Delirium detection using wearable sensors and machine learning in patients with intracerebral hemorrhage
ObjectiveDelirium is associated with worse outcomes in patients with stroke and neurocritical illness, but delirium detection in these patients can be challenging with existing screening tools. To address this gap, we aimed to develop and evaluate machine learning models that detect episodes of post...
| Main Authors: | , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2023-06-01
|
| Series: | Frontiers in Neurology |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fneur.2023.1135472/full |
| _version_ | 1797808343877681152 |
|---|---|
| author | Abdullah Ahmed Augusto Garcia-Agundez Augusto Garcia-Agundez Ivana Petrovic Fatemeh Radaei James Fife John Zhou Hunter Karas Scott Moody Jonathan Drake Richard N. Jones Carsten Eickhoff Michael E. Reznik |
| author_facet | Abdullah Ahmed Augusto Garcia-Agundez Augusto Garcia-Agundez Ivana Petrovic Fatemeh Radaei James Fife John Zhou Hunter Karas Scott Moody Jonathan Drake Richard N. Jones Carsten Eickhoff Michael E. Reznik |
| author_sort | Abdullah Ahmed |
| collection | DOAJ |
| description | ObjectiveDelirium is associated with worse outcomes in patients with stroke and neurocritical illness, but delirium detection in these patients can be challenging with existing screening tools. To address this gap, we aimed to develop and evaluate machine learning models that detect episodes of post-stroke delirium based on data from wearable activity monitors in conjunction with stroke-related clinical features.DesignProspective observational cohort study.SettingNeurocritical Care and Stroke Units at an academic medical center.PatientsWe recruited 39 patients with moderate-to-severe acute intracerebral hemorrhage (ICH) and hemiparesis over a 1-year period [mean (SD) age 71.3 (12.20), 54% male, median (IQR) initial NIH Stroke Scale 14.5 (6), median (IQR) ICH score 2 (1)].Measurements and main resultsEach patient received daily assessments for delirium by an attending neurologist, while activity data were recorded throughout each patient's hospitalization using wrist-worn actigraph devices (on both paretic and non-paretic arms). We compared the predictive accuracy of Random Forest, SVM and XGBoost machine learning methods in classifying daily delirium status using clinical information alone and combined with actigraph data. Among our study cohort, 85% of patients (n = 33) had at least one delirium episode, while 71% of monitoring days (n = 209) were rated as days with delirium. Clinical information alone had a low accuracy in detecting delirium on a day-to-day basis [accuracy mean (SD) 62% (18%), F1 score mean (SD) 50% (17%)]. Prediction performance improved significantly (p < 0.001) with the addition of actigraph data [accuracy mean (SD) 74% (10%), F1 score 65% (10%)]. Among actigraphy features, night-time actigraph data were especially relevant for classification accuracy.ConclusionsWe found that actigraphy in conjunction with machine learning models improves clinical detection of delirium in patients with stroke, thus paving the way to make actigraph-assisted predictions clinically actionable. |
| first_indexed | 2024-03-13T06:36:08Z |
| format | Article |
| id | doaj.art-8c1286582cec458fb6ad0082b5914c96 |
| institution | Directory Open Access Journal |
| issn | 1664-2295 |
| language | English |
| last_indexed | 2024-03-13T06:36:08Z |
| publishDate | 2023-06-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Neurology |
| spelling | doaj.art-8c1286582cec458fb6ad0082b5914c962023-06-09T05:05:40ZengFrontiers Media S.A.Frontiers in Neurology1664-22952023-06-011410.3389/fneur.2023.11354721135472Delirium detection using wearable sensors and machine learning in patients with intracerebral hemorrhageAbdullah Ahmed0Augusto Garcia-Agundez1Augusto Garcia-Agundez2Ivana Petrovic3Fatemeh Radaei4James Fife5John Zhou6Hunter Karas7Scott Moody8Jonathan Drake9Richard N. Jones10Carsten Eickhoff11Michael E. Reznik12Brown Center for Biomedical Informatics, Brown University, Providence, RI, United StatesBrown Center for Biomedical Informatics, Brown University, Providence, RI, United StatesIMDEA Networks Institute, Madrid, SpainBrown Center for Biomedical Informatics, Brown University, Providence, RI, United StatesBrown Center for Biomedical Informatics, Brown University, Providence, RI, United StatesBrown Center for Biomedical Informatics, Brown University, Providence, RI, United StatesBrown Center for Biomedical Informatics, Brown University, Providence, RI, United StatesBrown Center for Biomedical Informatics, Brown University, Providence, RI, United StatesDepartment of Neurology, Brown University, Providence, RI, United StatesDepartment of Neurology, Brown University, Providence, RI, United StatesDepartment of Psychiatry, Brown University, Providence, RI, United StatesBrown Center for Biomedical Informatics, Brown University, Providence, RI, United StatesDepartment of Neurology, Brown