Establishment of Airway Database for Surgical Patients

Artificial Intelligence Clinical Trial

Official title:

Study on the Method of Difficult Airway Prediction Based on Artificial Intelligence

Clinical Trial Details — Status: Not yet recruiting

Administrative data

• NCT number: NCT03125837
• Other study ID #: 2016-076
• Status: Not yet recruiting
• Phase: N/A
• First received: March 13, 2017
• Last updated: April 19, 2017
• Start date: May 2017
• Est. completion date: May 2022

Study information

• Verified date: April 2017
• Source: Second Affiliated Hospital, School of Medicine, Zhejiang University
• Contact: n/a
• Is FDA regulated: No
• Study type: Observational

Clinical Trial Summary

Difficult airway is a major reason of anesthesia related injuries with latent life threatening complications. Foresee difficult airway in the preoperative period is vital for the patient's safety. The aim of this study is to develop a computer algorithm that can detect whether the patient is a difficult airway based on photographs form six aspects. This method will be decreased potential complication related to difficult airway and increased patient safety.

Description:

Introduction:

The primary purpose of the study is to develop a computer algorithm that can detect whether the patient is a difficult airway based on photographs from six different aspects.

Methods:

This study is divided into two parts. In the first part, we collected the patients' airway assessment score who underwent general anesthesia with endotracheal intubation assessed by an experienced attending anesthesiologists before and after intubation. Evaluation of airway score after tracheal intubation as the gold standard for airway assessment. Digital photographs of the face of each patient in frontal neutral view and in profile neutrals were obtained. Details of the photographs, each corresponding to a facial motion: (1) Frontal, neutral. (2) Frontal, mouth open. (3)Frontal, extreme mouth open and tongue out. (4)Frontal, extreme upper lip bite (5)Profile, neutral. (6) Profile, neutral, maximum head back. The patient's photographs and the airway evaluation score after intubation were input to the computer to train the computer. In the second part, the trained computer was used to evaluate the airway score of the new patient compared with that of the patient after intubation, and calculated the sensitivity.

Recruitment information / eligibility

• Status: Not yet recruiting
• Enrollment: 50000
• Est. completion date: May 2022
• Est. primary completion date: May 2022
• Accepts healthy volunteers: No
• Gender: All
• Age group: 18 Years and older

Eligibility

Inclusion Criteria:

• General anesthesia-induced tracheal intubation in patients who undergoing elective surgical patients

Exclusion Criteria:

• Patients with multiple facial injuries Patients who had undergone head or neck surgery Patients who need emergency operation

Related Conditions & MeSH terms

• Artificial Intelligence
• Difficult Airway

Locations

Country	Name	City	State
n/a

Sponsors (2)

Lead Sponsor	Collaborator
Second Affiliated Hospital, School of Medicine, Zhejiang University	Zhejiang University

Outcome

Type	Measure	Description	Time frame	Safety issue
Primary	the sensitivity of artificial Intelligence to predict difficulty of facemask ventilation and endotracheal intubation	The outcome will be a computer algorithm that can detect whether the patient is a difficult airway based on photographs from six different aspects.Details of the photographs, each corresponding to a facial motion: (1) Frontal, neutral. (2) Frontal, mouth open. (3)Frontal, extreme mouth open and tongue out. (4)Frontal, extreme upper lip bite (5)Profile, neutral. (6) Profile, neutral, maximum head back.	5 years