K-Means Clustering for Shock Classification in Pediatric Intensive Care Units
Shock is described as an inadequate oxygen supply to the tissues and can be classified in multiple ways. In clinical practice still, old methods are used to discriminate these shock types. This article proposes the application of unsupervised classification methods for the stratification of these pa...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-08-01
|
| Series: | Diagnostics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2075-4418/12/8/1932 |
| _version_ | 1797410715662811136 |
|---|---|
| author | María Rollán-Martínez-Herrera Jon Kerexeta-Sarriegi Javier Gil-Antón Javier Pilar-Orive Iván Macía-Oliver |
| author_facet | María Rollán-Martínez-Herrera Jon Kerexeta-Sarriegi Javier Gil-Antón Javier Pilar-Orive Iván Macía-Oliver |
| author_sort | María Rollán-Martínez-Herrera |
| collection | DOAJ |
| description | Shock is described as an inadequate oxygen supply to the tissues and can be classified in multiple ways. In clinical practice still, old methods are used to discriminate these shock types. This article proposes the application of unsupervised classification methods for the stratification of these patients in order to treat them more appropriately. With a cohort of 90 patients admitted in pediatric intensive care units (PICU), the k-means algorithm was applied in the first 24 h data since admission (physiological and analytical variables and the need for devices), obtaining three main groups. Significant differences were found in variables used (e.g., mean diastolic arterial pressure <i>p</i> < 0.001, age <i>p</i> < 0.001) and not used for training (e.g., EtCO2 min <i>p</i> < 0.001, Troponin max <i>p</i> < 0.01), discharge diagnosis (<i>p</i> < 0.001) and outcomes (<i>p</i> < 0.05). Clustering classification equaled classical classification in its association with LOS (<i>p</i> = 0.01) and surpassed it in its association with mortality (<i>p</i> < 0.04 vs. <i>p</i> = 0.16). We have been able to classify shocked pediatric patients with higher outcome correlation than the clinical traditional method. These results support the utility of unsupervised learning algorithms for patient classification in PICU. |
| first_indexed | 2024-03-09T04:34:28Z |
| format | Article |
| id | doaj.art-f8fa69e41bda4eafb4df619622e55102 |
| institution | Directory Open Access Journal |
| issn | 2075-4418 |
| language | English |
| last_indexed | 2024-03-09T04:34:28Z |
| publishDate | 2022-08-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Diagnostics |
| spelling | doaj.art-f8fa69e41bda4eafb4df619622e551022023-12-03T13:31:59ZengMDPI AGDiagnostics2075-44182022-08-01128193210.3390/diagnostics12081932K-Means Clustering for Shock Classification in Pediatric Intensive Care UnitsMaría Rollán-Martínez-Herrera0Jon Kerexeta-Sarriegi1Javier Gil-Antón2Javier Pilar-Orive3Iván Macía-Oliver4Cruces University Hospital, 48903 Barakaldo, SpainVicomtech Foundation, Basque Research and Technology Alliance (BRTA), Mikeletegi 57, 20009 Donostia, SpainCruces University Hospital, 48903 Barakaldo, SpainCruces University Hospital, 48903 Barakaldo, SpainVicomtech Foundation, Basque Research and Technology Alliance (BRTA), Mikeletegi 57, 20009 Donostia, SpainShock is described as an inadequate oxygen supply to the tissues and can be classified in multiple ways. In clinical practice still, old methods are used to discriminate these shock types. This article proposes the application of unsupervised classification methods for the stratification of these patients in order to treat them more appropriately. With a cohort of 90 patients admitted in pediatric intensive care units (PICU), the k-means algorithm was applied in the first 24 h data since admission (physiological and analytical variables and the need for devices), obtaining three main groups. Significant differences were found in variables used (e.g., mean diastolic arterial pressure <i>p</i> < 0.001, age <i>p</i> < 0.001) and not used for training (e.g., EtCO2 min <i>p</i> < 0.001, Troponin max <i>p</i> < 0.01), discharge diagnosis (<i>p</i> < 0.001) and outcomes (<i>p</i> < 0.05). Clustering classification equaled classical classification in its association with LOS (<i>p</i> = 0.01) and surpassed it in its association with mortality (<i>p</i> < 0.04 vs. <i>p</i> = 0.16). We have been able to classify shocked pediatric patients with higher outcome correlation than the clinical traditional method. These results support the utility of unsupervised learning algorithms for patient classification in PICU.https://www.mdpi.com/2075-4418/12/8/1932shockpediatricunsupervised learningk-meansstratification |
| spellingShingle | María Rollán-Martínez-Herrera Jon Kerexeta-Sarriegi Javier Gil-Antón Javier Pilar-Orive Iván Macía-Oliver K-Means Clustering for Shock Classification in Pediatric Intensive Care Units Diagnostics shock pediatric unsupervised learning k-means stratification |
| title | K-Means Clustering for Shock Classification in Pediatric Intensive Care Units |
| title_full | K-Means Clustering for Shock Classification in Pediatric Intensive Care Units |
| title_fullStr | K-Means Clustering for Shock Classification in Pediatric Intensive Care Units |
| title_full_unstemmed | K-Means Clustering for Shock Classification in Pediatric Intensive Care Units |
| title_short | K-Means Clustering for Shock Classification in Pediatric Intensive Care Units |
| title_sort | k means clustering for shock classification in pediatric intensive care units |
| topic | shock pediatric unsupervised learning k-means stratification |
| url | https://www.mdpi.com/2075-4418/12/8/1932 |
| work_keys_str_mv | AT mariarollanmartinezherrera kmeansclusteringforshockclassificationinpediatricintensivecareunits AT jonkerexetasarriegi kmeansclusteringforshockclassificationinpediatricintensivecareunits AT javiergilanton kmeansclusteringforshockclassificationinpediatricintensivecareunits AT javierpilarorive kmeansclusteringforshockclassificationinpediatricintensivecareunits AT ivanmaciaoliver kmeansclusteringforshockclassificationinpediatricintensivecareunits |