Probabilistic patient monitoring using extreme value theory: A multivariate, multimodal methodology for detecting patient deterioration
Conventional patient monitoring is performed by generating alarms when vital signs exceed pie-determined thresholds, but the false-alarm rate of such monitors in hospitals is so high that alarms are typically ignored. We propose a principled, probabilistic method for combining vital signs into a mul...
| Main Authors: | , , |
|---|---|
| Format: | Journal article |
| Language: | English |
| Published: |
2010
|
| _version_ | 1797054352793272320 |
|---|---|
| author | Hugueny, S Clifton, D Tarassenko, L |
| author_facet | Hugueny, S Clifton, D Tarassenko, L |
| author_sort | Hugueny, S |
| collection | OXFORD |
| description | Conventional patient monitoring is performed by generating alarms when vital signs exceed pie-determined thresholds, but the false-alarm rate of such monitors in hospitals is so high that alarms are typically ignored. We propose a principled, probabilistic method for combining vital signs into a multivariate model of patient state, using extreme value theory (EVT) to generate robust alarms if a patient's vital signs are deemed to have become sufficiently "extreme". Our proposed formulation operates many orders of magnitude faster than existing methods, allowing on-line learning of models, leading ultimately to patient-specific monitoring. |
| first_indexed | 2024-03-06T18:56:04Z |
| format | Journal article |
| id | oxford-uuid:11e240fc-e62a-4a5e-b795-9be2fa8443f4 |
| institution | University of Oxford |
| language | English |
| last_indexed | 2024-03-06T18:56:04Z |
| publishDate | 2010 |
| record_format | dspace |
| spelling | oxford-uuid:11e240fc-e62a-4a5e-b795-9be2fa8443f42022-03-26T10:04:42ZProbabilistic patient monitoring using extreme value theory: A multivariate, multimodal methodology for detecting patient deteriorationJournal articlehttp://purl.org/coar/resource_type/c_dcae04bcuuid:11e240fc-e62a-4a5e-b795-9be2fa8443f4EnglishSymplectic Elements at Oxford2010Hugueny, SClifton, DTarassenko, LConventional patient monitoring is performed by generating alarms when vital signs exceed pie-determined thresholds, but the false-alarm rate of such monitors in hospitals is so high that alarms are typically ignored. We propose a principled, probabilistic method for combining vital signs into a multivariate model of patient state, using extreme value theory (EVT) to generate robust alarms if a patient's vital signs are deemed to have become sufficiently "extreme". Our proposed formulation operates many orders of magnitude faster than existing methods, allowing on-line learning of models, leading ultimately to patient-specific monitoring. |
| spellingShingle | Hugueny, S Clifton, D Tarassenko, L Probabilistic patient monitoring using extreme value theory: A multivariate, multimodal methodology for detecting patient deterioration |
| title | Probabilistic patient monitoring using extreme value theory: A multivariate, multimodal methodology for detecting patient deterioration |
| title_full | Probabilistic patient monitoring using extreme value theory: A multivariate, multimodal methodology for detecting patient deterioration |
| title_fullStr | Probabilistic patient monitoring using extreme value theory: A multivariate, multimodal methodology for detecting patient deterioration |
| title_full_unstemmed | Probabilistic patient monitoring using extreme value theory: A multivariate, multimodal methodology for detecting patient deterioration |
| title_short | Probabilistic patient monitoring using extreme value theory: A multivariate, multimodal methodology for detecting patient deterioration |
| title_sort | probabilistic patient monitoring using extreme value theory a multivariate multimodal methodology for detecting patient deterioration |
| work_keys_str_mv | AT huguenys probabilisticpatientmonitoringusingextremevaluetheoryamultivariatemultimodalmethodologyfordetectingpatientdeterioration AT cliftond probabilisticpatientmonitoringusingextremevaluetheoryamultivariatemultimodalmethodologyfordetectingpatientdeterioration AT tarassenkol probabilisticpatientmonitoringusingextremevaluetheoryamultivariatemultimodalmethodologyfordetectingpatientdeterioration |