University, Providence, RI, United StatesObjectiveDelirium is associated with worse outcomes in patients with stroke and neurocritical illness, but delirium detection in these patients can be challenging with existing screening tools. To address this gap, we aimed to develop and evaluate machine learning models that detect episodes of post-stroke delirium based on data from wearable activity monitors in conjunction with stroke-related clinical features.DesignProspective observational cohort study.SettingNeurocritical Care and Stroke Units at an academic medical center.PatientsWe recruited 39 patients with moderate-to-severe acute intracerebral hemorrhage (ICH) and hemiparesis over a 1-year period [mean (SD) age 71.3 (12.20), 54% male, median (IQR) initial NIH Stroke Scale 14.5 (6), median (IQR) ICH score 2 (1)].Measurements and main resultsEach patient received daily assessments for delirium by an attending neurologist, while activity data were recorded throughout each patient's hospitalization using wrist-worn actigraph devices (on both paretic and non-paretic arms). We compared the predictive accuracy of Random Forest, SVM and XGBoost machine learning methods in classifying daily delirium status using clinical information alone and combined with actigraph data. Among our study cohort, 85% of patients (n = 33) had at least one delirium episode, while 71% of monitoring days (n = 209) were rated as days with delirium. Clinical information alone had a low accuracy in detecting delirium on a day-to-day basis [accuracy mean (SD) 62% (18%), F1 score mean (SD) 50% (17%)]. Prediction performance improved significantly (p < 0.001) with the addition of actigraph data [accuracy mean (SD) 74% (10%), F1 score 65% (10%)]. Among actigraphy features, night-time actigraph data were especially relevant for classification accuracy.ConclusionsWe found that actigraphy in conjunction with machine learning models improves clinical detection of delirium in patients with stroke, thus paving the way to make actigraph-assisted predictions clinically actionable.https://www.frontiersin.org/articles/10.3389/fneur.2023.1135472/fulldeliriumneurocritical carestrokeintracerebral hemorrhageactigraphymachine learning |
| spellingShingle | Abdullah Ahmed Augusto Garcia-Agundez Augusto Garcia-Agundez Ivana Petrovic Fatemeh Radaei James Fife John Zhou Hunter Karas Scott Moody Jonathan Drake Richard N. Jones Carsten Eickhoff Michael E. Reznik Delirium detection using wearable sensors and machine learning in patients with intracerebral hemorrhage Frontiers in Neurology delirium neurocritical care stroke intracerebral hemorrhage actigraphy machine learning |
| title | Delirium detection using wearable sensors and machine learning in patients with intracerebral hemorrhage |
| title_full | Delirium detection using wearable sensors and machine learning in patients with intracerebral hemorrhage |
| title_fullStr | Delirium detection using wearable sensors and machine learning in patients with intracerebral hemorrhage |
| title_full_unstemmed | Delirium detection using wearable sensors and machine learning in patients with intracerebral hemorrhage |
| title_short | Delirium detection using wearable sensors and machine learning in patients with intracerebral hemorrhage |
| title_sort | delirium detection using wearable sensors and machine learning in patients with intracerebral hemorrhage |
| topic | delirium neurocritical care stroke intracerebral hemorrhage actigraphy machine learning |
| url | https://www.frontiersin.org/articles/10.3389/fneur.2023.1135472/full |
| work_keys_str_mv | AT abdullahahmed deliriumdetectionusingwearablesensorsandmachinelearninginpatientswithintracerebralhemorrhage AT augustogarciaagundez deliriumdetectionusingwearablesensorsandmachinelearninginpatientswithintracerebralhemorrhage AT augustogarciaagundez deliriumdetectionusingwearablesensorsandmachinelearninginpatientswithintracerebralhemorrhage AT ivanapetrovic deliriumdetectionusingwearablesensorsandmachinelearninginpatientswithintracerebralhemorrhage AT fatemehradaei deliriumdetectionusingwearablesensorsandmachinelearninginpatientswithintracerebralhemorrhage AT jamesfife deliriumdetectionusingwearablesensorsandmachinelearninginpatientswithintracerebralhemorrhage AT johnzhou deliriumdetectionusingwearablesensorsandmachinelearninginpatientswithintracerebralhemorrhage AT hunterkaras deliriumdetectionusingwearablesensorsandmachinelearninginpatientswithintracerebralhemorrhage AT scottmoody deliriumdetectionusingwearablesensorsandmachinelearninginpatientswithintracerebralhemorrhage AT jonathandrake deliriumdetectionusingwearablesensorsandmachinelearninginpatientswithintracerebralhemorrhage AT richardnjones deliriumdetectionusingwearablesensorsandmachinelearninginpatientswithintracerebralhemorrhage AT carsteneickhoff deliriumdetectionusingwearablesensorsandmachinelearninginpatientswithintracerebralhemorrhage AT michaelereznik deliriumdetectionusingwearablesensorsandmachinelearninginpatientswithintracerebralhemorrhage |