ARDS (Acute Respiratory Distress Syndrome)NCT07328997
Exploration of Diagnosis and Treatment Strategies and Prognostic Prediction Models for Acute Respiratory Distress Syndrome Based on Radiographic Evaluations Assessed by Artificial Intelligence
An observational study of ARDS (Acute Respiratory Distress Syndrome) and AI (Artificial Intelligence), run by Shanghai Zhongshan Hospital at 1 site in China.
Trial summary
By using multi-center chest CT data, an intelligent assessment model for the severity of ARDS was constructed. Based on CT quantitative features and clinical characteristics, a prediction model for short-term critical events (such as mechanical ventilation decisions, prone position strategies, death, ECMO use, etc.)… Read more about this study
- Who is it for?
- Ages 18 to 100 years
- Conditions the study is for
- ARDS (Acute Respiratory Distress Syndrome), AI (Artificial Intelligence)
- What is being tested?
- CT scan
- How many people
- 400 participants
- Where
- 1 siteChina
- Study dates
- 31 May 2024to 30 Nov 2025
- Phase
- Observational
- Are there results?
- None posted in the registry
Key requirements
Who can take part, as listed by the sponsor. Only the study team can confirm whether you are eligible, so take this list to your doctor.
- Ages 18 to 100 years
- All sexes
- No healthy volunteers
- Adult, Older adult
Study population: The study population consists of adult patients diagnosed with acute respiratory distress syndrome (ARDS) who were admitted to the intensive care units (ICUs) of three tertiary comprehensive hospitals in China. Eligible participants are retrospectively identified from electronic medical records between January 2020 and December 2024. All included patients meet the predefined inclusion and exclusion criteria based on the 2023 Global ARDS Definition and have available chest CT imaging and corresponding clinical data. This cohort represents a real-world ICU population with diverse etiologies of ARDS and varying degrees of disease severity.
Who may be able to join
- Meets the diagnostic criteria for ARDS
- Be admitted to the intensive care unit
- There are chest CT images
Who cannot join
- Age less than 18 years old
- Missing medical records
- No chest CT images
Trial locations and contacts
1 site in 1 country.
China1 site
Department of critical care medicine, Zhongshan Hospital, Fudan University
Shanghai, Fengling Rd, 200032, P. R.
About this study
By using multi-center chest CT data, an intelligent assessment model for the severity of ARDS was constructed. Based on CT quantitative features and clinical characteristics, a prediction model for short-term critical events (such as mechanical ventilation decisions, prone position strategies, death, ECMO use, etc.) was established. The disease was staged and quantified, and a diagnosis and risk stratification model for ARDS was developed to assist in guiding the diagnosis and treatment strategies for ARDS.
- Being tested
- CT scan
- Keywords
- ARDS, chest CT, AI models
- Medical subject headings
- Acute Lung Injury
What participants receive
What participants receive. In a randomised study each person is assigned to one group by chance.
CT scanDiagnostic test
CT scan
| Study group | Role | What they receive |
|---|---|---|
| Training group, testing group, validation group | The study adopts a stratified random sampling strategy with an 8:2 split to construct training and internal validation datasets, together with an independent external test cohort from a separate center. No randomization of clinical interventions or treatments is involved. The model will be developed and evaluated using observational data derived from real-world clinical pathways and outcomes, with the objectives of assessing performance in disease severity stratification, treatment recommendation, and mortality prediction. Model performance will be compared with established ICU severity scores and existing AI-based approaches according to a prespecified statistical analysis plan. Receives: Diagnostic Test: CT scan |
What is being measured
The primary outcome is the main question the study is designed to answer.
Accuracy of ARDS severity classification
Accuracy of the artificial intelligence-based model in classifying ARDS severity (mild, moderate, or severe), using the reference clinical classification defined by the 2023 global ARDS criteria as the ground truth.
Measured: Baseline, defined as within 24 hours of index chest CT acquisition during ICU admission.
Treatment plan matching rate between model-recommended and actual clinical management.
Concordance rate between model-recommended treatment strategies and actual clinical management decisions across five predefined intervention modalities: mechanical ventilation, high-flow nasal oxygen therapy, non-invasive ventilation, prone positioning, and neuromuscular blockade.
Measured: Baseline, defined as within 24 hours of index chest CT acquisition during ICU admission.
