Patient-specific physiological monitoring and prediction using structured Gaussian processes
The management of patient well-being can be performed by monitoring continuous time-series vital-sign data via low-cost wearable devices. Automated algorithms may then be used with the resulting data to provide early warning of deterioration of the health of an individual. Such algorithms are typica...
| Main Authors: | , , , , , , |
|---|---|
| Formato: | Journal article |
| Publicado em: |
Institute of Electrical and Electronics Engineers
2019
|
| _version_ | 1826257925096603648 |
|---|---|
| author | Zhu, T Wright Colopy, G Macewen, C Niehaus, K Yang, Y Pugh, C Clifton, D |
| author_facet | Zhu, T Wright Colopy, G Macewen, C Niehaus, K Yang, Y Pugh, C Clifton, D |
| author_sort | Zhu, T |
| collection | OXFORD |
| description | The management of patient well-being can be performed by monitoring continuous time-series vital-sign data via low-cost wearable devices. Automated algorithms may then be used with the resulting data to provide early warning of deterioration of the health of an individual. Such algorithms are typically trained for a large population without considering the time-variability and inter-subject variability of the data being collected. In the case where limited numbers of subjects are available, it is difficult to create a generalized population model from a small sample size. Furthermore, some “normal” patients may exhibit different physiological patterns when compared to other “normal” patients, forming multiple “normal” clusters/subgroups. This also makes inferring a population model difficult. It is, therefore, preferable to develop patient/subgroup-specific time-series models to overcome these challenges. We propose using Bayesian hierarchical Gaussian processes to infer the hidden latent structure of the vital sign's trajectory for each individual patient or group of patients who share similar patterns. We further demonstrate the feasibility of such a model in novelty detection, using the symmetric Kullback-Leibler divergence. This allows us to identify any patterns that correspond to “normal” or “abnormal” physiology, and further classifying “abnormal” patterns from a model of “normal” latent trajectories. We tested our approach using two real datasets for different monitoring scenarios. Our model was compared to the performance of the state-of-the-art unsupervised clustering algorithms, demonstrating at least 10% improvement in accuracy. We further benchmarked against two one-class classifiers and showed at least 5% accuracy improvement when using the proposed metrics in identifying abnormal physiological episodes. |
| first_indexed | 2024-03-06T18:25:53Z |
| format | Journal article |
| id | oxford-uuid:07f39a69-1502-425e-92b6-1b2132ca840d |
| institution | University of Oxford |
| last_indexed | 2024-03-06T18:25:53Z |
| publishDate | 2019 |
| publisher | Institute of Electrical and Electronics Engineers |
| record_format | dspace |
| spelling | oxford-uuid:07f39a69-1502-425e-92b6-1b2132ca840d2022-03-26T09:10:14ZPatient-specific physiological monitoring and prediction using structured Gaussian processesJournal articlehttp://purl.org/coar/resource_type/c_dcae04bcuuid:07f39a69-1502-425e-92b6-1b2132ca840dSymplectic Elements at OxfordInstitute of Electrical and Electronics Engineers2019Zhu, TWright Colopy, GMacewen, CNiehaus, KYang, YPugh, CClifton, DThe management of patient well-being can be performed by monitoring continuous time-series vital-sign data via low-cost wearable devices. Automated algorithms may then be used with the resulting data to provide early warning of deterioration of the health of an individual. Such algorithms are typically trained for a large population without considering the time-variability and inter-subject variability of the data being collected. In the case where limited numbers of subjects are available, it is difficult to create a generalized population model from a small sample size. Furthermore, some “normal” patients may exhibit different physiological patterns when compared to other “normal” patients, forming multiple “normal” clusters/subgroups. This also makes inferring a population model difficult. It is, therefore, preferable to develop patient/subgroup-specific time-series models to overcome these challenges. We propose using Bayesian hierarchical Gaussian processes to infer the hidden latent structure of the vital sign's trajectory for each individual patient or group of patients who share similar patterns. We further demonstrate the feasibility of such a model in novelty detection, using the symmetric Kullback-Leibler divergence. This allows us to identify any patterns that correspond to “normal” or “abnormal” physiology, and further classifying “abnormal” patterns from a model of “normal” latent trajectories. We tested our approach using two real datasets for different monitoring scenarios. Our model was compared to the performance of the state-of-the-art unsupervised clustering algorithms, demonstrating at least 10% improvement in accuracy. We further benchmarked against two one-class classifiers and showed at least 5% accuracy improvement when using the proposed metrics in identifying abnormal physiological episodes. |
| spellingShingle | Zhu, T Wright Colopy, G Macewen, C Niehaus, K Yang, Y Pugh, C Clifton, D Patient-specific physiological monitoring and prediction using structured Gaussian processes |
| title | Patient-specific physiological monitoring and prediction using structured Gaussian processes |
| title_full | Patient-specific physiological monitoring and prediction using structured Gaussian processes |
| title_fullStr | Patient-specific physiological monitoring and prediction using structured Gaussian processes |
| title_full_unstemmed | Patient-specific physiological monitoring and prediction using structured Gaussian processes |
| title_short | Patient-specific physiological monitoring and prediction using structured Gaussian processes |
| title_sort | patient specific physiological monitoring and prediction using structured gaussian processes |
| work_keys_str_mv | AT zhut patientspecificphysiologicalmonitoringandpredictionusingstructuredgaussianprocesses AT wrightcolopyg patientspecificphysiologicalmonitoringandpredictionusingstructuredgaussianprocesses AT macewenc patientspecificphysiologicalmonitoringandpredictionusingstructuredgaussianprocesses AT niehausk patientspecificphysiologicalmonitoringandpredictionusingstructuredgaussianprocesses AT yangy patientspecificphysiologicalmonitoringandpredictionusingstructuredgaussianprocesses AT pughc patientspecificphysiologicalmonitoringandpredictionusingstructuredgaussianprocesses AT cliftond patientspecificphysiologicalmonitoringandpredictionusingstructuredgaussianprocesses |