DeliriumNCT07136207
Development and Validation of Delirium Recognition Using Computer Vision in Neuro-critical Patients
An observational study of Delirium and Artificial Intelligence (AI), run by Beijing Tiantan Hospital at 1 site in China.
Trial summary
This research project employs machine learning algorithms integrated with computer vision, image processing, and pattern recognition technologies to perform digital analysis of facial expression behaviors in neurocritical care patients with delirium. Read more about this study
- Who is it for?
- Ages 18 to 80 years
- Conditions the study is for
- Delirium, Artificial Intelligence (AI)
- How many people
- 1,000 participants
- Where
- 1 siteChina
- Study dates
- 30 Aug 2025to 30 Jan 2026
- Phase
- Observational
- Are there results?
- Not yet: the study is under way
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 80 years
- All sexes
- No healthy volunteers
- Adult, Older adult
Study population: This study selects neurocritical patients as the population and collects facial expression data from both delirium and non-delirium patients.
Who may be able to join
- Neurocritical patients admitted to the ICU, including postoperative neurosurgical patients, stroke patients, and those receiving ICU care due to other neurological conditions.
- Age over 18 years.
- Signed informed consent.
Who cannot join
- Age under 18 years.
- Persistent coma (GCS ≤ 8) within 7 days pre- and post-surgery, making delirium assessment impossible.
- Did not survive more than 24 hours in the ICU.
- Patients with facial paralysis, post-traumatic facial disfigurement, or other conditions that could significantly affect facial recognition.
- Exclusion of patients with severe dementia, Parkinson's disease, depression, or other conditions that might impact facial emotional expressions.
Trial locations and contacts
1 site in 1 country. Contact the study team or a site near you to ask about taking part.
China1 site, 1 recruiting
- Recruiting
Beijing Tiantan Hospital
Beijing, Beijing Municipality, 100000
Shi Guangzhi Department Director, Doctoral degree, +8613599058877, huanghw0403@163.com
About this study
Official title: Research on Delirium Recognition in Neurocritical Patients Based on Facial Expression Behavior Patterns
This research project employs machine learning algorithms integrated with computer vision, image processing, and pattern recognition technologies to perform digital analysis of facial expression behaviors in neurocritical care patients with delirium. By constructing multidimensional high-level features of delirium, the investigators have established a classification model based on behavioral. The primary objective of this study is to address the critical challenge of achieving precise and efficient delirium diagnosis in neurologically critically ill patients through automated facial expression behavior recognition.
Read the detailed description
This study is a prospective cohort study approved by the Ethics Committee of Beijing Tiantan Hospital. It aims to support the accurate and efficient diagnosis of delirium in neurocritical patients through a facial expression recognition system. A mobile application was developed for this study, collaboratively designed by senior clinicians and engineers from the Institute of Computing Technology, Chinese Academy of Sciences. The application is based on a stimulus paradigm designed using CAM-ICU (Confusion Assessment Method for the Intensive Care Unit) questions to record dynamic facial videos of neurocritical patients following delirium evaluation based on the DSM-V criteria.
Patients were assessed for delirium and facial expression behavior data were collected twice daily during ICU admission, in two time slots: 8:00-10:00 AM and 8:00-10:00 PM, following the study's inclusion and exclusion criteria. A trained and experienced specialist used the gold standard DSM-V to diagnose delirium. Within five minutes after completing the assessment, dynamic facial behavior video data were collected to prepare images for subsequent model development.
Various image preprocessing and data augmentation techniques were employed to prepare the images for the VGG16 model. These techniques are standard for running convolutional neural network (CNN) models. Using the "preprocess_input"function from the Keras VGGFace module, the investigators standardized image color and size to ensure that each image met the expected input requirements for model training. For data augmentation, the investigators applied TensorFlow's "ImageDataGenerator" function to perform horizontal flipping, rotation, scaling, width and height shifting, and shearing. These augmentation techniques created a more diverse dataset, helping to prevent overfitting and improving the model's generalizability to new faces.