Accuracy of 28-day in-hospital mortality prediction.
Accuracy of the model in predicting all-cause in-hospital mortality within 28 days, based on integrated chest CT imaging features and clinical variables.
Measured: Up to 28 days from ICU admission, or until hospital discharge, whichever occurs first.
Secondary outcomes (4)
Comparative performance improvement over baseline AI models.
Absolute performance improvement of the proposed model compared with three commonly used baseline artificial intelligence models across ARDS severity classification, treatment plan matching, and 28-day mortality prediction.
Measured: Baseline for severity classification and treatment plan matching; up to 28 days from ICU admission for mortality prediction
Calibration performance of 28-day mortality prediction.
Calibration of the mortality prediction model assessed using calibration curves and calibration statistics to evaluate agreement between predicted and observed 28-day in-hospital mortality.
Measured: Up to 28 days from ICU admission, or until hospital discharge, whichever occurs first.
Model interpretability based on imaging and clinical feature contributions.
Quantification of the relative contributions of imaging-derived features and clinical variables to mortality prediction using Shapley Additive Explanations (SHAP). Feature importance will be analyzed overall and stratified by ARDS severity.
Measured: Baseline for feature extraction; up to 28 days from ICU admission for outcome association analysis.
Association between treatment concordance and 28-day in-hospital mortality.
Association between concordance of model-recommended interventions and actual clinical treatments and 28-day in-hospital mortality, evaluated using multivariable logistic regression adjusted for key imaging-derived structural metrics.
Measured: Up to 28 days from ICU admission, or until hospital discharge, whichever occurs first.
Timeline
From the start of the study to its completion; the mark is today. Estimated dates are the sponsor’s plan.
- First posted9 Jan 2026
- Study start31 May 2024
- Primary completion30 Nov 2025
- Study completion30 Nov 2025
Resources
- Official record on ClinicalTrials.govThe registry entry this page is built from, with its full change history.
- Publications (30)Papers the sponsor has linked to this study.
- Other trials for ARDS (Acute Respiratory Distress Syndrome)Studies of the same condition, open and under way.
- More trials run by Shanghai Zhongshan HospitalThe sponsor's other registered studies.
- What the trial phases meanHow a treatment moves from first tests in people to approval.
More details
For doctors and researchers: how the study is designed, who runs it, and its registry IDs.
Study design
- Study type
- Observational
- Observational model
- Case only
- Time perspective
- Retrospective
- Enrolment
- 400 participants (actual)
Sponsor and investigators
- Lead sponsor
- Shanghai Zhongshan Hospital
- Sponsor type
- University, hospital or other
- Responsible party
- Sponsor
IDs, dates and publications
- ClinicalTrials.gov ID
- NCT07328997
- Sponsor's protocol ID
- B2024-180
- Status last verified
- Dec 2025
- First submitted
- 13 Dec 2025
- Last updated
- 9 Jan 2026
- Results posted
- No
- Data monitoring committee
- Yes
- FDA-regulated drug
- No
- FDA-regulated device
- No
- Individual data sharing
- Yes: Individual participant data that underlie the results of the study, including de-identified demographic information, clinical variables, laboratory findings, ventilator or HFNC parameters, and AI-derived quantitative CT features, will be shared. All data will be fully de-identified in accordance with applicable regulations before sharing. Imaging data (CT scans) will also be provided in de-identified format when permitted by participating sites.
Publications
Ding XF, Li JB, Liang HY, Wang ZY, Jiao TT, Liu Z, Yi L, Bian WS, Wang SP, Zhu X, Sun TW. Predictive model for acute respiratory distress syndrome events in ICU patients in China using machine learning algorithms: a secondary analysis of a cohort study. J Transl Med. 2019 Oct 1;17(1):326. doi: 10.1186/s12967-019-2075-0.
PubMed 31570096Zhang Z. Prediction model for patients with acute respiratory distress syndrome: use of a genetic algorithm to develop a neural network model. PeerJ. 2019 Sep 16;7:e7719. doi: 10.7717/peerj.7719. eCollection 2019.
PubMed 31576250Zeiberg D, Prahlad T, Nallamothu BK, Iwashyna TJ, Wiens J, Sjoding MW. Machine learning for patient risk stratification for acute respiratory distress syndrome. PLoS One. 2019 Mar 28;14(3):e0214465. doi: 10.1371/journal.pone.0214465. eCollection 2019.
