a b s t r a c tThe interest in statistical classification for critical applications such as diagnoses of patient samples based on supervised learning is rapidly growing. To gain acceptance in applications where the subsequent decisions have serious consequences, e.g. choice of cancer therapy, any such decision support system must come with a reliable performance estimate. Tailored for small sample problems, cross-validation (CV) and bootstrapping (BTS) have been the most commonly used methods to determine such estimates in virtually all branches of science for the last 20 years. Here, we address the often overlooked fact that the uncertainty in a point estimate obtained with CV and BTS is unknown and quite large for small sample classification problems encountered in biomedical applications and elsewhere. To avoid this fundamental problem of employing CV and BTS, until improved alternatives have been established, we suggest that the final classification performance always should be reported in the form of a Bayesian confidence interval obtained from a simple holdout test or using some other method that yields conservative measures of the uncertainty.
Exit sites associated with scar-related reentrant arrhythmias represent important targets for catheter ablation therapy. However, their accurate location in a safe and robust manner remains a significant clinical challenge. We recently proposed a novel quantitative metric (termed the Reentry Vulnerability Index, RVI) to determine the difference between activation and repolarisation intervals measured from pairs of spatial locations during premature stimulation to accurately locate the critical site of reentry formation. In the clinic, the method showed potential to identify regions of low RVI corresponding to areas vulnerable to reentry, subsequently identified as ventricular tachycardia (VT) circuit exit sites. Here, we perform an in silico investigation of the RVI metric in order to aid the acquisition and interpretation of RVI maps and optimise its future usage within the clinic. Within idealised 2D sheet models we show that the RVI produces lower values under correspondingly more arrhythmogenic conditions, with even low resolution (8 mm electrode separation) recordings still able to locate vulnerable regions. When applied to models of infarct scars, the surface RVI maps successfully identified exit sites of the reentrant circuit, even in scenarios where the scar was wholly intramural. Within highly complex infarct scar anatomies with multiple reentrant pathways, the identified exit sites were dependent upon the specific pacing location used to compute the endocardial RVI maps. However, simulated ablation of these sites successfully prevented the reentry re-initiation. We conclude that endocardial surface RVI maps are able to successfully locate regions vulnerable to reentry corresponding to critical exit sites during sustained scar-related VT. The method is robust against highly complex and intramural scar anatomies and low resolution clinical data acquisition. Optimal location of all relevant sites requires RVI maps to be computed from multiple pacing locations.
Both biomedical research and clinical practice rely on complex datasets for the physiological and genetic characterization of human hearts in health and disease. Given the complexity and variety of approaches and recordings, there is now growing recognition of the need to embed computational methods in cardiovascular medicine and science for analysis, integration and prediction. This paper describes a Workshop on Computational Cardiovascular Science that created an international, interdisciplinary and inter-sectorial forum to define the next steps for a human-based approach to disease supported by computational methodologies. The main ideas highlighted were (i) a shift towards human-based methodologies, spurred by advances in new in silico, in vivo, in vitro, and ex vivo techniques and the increasing acknowledgement of the limitations of animal models. (ii) Computational approaches complement, expand, bridge, and integrate in vitro, in vivo, and ex vivo experimental and clinical data and methods, and as such they are an integral part of human-based methodologies in pharmacology and medicine. (iii) The effective implementation of multi- and interdisciplinary approaches, teams, and training combining and integrating computational methods with experimental and clinical approaches across academia, industry, and healthcare settings is a priority. (iv) The human-based cross-disciplinary approach requires experts in specific methodologies and domains, who also have the capacity to communicate and collaborate across disciplines and cross-sector environments. (v) This new translational domain for human-based cardiology and pharmacology requires new partnerships supported financially and institutionally across sectors. Institutional, organizational, and social barriers must be identified, understood and overcome in each specific setting.
The bidomain and monodomain equations are well established as the standard set of equations for the simulation of cardiac electrophysiological behavior. However, the computational cost of detailed bidomain/monodomain simulations limits their applicability in scenarios where a large number of simulations needs to be performed (e.g., parameter estimation). In this study, we present a graph-based method, which relies on point-to-point path finding to estimate activation times for single points in cardiac tissue with minimal computational costs. To validate our approach, activation times are compared to monodomain simulation results for an anatomically based rabbit ventricular model, incorporating realistic fiber orientation and conduction heterogeneities. Differences in activation times between the graph-based method and monodomain results are less than 10% of the total activation time, and computational performance is orders of magnitude faster with the proposed method when calculating activation times at single points. These results suggest that the graph-based method is well suited for estimating activation times when the need for fast performance justifies a limited loss of accuracy.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.