The investigators developed a binary classification model to identify delirium using a CNN with a pretrained backbone. The VGG16 model, based on deep learning, was adopted, leveraging transfer learning from VGGFace2, which possesses pre-existing facial feature recognition capabilities. Transfer learning allowed us to utilize prior knowledge to detect features more quickly, accurately, and with lower computational cost. The VGGFace2 model was employed for training.
Model performance was evaluated through internal validation at Beijing Tiantan Hospital and external validation at Guiyang Second People's Hospital, with metrics including accuracy, sensitivity, specificity, and F1 score. Additionally, to address the "black box" issue of machine learning, occlusion heatmap techniques were used to identify the most critical facial regions for delirium assessment, with the results visualized on a virtual face.
This model aims to support precise and efficient identification of delirium in neurocritical care units.
- Conditions
- Delirium, Artificial Intelligence (AI)
- Keywords
- Delirium, Artificial Intelligence (AI), neurocritical care units, Facial recognition
- Medical subject headings
- Delirium
What participants receive
What participants receive. In a randomised study each person is assigned to one group by chance.
| Study group | Role | What they receive |
|---|---|---|
| Neurocritical non-delirium patients | For neurocritical non-delirium patients, the investigators record facial expression videos, which are used during model development to compare with the facial expressions of delirium patients. | |
| Neurocritical delirium patients | The investigators record facial expression videos of neurocritical delirium patients and perform frame sampling on the videos to analyze and extract the facial expression features specific to delirium. Based on this analysis, the investigators develop a model for delirium recognition in neurocritical patients. |
What is being measured
The primary outcome is the main question the study is designed to answer.
Accuracy of the delirium prediction model
The accuracy of the delirium prediction model will be calculated as the proportion of correct predictions among total predictions.
Measured: Through study completion, an average of 1 year
Sensitivity of the delirium prediction model
Sensitivity (true positive rate) will be assessed as the proportion of actual delirium cases correctly identified by the model.
Measured: Through study completion, an average of 1 year
Specificity of the delirium prediction model
Specificity (true negative rate) will be calculated as the proportion of non-delirium cases correctly identified by the model.
Measured: Through study completion, an average of 1 year
Secondary outcomes (2)
F1 Score of the delirium prediction model
The F1 score, the harmonic mean of precision and recall, will be used to evaluate the balance between sensitivity and specificity.
Measured: Through study completion, an average of 1 year
AUC of the facial feature curve for delirium patients
The area under the curve (AUC) of the receiver operating characteristic (ROC) curve derived from facial features will be used to assess the discriminatory performance of the model.
Measured: Through study completion, an average of 1 year
Timeline
From the start of the study to its completion; the mark is today. Estimated dates are the sponsor’s plan.
- First posted22 Aug 2025
- Study start30 Aug 2025 (estimated)
- Primary completion30 Dec 2025 (estimated)
- Study completion30 Jan 2026 (estimated)
Resources
- Official record on ClinicalTrials.govThe registry entry this page is built from, with its full change history.
- Publications (10)Papers the sponsor has linked to this study.
- Other trials for DeliriumStudies of the same condition, open and under way.
- 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
- Cohort
- Time perspective
- Prospective
- Enrolment
- 1,000 participants (estimated)
Sponsor and investigators
- Lead sponsor
- Beijing Tiantan Hospital
- Sponsor type
- University, hospital or other
- Collaborators
- Affiliated Jinyang Hospital of Guizhou Medical University, The Affiliated Hospital Of Guizhou Medical University
- Responsible party
- Sponsor
IDs, dates and publications
- ClinicalTrials.gov ID
- NCT07136207
- Sponsor's protocol ID
- PX2023021
- Status last verified
- Aug 2025
- First submitted
- 18 Jul 2025
- Last updated
- 22 Aug 2025
- Results posted
- No
- Data monitoring committee
- No
- FDA-regulated drug
- No
- FDA-regulated device
- No
- Individual data sharing
- No: This study involves collecting facial information of patients, which pertains to their privacy. To protect participants' confidentiality, all data will be uniformly destroyed after the study is completed. The investigators will not share or disclose patients' information to other researchers.