PubMed 30921400Zhou Y, Feng J, Mei S, Tang R, Xing S, Qin S, Zhang Z, Xu Q, Gao Y, He Z. A deep learning model for predicting COVID-19 ARDS in critically ill patients. Front Med (Lausanne). 2023 Jul 25;10:1221711. doi: 10.3389/fmed.2023.1221711. eCollection 2023.
PubMed 37564041Chiumello D, Coppola S, Catozzi G, Danzo F, Santus P, Radovanovic D. Lung Imaging and Artificial Intelligence in ARDS. J Clin Med. 2024 Jan 5;13(2):305. doi: 10.3390/jcm13020305.
PubMed 38256439Albahri OS, Zaidan AA, Albahri AS, Zaidan BB, Abdulkareem KH, Al-Qaysi ZT, Alamoodi AH, Aleesa AM, Chyad MA, Alesa RM, Kem LC, Lakulu MM, Ibrahim AB, Rashid NA. Systematic review of artificial intelligence techniques in the detection and classification of COVID-19 medical images in terms of evaluation and benchmarking: Taxonomy analysis, challenges, future solutions and methodological aspects. J Infect Public Health. 2020 Oct;13(10):1381-1396. doi: 10.1016/j.jiph.2020.06.028. Epub 2020 Jul 1.
PubMed 32646771Amisha, Malik P, Pathania M, Rathaur VK. Overview of artificial intelligence in medicine. J Family Med Prim Care. 2019 Jul;8(7):2328-2331. doi: 10.4103/jfmpc.jfmpc_440_19.
PubMed 31463251Gutierrez G. Artificial Intelligence in the Intensive Care Unit. Crit Care. 2020 Mar 24;24(1):101. doi: 10.1186/s13054-020-2785-y.
PubMed 32204716Ahuja AS. The impact of artificial intelligence in medicine on the future role of the physician. PeerJ. 2019 Oct 4;7:e7702. doi: 10.7717/peerj.7702. eCollection 2019.
PubMed 31592346Shenoy S, Rajan AK, Rashid M, Chandran VP, Poojari PG, Kunhikatta V, Acharya D, Nair S, Varma M, Thunga G. Artificial intelligence in differentiating tropical infections: A step ahead. PLoS Negl Trop Dis. 2022 Jun 30;16(6):e0010455. doi: 10.1371/journal.pntd.0010455. eCollection 2022 Jun.
PubMed 35771774Chiumello D, Marino A, Brioni M, Menga F, Cigada I, Lazzerini M, Andrisani MC, Biondetti P, Cesana B, Gattinoni L. Visual anatomical lung CT scan assessment of lung recruitability. Intensive Care Med. 2013 Jan;39(1):66-73. doi: 10.1007/s00134-012-2707-9. Epub 2012 Sep 19.
PubMed 22990871Raghavendran K, Davidson BA, Woytash JA, Helinski JD, Marschke CJ, Manderscheid PA, Notter RH, Knight PR. The evolution of isolated bilateral lung contusion from blunt chest trauma in rats: cellular and cytokine responses. Shock. 2005 Aug;24(2):132-8. doi: 10.1097/01.shk.0000169725.80068.4a.
PubMed 16044083Gattinoni L, Caironi P, Pelosi P, Goodman LR. What has computed tomography taught us about the acute respiratory distress syndrome? Am J Respir Crit Care Med. 2001 Nov 1;164(9):1701-11. doi: 10.1164/ajrccm.164.9.2103121. No abstract available.
PubMed 11719313Gattinoni L, Pesenti A. The concept of "baby lung". Intensive Care Med. 2005 Jun;31(6):776-84. doi: 10.1007/s00134-005-2627-z. Epub 2005 Apr 6.
PubMed 15812622Xirouchaki N, Magkanas E, Vaporidi K, Kondili E, Plataki M, Patrianakos A, Akoumianaki E, Georgopoulos D. Lung ultrasound in critically ill patients: comparison with bedside chest radiography. Intensive Care Med. 2011 Sep;37(9):1488-93. doi: 10.1007/s00134-011-2317-y. Epub 2011 Aug 2.
