{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/learning-disentangled-representation-from-12","title":"Learning disentangled representation from 12-lead electrograms: application in localizing the origin of Ventricular Tachycardia","arxiv_id":"1808.01524","date":"2018-08-04","proceeding":null,"authors":["Prashnna K Gyawali","B. Milan Horacek","John L. Sapp","Linwei Wang"],"abstract":"The increasing availability of electrocardiogram (ECG) data has motivated the\nuse of data-driven models for automating various clinical tasks based on ECG\ndata. The development of subject-specific models are limited by the cost and\ndifficulty of obtaining sufficient training data for each individual. The\nalternative of population model, however, faces challenges caused by the\nsignificant inter-subject variations within the ECG data. We address this\nchallenge by investigating for the first time the problem of learning\nrepresentations for clinically-informative variables while disentangling other\nfactors of variations within the ECG data. In this work, we present a\nconditional variational autoencoder (VAE) to extract the subject-specific\nadjustment to the ECG data, conditioned on task-specific representations\nlearned from a deterministic encoder. To encourage the representation for\ninter-subject variations to be independent from the task-specific\nrepresentation, maximum mean discrepancy is used to match all the moments\nbetween the distributions learned by the VAE conditioning on the code from the\ndeterministic encoder. The learning of the task-specific representation is\nregularized by a weak supervision in the form of contrastive regularization. We\napply the proposed method to a novel yet important clinical task of classifying\nthe origin of ventricular tachycardia (VT) into pre-defined segments,\ndemonstrating the efficacy of the proposed method against the standard VAE.","url_abs":"http://arxiv.org/abs/1808.01524v1","url_pdf":"http://arxiv.org/pdf/1808.01524v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"learning-disentangled-representation-from-12","repo_url":"https://github.com/Prasanna1991/MMD_VAE","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}