Publications
Heintz TA, Badathala A, Wooten A, Cu CW, Wallace A, Pham B, Wallace AW, Cobert J. Preliminary Development and Validation of Automated Nociception Recognition Using Computer Vision in Perioperative Patients. Anesthesiology. 2025 Apr 1;142(4):726-737. doi: 10.1097/ALN.0000000000005370. Epub 2025 Jan 13.
PubMed 39804295Atee M, Hoti K, Parsons R, Hughes JD. A novel pain assessment tool incorporating automated facial analysis: interrater reliability in advanced dementia. Clin Interv Aging. 2018 Jul 16;13:1245-1258. doi: 10.2147/CIA.S168024. eCollection 2018.
PubMed 30038491Goldberg TE, Chen C, Wang Y, Jung E, Swanson A, Ing C, Garcia PS, Whittington RA, Moitra V. Association of Delirium With Long-term Cognitive Decline: A Meta-analysis. JAMA Neurol. 2020 Nov 1;77(11):1373-1381. doi: 10.1001/jamaneurol.2020.2273.
PubMed 32658246Aldecoa C, Bettelli G, Bilotta F, Sanders RD, Audisio R, Borozdina A, Cherubini A, Jones C, Kehlet H, MacLullich A, Radtke F, Riese F, Slooter AJ, Veyckemans F, Kramer S, Neuner B, Weiss B, Spies CD. European Society of Anaesthesiology evidence-based and consensus-based guideline on postoperative delirium. Eur J Anaesthesiol. 2017 Apr;34(4):192-214. doi: 10.1097/EJA.0000000000000594.
PubMed 28187050Ely EW, Margolin R, Francis J, May L, Truman B, Dittus R, Speroff T, Gautam S, Bernard GR, Inouye SK. Evaluation of delirium in critically ill patients: validation of the Confusion Assessment Method for the Intensive Care Unit (CAM-ICU). Crit Care Med. 2001 Jul;29(7):1370-9. doi: 10.1097/00003246-200107000-00012.
PubMed 11445689Ahmed A, Garcia-Agundez A, Petrovic I, Radaei F, Fife J, Zhou J, Karas H, Moody S, Drake J, Jones RN, Eickhoff C, Reznik ME. Delirium detection using wearable sensors and machine learning in patients with intracerebral hemorrhage. Front Neurol. 2023 Jun 9;14:1135472. doi: 10.3389/fneur.2023.1135472. eCollection 2023.
PubMed 37360342Al-Hindawi A, Vizcaychipi M, Demiris Y. A Dual-Camera Eye-Tracking Platform for Rapid Real-Time Diagnosis of Acute Delirium: A Pilot Study. IEEE J Transl Eng Health Med. 2024 May 7;12:488-498. doi: 10.1109/JTEHM.2024.3397737. eCollection 2024.
PubMed 39050621Oh J, Cho D, Park J, Na SH, Kim J, Heo J, Shin CS, Kim JJ, Park JY, Lee B. Prediction and early detection of delirium in the intensive care unit by using heart rate variability and machine learning. Physiol Meas. 2018 Mar 27;39(3):035004. doi: 10.1088/1361-6579/aaab07.
PubMed 29376502Eeles E, Tronstad O, Teodorczuk A, Flaws D, Fraser JF, Dissanayaka N. Face and content validity of a mobile delirium screening tool adapted for use in the medical setting (eDIS-MED): Welcome to the machine. Australas J Ageing. 2024 Jun;43(2):415-419. doi: 10.1111/ajag.13288. Epub 2024 Feb 28.
PubMed 38415380Nejati V, Khorrami AS, Fonoudi M. Neuromodulation of facial emotion recognition in health and disease: A systematic review. Neurophysiol Clin. 2022 Jun;52(3):183-201. doi: 10.1016/j.neucli.2022.03.005. Epub 2022 Apr 12.