PubMed 21809107
Study at a glance
From the registry record, last updated by the sponsor on 9 Jan 2026. Recruitment status is as the sponsor reported it: confirm with the study team.
- Official title
- Exploration of Diagnosis and Treatment Strategies and Prognostic Prediction Models for Acute Respiratory Distress Syndrome Based on Radiographic Evaluations Assessed by Artificial Intelligence
- ClinicalTrials.gov ID
- NCT07328997
- Acronym / sponsor ID
- B2024-180
- Keywords
- ARDS, chest CT, AI models
- Study type
- Observational
- Recruitment status
- Completed
- Enrolment
- 400 participants (actual)
- Lead sponsor
- Shanghai Zhongshan HospitalUniversity, hospital or other
- Study start
- 31 May 2024 (actual)
- Primary completion
- 30 Nov 2025 (actual)
- Study completion
- 30 Nov 2025 (actual)
- First posted
- 9 Jan 2026 (actual)
- Results
- Not posted in the source record
Clinical summary
The research question as the sponsor registered it. This describes what the study plans to test, not what it found.
By using multi-center chest CT data, an intelligent assessment model for the severity of ARDS was constructed. Based on CT quantitative features and clinical characteristics, a prediction model for short-term critical events (such as mechanical ventilation decisions, prone position strategies, death, ECMO use, etc.) was established. The disease was staged and quantified, and a diagnosis and risk stratification model for ARDS was developed to assist in guiding the diagnosis and treatment strategies for ARDS.
Study design
- Observational model
- Case only
- Time perspective
- Retrospective
- Number of arms
- 1
| Arm | Type | Receives |
|---|---|---|
| Training group, testing group, validation group | – | Diagnostic Test: CT scanThe study adopts a stratified random sampling strategy with an 8:2 split to construct training and internal validation datasets, together with an independent external test cohort from a separate center. No randomization of clinical interventions or treatments is involved. The model will be developed and evaluated using observational data derived from real-world clinical pathways and outcomes, with the objectives of assessing performance in disease severity stratification, treatment recommendation, and mortality prediction. Model performance will be compared with established ICU severity scores and existing AI-based approaches according to a prespecified statistical analysis plan. |
Who may qualify
- Ages
- 18 to 100 years
- Sex
- All
- Healthy volunteers
- Not accepted
- Age groups
- Adult, Older adult
- Study population
- The study population consists of adult patients diagnosed with acute respiratory distress syndrome (ARDS) who were admitted to the intensive care units (ICUs) of three tertiary comprehensive hospitals in China. Eligible participants are retrospectively identified from electronic medical records between January 2020 and December 2024. All included patients meet the predefined inclusion and exclusion criteria based on the 2023 Global ARDS Definition and have available chest CT imaging and corresponding clinical data. This cohort represents a real-world ICU population with diverse etiologies of ARDS and varying degrees of disease severity.
Key inclusion criteria (3 listed)
- Meets the diagnostic criteria for ARDS
- Be admitted to the intensive care unit
- There are chest CT images
Key exclusion criteria (3 listed)
- Age less than 18 years old
- Missing medical records
- No chest CT images
Eligibility is decided by the study team after screening; this overview does not establish that a patient qualifies. Full inclusion and exclusion criteria.
Interventions
| Intervention | Type | Arms |
|---|---|---|
| CT scanCT scan | Diagnostic test | Training group, testing group, validation group |
Primary outcome
Accuracy of ARDS severity classification
Time frame: Baseline, defined as within 24 hours of index chest CT acquisition during ICU admission.
Accuracy of the artificial intelligence-based model in classifying ARDS severity (mild, moderate, or severe), using the reference clinical classification defined by the 2023 global ARDS criteria as the ground truth.
Treatment plan matching rate between model-recommended and actual clinical management.
Time frame: Baseline, defined as within 24 hours of index chest CT acquisition during ICU admission.
Concordance rate between model-recommended treatment strategies and actual clinical management decisions across five predefined intervention modalities: mechanical ventilation, high-flow nasal oxygen therapy, non-invasive ventilation, prone positioning, and neuromuscular blockade.
Accuracy of 28-day in-hospital mortality prediction.
Time frame: Up to 28 days from ICU admission, or until hospital discharge, whichever occurs first.
Accuracy of the model in predicting all-cause in-hospital mortality within 28 days, based on integrated chest CT imaging features and clinical variables.
