{"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/measuring-patient-similarities-via-a-deep","title":"Measuring Patient Similarities via a Deep Architecture with Medical Concept Embedding","arxiv_id":"1902.03376","date":"2019-02-09","proceeding":null,"authors":["Zihao Zhu","Changchang Yin","Buyue Qian","Yu Cheng","Jishang Wei","Fei Wang"],"abstract":"Evaluating the clinical similarities between pairwise patients is a\nfundamental problem in healthcare informatics. A proper patient similarity\nmeasure enables various downstream applications, such as cohort study and\ntreatment comparative effectiveness research. One major carrier for conducting\npatient similarity research is Electronic Health Records(EHRs), which are\nusually heterogeneous, longitudinal, and sparse. Though existing studies on\nlearning patient similarity from EHRs have shown being useful in solving real\nclinical problems, their applicability is limited due to the lack of medical\ninterpretations. Moreover, most previous methods assume a vector-based\nrepresentation for patients, which typically requires aggregation of medical\nevents over a certain time period. As a consequence, temporal information will\nbe lost. In this paper, we propose a patient similarity evaluation framework\nbased on the temporal matching of longitudinal patient EHRs. Two efficient\nmethods are presented, unsupervised and supervised, both of which preserve the\ntemporal properties in EHRs. The supervised scheme takes a convolutional neural\nnetwork architecture and learns an optimal representation of patient clinical\nrecords with medical concept embedding. The empirical results on real-world\nclinical data demonstrate substantial improvement over the baselines. We make\nour code and sample data available for further study.","url_abs":"http://arxiv.org/abs/1902.03376v1","url_pdf":"http://arxiv.org/pdf/1902.03376v1.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":"measuring-patient-similarities-via-a-deep","repo_url":"https://github.com/yinchangchang/patient_similarity","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"tf","reach":{"status":"ok"}}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1902.03376","atlas_url":"https://app.syntology.ai/?focus=1902.03376","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}