Machine learning predicts lung recruitment in acute respiratory distress syndrome using single lung CT scan
Abstract Background To develop and validate classifier models that could be used to identify patients with a high percentage of potentially recruitable lung from readily available clinical data and from single CT scan quantitative analysis at intensive care unit admission. 221 retrospectively enroll...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
SpringerOpen
2023-07-01
|
| Series: | Annals of Intensive Care |
| Subjects: | |
| Online Access: | https://doi.org/10.1186/s13613-023-01154-5 |
| _version_ | 1827708285558456320 |
|---|---|
| author | Francesca Pennati Andrea Aliverti Tommaso Pozzi Simone Gattarello Fabio Lombardo Silvia Coppola Davide Chiumello |
| author_facet | Francesca Pennati Andrea Aliverti Tommaso Pozzi Simone Gattarello Fabio Lombardo Silvia Coppola Davide Chiumello |
| author_sort | Francesca Pennati |
| collection | DOAJ |
| description | Abstract Background To develop and validate classifier models that could be used to identify patients with a high percentage of potentially recruitable lung from readily available clinical data and from single CT scan quantitative analysis at intensive care unit admission. 221 retrospectively enrolled mechanically ventilated, sedated and paralyzed patients with acute respiratory distress syndrome (ARDS) underwent a PEEP trial at 5 and 15 cmH2O of PEEP and two lung CT scans performed at 5 and 45 cmH2O of airway pressure. Lung recruitability was defined at first as percent change in not aerated tissue between 5 and 45 cmH2O (radiologically defined; recruiters: Δ45-5non-aerated tissue > 15%) and secondly as change in PaO2 between 5 and 15 cmH2O (gas exchange-defined; recruiters: Δ15-5PaO2 > 24 mmHg). Four machine learning (ML) algorithms were evaluated as classifiers of radiologically defined and gas exchange-defined lung recruiters using different models including different variables, separately or combined, of lung mechanics, gas exchange and CT data. Results ML algorithms based on CT scan data at 5 cmH2O classified radiologically defined lung recruiters with similar AUC as ML based on the combination of lung mechanics, gas exchange and CT data. ML algorithm based on CT scan data classified gas exchange-defined lung recruiters with the highest AUC. Conclusions ML based on a single CT data at 5 cmH2O represented an easy-to-apply tool to classify ARDS patients in recruiters and non-recruiters according to both radiologically defined and gas exchange-defined lung recruitment within the first 48 h from the start of mechanical ventilation. |
| first_indexed | 2024-03-10T17:02:30Z |
| format | Article |
| id | doaj.art-0bbbee0d982d4ac9be49c94fc1629384 |
| institution | Directory Open Access Journal |
| issn | 2110-5820 |
| language | English |
| last_indexed | 2024-03-10T17:02:30Z |
| publishDate | 2023-07-01 |
| publisher | SpringerOpen |
| record_format | Article |
| series | Annals of Intensive Care |
| spelling | doaj.art-0bbbee0d982d4ac9be49c94fc16293842023-11-20T10:54:39ZengSpringerOpenAnnals of Intensive Care2110-58202023-07-0113111110.1186/s13613-023-01154-5Machine learning predicts lung recruitment in acute respiratory distress syndrome using single lung CT scanFrancesca Pennati0Andrea Aliverti1Tommaso Pozzi2Simone Gattarello3Fabio Lombardo4Silvia Coppola5Davide Chiumello6Ipartimento di Elettronica, Informazione e Bioingegneria, Politecnico di MilanoIpartimento di Elettronica, Informazione e Bioingegneria, Politecnico di MilanoDepartment of Health Sciences, University of MilanDepartment of Anesthesiology, University Medical Center GöttingenDepartment of Anesthesiology, University Medical Center GöttingenDepartment of Anesthesia and Intensive Care, ASST Santi Paolo e Carlo, San Paolo University HospitalDepartment of Health Sciences, University of MilanAbstract Background To develop and validate classifier models that could be used to identify patients with a high percentage of potentially recruitable lung from readily available clinical data and from single CT scan quantitative analysis at intensive care unit admission. 221 retrospectively enrolled mechanically ventilated, sedated and paralyzed patients with acute respiratory distress syndrome (ARDS) underwent a PEEP trial at 5 and 15 cmH2O of PEEP and two lung CT scans performed at 5 and 45 cmH2O of airway pressure. Lung recruitability was defined at first as percent change in not aerated tissue between 5 and 45 cmH2O (radiologically defined; recruiters: Δ45-5non-aerated tissue > 15%) and secondly as change in PaO2 between 5 and 15 cmH2O (gas exchange-defined; recruiters: Δ15-5PaO2 > 24 mmHg). Four machine learning (ML) algorithms were evaluated as classifiers of radiologically defined and gas exchange-defined lung recruiters using different models including different variables, separately or combined, of lung mechanics, gas exchange and CT data. Results ML algorithms based on CT scan data at 5 cmH2O classified radiologically defined lung recruiters with similar AUC as ML based on the combination of lung mechanics, gas exchange and CT data. ML algorithm based on CT scan data classified gas exchange-defined lung recruiters with the highest AUC. Conclusions ML based on a single CT data at 5 cmH2O represented an easy-to-apply tool to classify ARDS patients in recruiters and non-recruiters according to both radiologically defined and gas exchange-defined lung recruitment within the first 48 h from the start of mechanical ventilation.https://doi.org/10.1186/s13613-023-01154-5ARDSMachine learningTomography |
| spellingShingle | Francesca Pennati Andrea Aliverti Tommaso Pozzi Simone Gattarello Fabio Lombardo Silvia Coppola Davide Chiumello Machine learning predicts lung recruitment in acute respiratory distress syndrome using single lung CT scan Annals of Intensive Care ARDS Machine learning Tomography |
| title | Machine learning predicts lung recruitment in acute respiratory distress syndrome using single lung CT scan |
| title_full | Machine learning predicts lung recruitment in acute respiratory distress syndrome using single lung CT scan |
| title_fullStr | Machine learning predicts lung recruitment in acute respiratory distress syndrome using single lung CT scan |
| title_full_unstemmed | Machine learning predicts lung recruitment in acute respiratory distress syndrome using single lung CT scan |
| title_short | Machine learning predicts lung recruitment in acute respiratory distress syndrome using single lung CT scan |
| title_sort | machine learning predicts lung recruitment in acute respiratory distress syndrome using single lung ct scan |
| topic | ARDS Machine learning Tomography |
| url | https://doi.org/10.1186/s13613-023-01154-5 |
| work_keys_str_mv | AT francescapennati machinelearningpredictslungrecruitmentinacuterespiratorydistresssyndromeusingsinglelungctscan AT andreaaliverti machinelearningpredictslungrecruitmentinacuterespiratorydistresssyndromeusingsinglelungctscan AT tommasopozzi machinelearningpredictslungrecruitmentinacuterespiratorydistresssyndromeusingsinglelungctscan AT simonegattarello machinelearningpredictslungrecruitmentinacuterespiratorydistresssyndromeusingsinglelungctscan AT fabiolombardo machinelearningpredictslungrecruitmentinacuterespiratorydistresssyndromeusingsinglelungctscan AT silviacoppola machinelearningpredictslungrecruitmentinacuterespiratorydistresssyndromeusingsinglelungctscan AT davidechiumello machinelearningpredictslungrecruitmentinacuterespiratorydistresssyndromeusingsinglelungctscan |