{"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/sparse-multi-output-gaussian-processes-for","title":"Sparse Multi-Output Gaussian Processes for Medical Time Series Prediction","arxiv_id":"1703.09112","date":"2017-03-27","proceeding":null,"authors":["Li-Fang Cheng","Gregory Darnell","Bianca Dumitrascu","Corey Chivers","Michael E Draugelis","Kai Li","Barbara E. Engelhardt"],"abstract":"In the scenario of real-time monitoring of hospital patients, high-quality\ninference of patients' health status using all information available from\nclinical covariates and lab tests is essential to enable successful medical\ninterventions and improve patient outcomes. Developing a computational\nframework that can learn from observational large-scale electronic health\nrecords (EHRs) and make accurate real-time predictions is a critical step. In\nthis work, we develop and explore a Bayesian nonparametric model based on\nGaussian process (GP) regression for hospital patient monitoring. We propose\nMedGP, a statistical framework that incorporates 24 clinical and lab covariates\nand supports a rich reference data set from which relationships between\nobserved covariates may be inferred and exploited for high-quality inference of\npatient state over time. To do this, we develop a highly structured sparse GP\nkernel to enable tractable computation over tens of thousands of time points\nwhile estimating correlations among clinical covariates, patients, and\nperiodicity in patient observations. MedGP has a number of benefits over\ncurrent methods, including (i) not requiring an alignment of the time series\ndata, (ii) quantifying confidence regions in the predictions, (iii) exploiting\na vast and rich database of patients, and (iv) inferring interpretable\nrelationships among clinical covariates. We evaluate and compare results from\nMedGP on the task of online prediction for three patient subgroups from two\nmedical data sets across 8,043 patients. We found MedGP improves online\nprediction over baseline methods for nearly all covariates across different\ndisease subgroups and studies. The publicly available code is at\nhttps://github.com/bee-hive/MedGP.","url_abs":"http://arxiv.org/abs/1703.09112v2","url_pdf":"http://arxiv.org/pdf/1703.09112v2.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":"sparse-multi-output-gaussian-processes-for","repo_url":"https://github.com/bee-hive/MedGP","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[{"task_slug":"gaussian-processes","task_name":"Gaussian Processes"},{"task_slug":"time-series-1","task_name":"Time Series"},{"task_slug":"time-series","task_name":"Time Series Analysis"},{"task_slug":"time-series-prediction","task_name":"Time Series Prediction"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1703.09112","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}