4 secondary outcomes
Comparative performance improvement over baseline AI models.
Time frame: Baseline for severity classification and treatment plan matching; up to 28 days from ICU admission for mortality prediction
Calibration performance of 28-day mortality prediction.
Time frame: Up to 28 days from ICU admission, or until hospital discharge, whichever occurs first.
Model interpretability based on imaging and clinical feature contributions.
Time frame: Baseline for feature extraction; up to 28 days from ICU admission for outcome association analysis.
Association between treatment concordance and 28-day in-hospital mortality.
Time frame: Up to 28 days from ICU admission, or until hospital discharge, whichever occurs first.
Results and safety
The study is completed, but results are not available in the source record. That does not mean the treatment did or did not work.
Research links
- Official record on ClinicalTrials.govLast updated by the sponsor 9 Jan 2026.
- Results publication: PubMed 31570096Ding XF, Li JB, Liang HY, Wang ZY, Jiao TT, Liu Z, Yi L, Bian WS, Wang SP, Zhu X, Sun TW. Predictive model for acute respiratory distress syndrome events in ICU patients in China using machine learning algorithms: a secondary analysis of a cohort study. J Transl Med. 2019 Oct 1;17(1):326. doi: 10.1186/s12967-019-2075-0.
- Results publication: PubMed 31576250Zhang Z. Prediction model for patients with acute respiratory distress syndrome: use of a genetic algorithm to develop a neural network model. PeerJ. 2019 Sep 16;7:e7719. doi: 10.7717/peerj.7719. eCollection 2019.
- Results publication: PubMed 30921400Zeiberg D, Prahlad T, Nallamothu BK, Iwashyna TJ, Wiens J, Sjoding MW. Machine learning for patient risk stratification for acute respiratory distress syndrome. PLoS One. 2019 Mar 28;14(3):e0214465. doi: 10.1371/journal.pone.0214465. eCollection 2019.
- Results publication: PubMed 37564041Zhou Y, Feng J, Mei S, Tang R, Xing S, Qin S, Zhang Z, Xu Q, Gao Y, He Z. A deep learning model for predicting COVID-19 ARDS in critically ill patients. Front Med (Lausanne). 2023 Jul 25;10:1221711. doi: 10.3389/fmed.2023.1221711. eCollection 2023.
- Results publication: PubMed 38256439Chiumello D, Coppola S, Catozzi G, Danzo F, Santus P, Radovanovic D. Lung Imaging and Artificial Intelligence in ARDS. J Clin Med. 2024 Jan 5;13(2):305. doi: 10.3390/jcm13020305.
- Results publication: PubMed 32646771Albahri OS, Zaidan AA, Albahri AS, Zaidan BB, Abdulkareem KH, Al-Qaysi ZT, Alamoodi AH, Aleesa AM, Chyad MA, Alesa RM, Kem LC, Lakulu MM, Ibrahim AB, Rashid NA. Systematic review of artificial intelligence techniques in the detection and classification of COVID-19 medical images in terms of evaluation and benchmarking: Taxonomy analysis, challenges, future solutions and methodological aspects. J Infect Public Health. 2020 Oct;13(10):1381-1396. doi: 10.1016/j.jiph.2020.06.028. Epub 2020 Jul 1.
- Results publication: PubMed 31463251Amisha, Malik P, Pathania M, Rathaur VK. Overview of artificial intelligence in medicine. J Family Med Prim Care. 2019 Jul;8(7):2328-2331. doi: 10.4103/jfmpc.jfmpc_440_19.
- Results publication: PubMed 32204716Gutierrez G. Artificial Intelligence in the Intensive Care Unit. Crit Care. 2020 Mar 24;24(1):101. doi: 10.1186/s13054-020-2785-y.
- Results publication: PubMed 31592346Ahuja AS. The impact of artificial intelligence in medicine on the future role of the physician. PeerJ. 2019 Oct 4;7:e7702. doi: 10.7717/peerj.7702. eCollection 2019.
- Results publication: PubMed 35771774Shenoy S, Rajan AK, Rashid M, Chandran VP, Poojari PG, Kunhikatta V, Acharya D, Nair S, Varma M, Thunga G. Artificial intelligence in differentiating tropical infections: A step ahead. PLoS Negl Trop Dis. 2022 Jun 30;16(6):e0010455. doi: 10.1371/journal.pntd.0010455. eCollection 2022 Jun.