PubMed 35428551
Study at a glance
From the registry record, last updated by the sponsor on 22 Aug 2025. Recruitment status is as the sponsor reported it: confirm with the study team.
- Official title
- Research on Delirium Recognition in Neurocritical Patients Based on Facial Expression Behavior Patterns
- ClinicalTrials.gov ID
- NCT07136207
- Acronym / sponsor ID
- PX2023021
- Conditions
- Delirium, Artificial Intelligence (AI)
- Keywords
- Delirium, Artificial Intelligence (AI), neurocritical care units, Facial recognition
- Study type
- Observational
- Recruitment status
- Recruiting
- Enrolment
- 1,000 participants (estimated)
- Lead sponsor
- Beijing Tiantan HospitalUniversity, hospital or other
- Collaborators
- Affiliated Jinyang Hospital of Guizhou Medical University, The Affiliated Hospital Of Guizhou Medical University
- Study start
- 30 Aug 2025 (estimated)
- Primary completion
- 30 Dec 2025 (estimated)
- Study completion
- 30 Jan 2026 (estimated)
- First posted
- 22 Aug 2025 (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.
This research project employs machine learning algorithms integrated with computer vision, image processing, and pattern recognition technologies to perform digital analysis of facial expression behaviors in neurocritical care patients with delirium. By constructing multidimensional high-level features of delirium, the investigators have established a classification model based on behavioral. The primary objective of this study is to address the critical challenge of achieving precise and efficient delirium diagnosis in neurologically critically ill patients through automated facial expression behavior recognition.
Detailed description
This study is a prospective cohort study approved by the Ethics Committee of Beijing Tiantan Hospital. It aims to support the accurate and efficient diagnosis of delirium in neurocritical patients through a facial expression recognition system. A mobile application was developed for this study, collaboratively designed by senior clinicians and engineers from the Institute of Computing Technology, Chinese Academy of Sciences. The application is based on a stimulus paradigm designed using CAM-ICU (Confusion Assessment Method for the Intensive Care Unit) questions to record dynamic facial videos of neurocritical patients following delirium evaluation based on the DSM-V criteria.
Patients were assessed for delirium and facial expression behavior data were collected twice daily during ICU admission, in two time slots: 8:00-10:00 AM and 8:00-10:00 PM, following the study's inclusion and exclusion criteria. A trained and experienced specialist used the gold standard DSM-V to diagnose delirium. Within five minutes after completing the assessment, dynamic facial behavior video data were collected to prepare images for subsequent model development.
Various image preprocessing and data augmentation techniques were employed to prepare the images for the VGG16 model. These techniques are standard for running convolutional neural network (CNN) models. Using the "preprocess_input"function from the Keras VGGFace module, the investigators standardized image color and size to ensure that each image met the expected input requirements for model training. For data augmentation, the investigators applied TensorFlow's "ImageDataGenerator" function to perform horizontal flipping, rotation, scaling, width and height shifting, and shearing. These augmentation techniques created a more diverse dataset, helping to prevent overfitting and improving the model's generalizability to new faces.
The investigators developed a binary classification model to identify delirium using a CNN with a pretrained backbone. The VGG16 model, based on deep learning, was adopted, leveraging transfer learning from VGGFace2, which possesses pre-existing facial feature recognition capabilities. Transfer learning allowed us to utilize prior knowledge to detect features more quickly, accurately, and with lower computational cost. The VGGFace2 model was employed for training.
Model performance was evaluated through internal validation at Beijing Tiantan Hospital and external validation at Guiyang Second People's Hospital, with metrics including accuracy, sensitivity, specificity, and F1 score. Additionally, to address the "black box" issue of machine learning, occlusion heatmap techniques were used to identify the most critical facial regions for delirium assessment, with the results visualized on a virtual face.
This model aims to support precise and efficient identification of delirium in neurocritical care units.
