Skip to content

Research on Early Warning and Solution System of Difficult Airway in Perioperative Period Based on Artificial Intelligence

Research on Early Warning and Solution System of Difficult Airway in Perioperative Period Based on Artificial Intelligence

Status
UNKNOWN
Phases
Unknown
Study type
Observational
Source
ClinicalTrials.gov
Registry ID
NCT04458220
Enrollment
16000
Registered
2020-07-07
Start date
2020-07-30
Completion date
2025-05-30
Last updated
2023-08-01

For informational purposes only — not medical advice. Sourced from public registries and may not reflect the latest updates. Terms

Conditions

Difficult Airway

Brief summary

The study is performed in Shanghai Ninth People's Hospital Affiliated with Shanghai Jiao Tong University School of Medicine . According to inclusion and exclusion criteria ,the investigators are expected to enroll 16000 participants who performed endotracheal intubation under general anesthesia. All enrolled participants must sign a written informed consent.In the preset studio,The 3D face scanner and camera is used to obtain 3D or 2-dimensional portrait images of patients in different positions and from different angles.The Hi-Fi Recorder is used to obtain sound samples of patients in different word.Then,Statistical experts use quantitative software to quantify the data.The investigators will put all the data and images into a confidential database in order to build a large database of difficult airways. Anesthesiologist will give every patient an endotracheal intubation as planned. After that the anesthesiologist will be asked to fill out the questionnaire immediately.This questionnaire allows obtaining a ground truth for the intubation difficulty.All data will be used for AI deep learning and intelligent analysis,Several of the most relevant landmarks will be selected to build an early warning model.The overall study did not involve any intervention in the process of anaesthesia and operation of the patients, only the three-dimensional facial images of the patients were obtained, without any trauma or injury.

Interventions

OTHERphoto and voice obtained

The 3D face scanner and camera is used to obtain 3D or 2-dimensional portrait images of patients in different positions and from different angles.The Hi-Fi Recorder is used to obtain sound samples of patients in different word.

Sponsors

Shanghai Ninth People's Hospital Affiliated to Shanghai Jiao Tong University
Lead SponsorOTHER

Study design

Observational model
CASE_CONTROL
Time perspective
CROSS_SECTIONAL

Eligibility

Sex/Gender
ALL
Healthy volunteers
Yes

Inclusion criteria

* Patients to be intubated under general anesthesia.

Exclusion criteria

* Patients with deaf-mutism or communication disorders. * Patients with language deficiency or non-native language. * Patients with mental or central nervous system disease. * Patients with stupefaction or disturbance of consciousness. * Patients with terrible injury. * Patients cannot follow instructions to perform standard actions. * Patients participated in other relevant clinical investigation in the past 3 months.

Design outcomes

Primary

MeasureTime frameDescription
Diagnosed as a difficult airwayJust after intubationC-L≥Ⅲ grade

Secondary

MeasureTime frameDescription
Difficult mask ventilation2 minutes after administration of RocuroniumThe difficult mask ventilation was defined as follows: (1) the inability of an unassisted anesthesiologist to maintain oxygen saturation, as measured by SpO2\<92% with 100% oxygen and positive-pressure mask ventilation; (2) significant gas flow leakage around the face mask; (3) the need to increase the gas flow to more than 15 L/min and use the oxygen flush valve more than twice (4) absence of visible chest movement; (5) the necessity to switch to a two-handed mask ventilation technique; (6) the need for operator substitution or addition.

Countries

China

Contacts

Primary ContactRen Zhou, Dr
zhouren77@126.com86-15121007303
Backup ContactMing Xia, Dr
jianghongjiuyuan@163.com86-15021306970

Outcome results

None listed

Source: ClinicalTrials.gov · Data processed: Feb 4, 2026