- Results publication: PubMed 22990871Chiumello D, Marino A, Brioni M, Menga F, Cigada I, Lazzerini M, Andrisani MC, Biondetti P, Cesana B, Gattinoni L. Visual anatomical lung CT scan assessment of lung recruitability. Intensive Care Med. 2013 Jan;39(1):66-73. doi: 10.1007/s00134-012-2707-9. Epub 2012 Sep 19.
- Results publication: PubMed 16044083Raghavendran K, Davidson BA, Woytash JA, Helinski JD, Marschke CJ, Manderscheid PA, Notter RH, Knight PR. The evolution of isolated bilateral lung contusion from blunt chest trauma in rats: cellular and cytokine responses. Shock. 2005 Aug;24(2):132-8. doi: 10.1097/01.shk.0000169725.80068.4a.
- Individual participant dataSharing planned: Yes. Individual participant data that underlie the results of the study, including de-identified demographic information, clinical variables, laboratory findings, ventilator or HFNC parameters, and AI-derived quantitative CT features, will be shared. All data will be fully de-identified in accordance with applicable regulations before sharing. Imaging data (CT scans) will also be provided in de-identified format when permitted by participating sites.
Related trials for ARDS (Acute Respiratory Distress Syndrome)
Enrolling by invitation
Personalized Ventilator Settings for Patients on ECMO
Acute Respiratory Distress Syndrome (ARDS), Extracorporeal Membrane Oxygenation
Not yet recruitingObservational study
Development and Multicenter Validation of an Explainable System for Neonatal ARDS-Compatible Phenotyping
Acute Respiratory Distress Syndrome
RecruitingObservational study
Assessment of Blood Viscosity in ARDS (Acute Respiratory Distress Syndrome)
ARDS
Recruiting
Comparing Surfactant Administration Through Supraglottic Airway and Thin Catheter for Preterm Infants
Surfactant, Respiratory Distress Syndrome (Neonatal)
Recruiting
REspiratory MEchanics for Delivering Individualised Exogenous Surfactant
Respiratory Distress Syndrome in Premature Infant
Referral and contact
AzTrials does not enrol participants or take referrals. Contact the study team or a site listed below; they decide who can take part. This study is not recruiting at the moment.
The record lists no contact. Use the official record to reach the sponsor.
1 site in 1 country: see every site and its status.
Questions about this trial
What is this clinical trial studying?
An observational study of ARDS (Acute Respiratory Distress Syndrome) and AI (Artificial Intelligence), run by Shanghai Zhongshan Hospital at 1 site in China.
The full registered description is in the About this study section and the official record.
Who may be eligible to take part?
The sponsor lists 3 inclusion and 3 exclusion criteria. Eligibility is decided by the study team after screening, so read the full criteria and contact a listed site.
Is this clinical trial currently recruiting?
The record says: finished, as of the sponsor's last update on 9 Jan 2026. Status can change between updates, so confirm with the study contact before making plans.
Where is the study taking place?
The record lists 1 site in China. Whether a site is taking part right now should be confirmed with that site. See every site.
What are the possible benefits and risks?
They depend on the study and the treatment, and no study can promise a benefit. The study team explains the known risks during informed consent, before anyone joins; the record itself does not list them.
Are results available for this study?
The study is completed, but results are not available in the source record. That does not mean the treatment did or did not work.
How can a doctor refer a patient or contact the research team?
The record lists no contact. The official record names the sponsor, who can be asked directly. AzTrials does not enrol participants or take referrals.
What does the primary outcome measure?
Accuracy of ARDS severity classification (time frame: Baseline, defined as within 24 hours of index chest CT acquisition during ICU admission.). The study lists 3 primary outcomes in all. It is what the study was designed to measure, not a finding.
Where can I read the original study information?
On the ClinicalTrials.gov record for NCT07328997, which also lists 30 linked publications.
Does a listing mean the treatment is approved or proven to work?
No. A listing means the study is registered. It does not show that the treatment is safe, works, or is approved by a regulator.
How often is this information updated?
AzTrials copies changed records from ClinicalTrials.gov every night and reloads the whole registry every week. The sponsor last updated this record on 9 Jan 2026.