Study design
- Observational model
- Cohort
- Time perspective
- Prospective
- Number of arms
- 2
| Arm | Type | Receives |
|---|---|---|
| Neurocritical non-delirium patients | – | –For neurocritical non-delirium patients, the investigators record facial expression videos, which are used during model development to compare with the facial expressions of delirium patients. |
| Neurocritical delirium patients | – | –The investigators record facial expression videos of neurocritical delirium patients and perform frame sampling on the videos to analyze and extract the facial expression features specific to delirium. Based on this analysis, the investigators develop a model for delirium recognition in neurocritical patients. |
Who may qualify
- Ages
- 18 to 80 years
- Sex
- All
- Healthy volunteers
- Not accepted
- Age groups
- Adult, Older adult
- Study population
- This study selects neurocritical patients as the population and collects facial expression data from both delirium and non-delirium patients.
Key inclusion criteria (3 listed)
- Neurocritical patients admitted to the ICU, including postoperative neurosurgical patients, stroke patients, and those receiving ICU care due to other neurological conditions.
- Age over 18 years.
- Signed informed consent.
Key exclusion criteria (5 listed)
- Age under 18 years.
- Persistent coma (GCS ≤ 8) within 7 days pre- and post-surgery, making delirium assessment impossible.
- Did not survive more than 24 hours in the ICU.
- Patients with facial paralysis, post-traumatic facial disfigurement, or other conditions that could significantly affect facial recognition.
- Exclusion of patients with severe dementia, Parkinson's disease, depression, or other conditions that might impact facial emotional expressions.
Eligibility is decided by the study team after screening; this overview does not establish that a patient qualifies. Full inclusion and exclusion criteria.
Primary outcome
Accuracy of the delirium prediction model
Time frame: Through study completion, an average of 1 year
The accuracy of the delirium prediction model will be calculated as the proportion of correct predictions among total predictions.
Sensitivity of the delirium prediction model
Time frame: Through study completion, an average of 1 year
Sensitivity (true positive rate) will be assessed as the proportion of actual delirium cases correctly identified by the model.
Specificity of the delirium prediction model
Time frame: Through study completion, an average of 1 year
Specificity (true negative rate) will be calculated as the proportion of non-delirium cases correctly identified by the model.
2 secondary outcomes
F1 Score of the delirium prediction model
Time frame: Through study completion, an average of 1 year
AUC of the facial feature curve for delirium patients
Time frame: Through study completion, an average of 1 year
Results and safety
This study is still under way. Results are not available in the source record yet.
Research links
- Official record on ClinicalTrials.govLast updated by the sponsor 22 Aug 2025.
- Background: PubMed 39804295Heintz TA, Badathala A, Wooten A, Cu CW, Wallace A, Pham B, Wallace AW, Cobert J. Preliminary Development and Validation of Automated Nociception Recognition Using Computer Vision in Perioperative Patients. Anesthesiology. 2025 Apr 1;142(4):726-737. doi: 10.1097/ALN.0000000000005370. Epub 2025 Jan 13.
- Background: PubMed 30038491Atee M, Hoti K, Parsons R, Hughes JD. A novel pain assessment tool incorporating automated facial analysis: interrater reliability in advanced dementia. Clin Interv Aging. 2018 Jul 16;13:1245-1258. doi: 10.2147/CIA.S168024. eCollection 2018.
- Background: PubMed 32658246Goldberg TE, Chen C, Wang Y, Jung E, Swanson A, Ing C, Garcia PS, Whittington RA, Moitra V. Association of Delirium With Long-term Cognitive Decline: A Meta-analysis. JAMA Neurol. 2020 Nov 1;77(11):1373-1381. doi: 10.1001/jamaneurol.2020.2273.
- Background: PubMed 28187050Aldecoa C, Bettelli G, Bilotta F, Sanders RD, Audisio R, Borozdina A, Cherubini A, Jones C, Kehlet H, MacLullich A, Radtke F, Riese F, Slooter AJ, Veyckemans F, Kramer S, Neuner B, Weiss B, Spies CD. European Society of Anaesthesiology evidence-based and consensus-based guideline on postoperative delirium. Eur J Anaesthesiol. 2017 Apr;34(4):192-214. doi: 10.1097/EJA.0000000000000594.
