in this work.

Reliable patient-specific ventricular repolarization times (RTs) can identify regions of functional block or afterdepolarizations, indicating arrhythmogenic cardiac tissue and the risk of sudden cardiac death. Unipolar electrograms (UEs) record electric potentials, and the Wyatt method has been shown to be accurate for estimating RT from a UE. High-pass filtering is an important step in processing UEs, however, it is known to distort the T-wave phase of the UE, which may compromise the accuracy of the Wyatt method. The aim of this study was to examine the effects of high-pass filtering, and improve RT estimates derived from filtered UEs. We first generated a comprehensive set of UEs, corresponding to early and late activation and repolarization, that were then high-pass filtered with settings that mimicked the CARTO filter. We trained a deep neural network (DNN) to output a probabilistic estimation of RT and a measure of confidence, using the filtered synthetic UEs and their true RTs. Unfiltered ex-vivo human UEs were also filtered and the trained DNN used to estimate RT. Even a modest 2 Hz high-pass filter imposes a significant error on RT estimation using the Wyatt method. The DNN outperformed the Wyatt method in 62.75% of cases, and produced a significantly lower absolute error (p=8.99E-13), with a median of 16.91 ms, on 102 ex-vivo UEs. We also applied the DNN to patient UEs from CARTO, from which an RT map was computed. In conclusion, DNNs trained on synthetic UEs improve the RT estimation from filtered UEs, which leads to more reliable repolarization maps that help to identify patient-specific repolarization abnormalities.

Integrating spatial information about atrial physiology and anatomy in a single patient from multimodal datasets, as well as generalizing these data across patients, requires a common coordinate system. In the atria, this is challenging due to the complexity and variability of the anatomy. We aimed to develop and validate a Universal Atrial Coordinate (UAC) system for the following applications: combination and assessment of multimodal data; comparison of spatial data across patients; 2D visualization; and construction of patient specific geometries to test mechanistic hypotheses. Left and right atrial LGE-MRI data were segmented and meshed. Two coordinates were calculated for each atrium by solving Laplace's equation, with boundary conditions assigned using five landmark points. The coordinate system was used to map spatial information between atrial meshes, including scalar fields measured using different mapping modalities, and atrial anatomic structures and fibre directions from a reference geometry. Average error in point transfer from a source mesh to a destination mesh and back again was less than 0.1 mm for the left atrium and 0.02 mm for the right atrium. Patient specific meshes were constructed using the coordinate system and phase singularity density maps from arrhythmia simulations were visualised in 2D. In conclusion, we have developed a universal atrial coordinate system allowing automatic registration of imaging and electroanatomic mapping data, 2D visualisation, and patient specific model creation.

Success rates for catheter ablation of persistent atrial fibrillation patients are currently low; however, there is a subset of patients for whom electrical isolation of the pulmonary veins alone is a successful treatment strategy. It is difficult to identify these patients because there are a multitude of factors affecting arrhythmia susceptibility and maintenance, and the individual contributions of these factors are difficult to determine clinically. We hypothesised that the combination of pulmonary vein (PV) electrophysiology and atrial body fibrosis determine driver location and effectiveness of pulmonary vein isolation (PVI). We used bilayer biatrial computer models based on patient geometries to investigate the effects of PV properties and atrial fibrosis on arrhythmia inducibility, maintenance mechanisms, and the outcome of PVI. Short PV action potential duration (APD) increased arrhythmia susceptibility, while longer PV APD was found to be protective. Arrhythmia inducibility increased with slower conduction velocity (CV) at the LA/PV junction, but not for cases with homogeneous CV changes or slower CV at the distal PV. Phase singularity (PS) density in the PV region for cases with PV fibrosis was increased. Arrhythmia dynamics depend on both PV properties and fibrosis distribution, varying from meandering rotors to PV reentry (in cases with baseline or long APD), to stable rotors at regions of high fibrosis density. Measurement of fibrosis and PV properties may indicate patient specific susceptibility to AF initiation and maintenance. PV PS density before PVI was higher for cases in which AF terminated or converted to a macroreentry; thus, high PV PS density may indicate likelihood of PVI success.