Early Prediction of Hemodynamic Shock in Pediatric Intensive Care Units With Deep Learning on Thermal Videos
Shock is one of the major killers in intensive care units, and early interventions can potentially reverse it. In this study, we advance a noncontact thermal imaging modality for continuous monitoring of hemodynamic shock working on 1,03,936 frames from 406 videos recorded longitudinally upon 22 ped...
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2022-07-01
|
| Series: | Frontiers in Physiology |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fphys.2022.862411/full |
| _version_ | 1818047302510575616 |
|---|---|
| author | Vanshika Vats Aditya Nagori Aditya Nagori Aditya Nagori Pradeep Singh Raman Dutt Harsh Bandhey Mahika Wason Rakesh Lodha Tavpritesh Sethi Tavpritesh Sethi |
| author_facet | Vanshika Vats Aditya Nagori Aditya Nagori Aditya Nagori Pradeep Singh Raman Dutt Harsh Bandhey Mahika Wason Rakesh Lodha Tavpritesh Sethi Tavpritesh Sethi |
| author_sort | Vanshika Vats |
| collection | DOAJ |
| description | Shock is one of the major killers in intensive care units, and early interventions can potentially reverse it. In this study, we advance a noncontact thermal imaging modality for continuous monitoring of hemodynamic shock working on 1,03,936 frames from 406 videos recorded longitudinally upon 22 pediatric patients. Deep learning was used to preprocess and extract the Center-to-Peripheral Difference (CPD) in temperature values from the videos. This time-series data along with the heart rate was finally analyzed using Long-Short Term Memory models to predict the shock status up to the next 6 h. Our models achieved the best area under the receiver operating characteristic curve of 0.81 ± 0.06 and area under the precision-recall curve of 0.78 ± 0.05 at 5 h, providing sufficient time to stabilize the patient. Our approach, thus, provides a reliable shock prediction using an automated decision pipeline that can provide better care and save lives. |
| first_indexed | 2024-12-10T10:03:38Z |
| format | Article |
| id | doaj.art-d773e44383f14a78bd2eab0edcdcd7d1 |
| institution | Directory Open Access Journal |
| issn | 1664-042X |
| language | English |
| last_indexed | 2024-12-10T10:03:38Z |
| publishDate | 2022-07-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Physiology |
| spelling | doaj.art-d773e44383f14a78bd2eab0edcdcd7d12022-12-22T01:53:17ZengFrontiers Media S.A.Frontiers in Physiology1664-042X2022-07-011310.3389/fphys.2022.862411862411Early Prediction of Hemodynamic Shock in Pediatric Intensive Care Units With Deep Learning on Thermal VideosVanshika Vats0Aditya Nagori1Aditya Nagori2Aditya Nagori3Pradeep Singh4Raman Dutt5Harsh Bandhey6Mahika Wason7Rakesh Lodha8Tavpritesh Sethi9Tavpritesh Sethi10Indraprastha Institute of Information Technology, Delhi, IndiaIndraprastha Institute of Information Technology, Delhi, IndiaCSIR-Institute of Genomics and Integrative Biology, New Delhi, IndiaAcademy of Scientific and Innovative Research (AcSIR), Ghaziabad, IndiaIndraprastha Institute of Information Technology, Delhi, IndiaComputer Science and Engineering, Shiv Nadar University, Greater Noida, IndiaIndraprastha Institute of Information Technology, Delhi, IndiaIndraprastha Institute of Information Technology, Delhi, IndiaDepartment of Pediatrics, All India Institute of Medical Sciences, New Delhi, IndiaIndraprastha Institute of Information Technology, Delhi, IndiaDepartment of Pediatrics, All India Institute of Medical Sciences, New Delhi, IndiaShock is one of the major killers in intensive care units, and early interventions can potentially reverse it. In this study, we advance a noncontact thermal imaging modality for continuous monitoring of hemodynamic shock working on 1,03,936 frames from 406 videos recorded longitudinally upon 22 pediatric patients. Deep learning was used to preprocess and extract the Center-to-Peripheral Difference (CPD) in temperature values from the videos. This time-series data along with the heart rate was finally analyzed using Long-Short Term Memory models to predict the shock status up to the next 6 h. Our models achieved the best area under the receiver operating characteristic curve of 0.81 ± 0.06 and area under the precision-recall curve of 0.78 ± 0.05 at 5 h, providing sufficient time to stabilize the patient. Our approach, thus, provides a reliable shock prediction using an automated decision pipeline that can provide better care and save lives.https://www.frontiersin.org/articles/10.3389/fphys.2022.862411/fullhemodynamic shockdeep learningICU—intensive care unitartificial intelligencethermal imagingcomputer vision |
| spellingShingle | Vanshika Vats Aditya Nagori Aditya Nagori Aditya Nagori Pradeep Singh Raman Dutt Harsh Bandhey Mahika Wason Rakesh Lodha Tavpritesh Sethi Tavpritesh Sethi Early Prediction of Hemodynamic Shock in Pediatric Intensive Care Units With Deep Learning on Thermal Videos Frontiers in Physiology hemodynamic shock deep learning ICU—intensive care unit artificial intelligence thermal imaging computer vision |
| title | Early Prediction of Hemodynamic Shock in Pediatric Intensive Care Units With Deep Learning on Thermal Videos |
| title_full | Early Prediction of Hemodynamic Shock in Pediatric Intensive Care Units With Deep Learning on Thermal Videos |
| title_fullStr | Early Prediction of Hemodynamic Shock in Pediatric Intensive Care Units With Deep Learning on Thermal Videos |
| title_full_unstemmed | Early Prediction of Hemodynamic Shock in Pediatric Intensive Care Units With Deep Learning on Thermal Videos |
| title_short | Early Prediction of Hemodynamic Shock in Pediatric Intensive Care Units With Deep Learning on Thermal Videos |
| title_sort | early prediction of hemodynamic shock in pediatric intensive care units with deep learning on thermal videos |
| topic | hemodynamic shock deep learning ICU—intensive care unit artificial intelligence thermal imaging computer vision |
| url | https://www.frontiersin.org/articles/10.3389/fphys.2022.862411/full |
| work_keys_str_mv | AT vanshikavats earlypredictionofhemodynamicshockinpediatricintensivecareunitswithdeeplearningonthermalvideos AT adityanagori earlypredictionofhemodynamicshockinpediatricintensivecareunitswithdeeplearningonthermalvideos AT adityanagori earlypredictionofhemodynamicshockinpediatricintensivecareunitswithdeeplearningonthermalvideos AT adityanagori earlypredictionofhemodynamicshockinpediatricintensivecareunitswithdeeplearningonthermalvideos AT pradeepsingh earlypredictionofhemodynamicshockinpediatricintensivecareunitswithdeeplearningonthermalvideos AT ramandutt earlypredictionofhemodynamicshockinpediatricintensivecareunitswithdeeplearningonthermalvideos AT harshbandhey earlypredictionofhemodynamicshockinpediatricintensivecareunitswithdeeplearningonthermalvideos AT mahikawason earlypredictionofhemodynamicshockinpediatricintensivecareunitswithdeeplearningonthermalvideos AT rakeshlodha earlypredictionofhemodynamicshockinpediatricintensivecareunitswithdeeplearningonthermalvideos AT tavpriteshsethi earlypredictionofhemodynamicshockinpediatricintensivecareunitswithdeeplearningonthermalvideos AT tavpriteshsethi earlypredictionofhemodynamicshockinpediatricintensivecareunitswithdeeplearningonthermalvideos |