- Background: PubMed 11445689Ely EW, Margolin R, Francis J, May L, Truman B, Dittus R, Speroff T, Gautam S, Bernard GR, Inouye SK. Evaluation of delirium in critically ill patients: validation of the Confusion Assessment Method for the Intensive Care Unit (CAM-ICU). Crit Care Med. 2001 Jul;29(7):1370-9. doi: 10.1097/00003246-200107000-00012.
- Background: PubMed 37360342Ahmed A, Garcia-Agundez A, Petrovic I, Radaei F, Fife J, Zhou J, Karas H, Moody S, Drake J, Jones RN, Eickhoff C, Reznik ME. Delirium detection using wearable sensors and machine learning in patients with intracerebral hemorrhage. Front Neurol. 2023 Jun 9;14:1135472. doi: 10.3389/fneur.2023.1135472. eCollection 2023.
- Background: PubMed 39050621Al-Hindawi A, Vizcaychipi M, Demiris Y. A Dual-Camera Eye-Tracking Platform for Rapid Real-Time Diagnosis of Acute Delirium: A Pilot Study. IEEE J Transl Eng Health Med. 2024 May 7;12:488-498. doi: 10.1109/JTEHM.2024.3397737. eCollection 2024.
- Background: PubMed 29376502Oh J, Cho D, Park J, Na SH, Kim J, Heo J, Shin CS, Kim JJ, Park JY, Lee B. Prediction and early detection of delirium in the intensive care unit by using heart rate variability and machine learning. Physiol Meas. 2018 Mar 27;39(3):035004. doi: 10.1088/1361-6579/aaab07.
- Background: PubMed 38415380Eeles E, Tronstad O, Teodorczuk A, Flaws D, Fraser JF, Dissanayaka N. Face and content validity of a mobile delirium screening tool adapted for use in the medical setting (eDIS-MED): Welcome to the machine. Australas J Ageing. 2024 Jun;43(2):415-419. doi: 10.1111/ajag.13288. Epub 2024 Feb 28.
- Background: PubMed 35428551Nejati V, Khorrami AS, Fonoudi M. Neuromodulation of facial emotion recognition in health and disease: A systematic review. Neurophysiol Clin. 2022 Jun;52(3):183-201. doi: 10.1016/j.neucli.2022.03.005. Epub 2022 Apr 12.
- Individual participant dataSharing planned: No. This study involves collecting facial information of patients, which pertains to their privacy. To protect participants' confidentiality, all data will be uniformly destroyed after the study is completed. The investigators will not share or disclose patients' information to other researchers.
Related trials for Delirium
Recruiting
Postoperative Delirium and Analgesia in Elderly Patients
Postoperative Delirium, Postoperative Pain, Acute
Not yet recruitingObservational study
Sugammadex Timing and Recovery Profile in Pediatric Adenotonsillectomy
Neuromuscular Blockade, Delayed Emergence From Anesthesia
Not yet recruitingObservational study
Biomarkers of Postoperative Complications After Cardiac Surgery
Postoperative Complications
Not yet recruitingPhase 3
Lactium Premedication to Prevent Emergence Agitation
Lactium, Premedication
Not yet recruiting
Creatine and Perioperative Brain Health
Neurocognitive Disorders, Postoperative Delirium
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.
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 Delirium and Artificial Intelligence (AI), run by Beijing Tiantan 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 5 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: looking for participants now, as of the sponsor's last update on 22 Aug 2025. 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?
This study is still under way. Results are not available in the source record yet.
How can a doctor refer a patient or contact the research team?
Contact Huang Huawei, Doctoral degree (+8613599058877 ext. 59978000, huanghw0403@163.com), or a site near the patient. AzTrials does not enrol participants or take referrals.
What does the primary outcome measure?
Accuracy of the delirium prediction model (time frame: Through study completion, an average of 1 year). 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 NCT07136207, which also lists 10 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 22 Aug 2025.