Pre-anesthesia Imaging-based Respiratory Assessment and Analysis

Kaohsiung Medical University

Status: Enrolling

Conditions: Clinical Decision Support System

Treatments: Procedure: intubation for general anesthesia

Study type: Observational

Funder types: Other

Identifiers

NCT06270797

KMUHIRB-E(I)-20230184

Details and patient eligibility

About

This study is to establish a preoperative respiratory imaging assessment database and develop a difficult intubation risk prediction model and further risk analysis. We attempt to construct it into a pre-anesthesia intubation risk assessment software as the clinical decision support system.

Full description

Anesthesia respiratory assessment is an important issue for anesthesiologists to evaluate the respiratory status and airway management of patients before surgery. The American Society of Anesthesiologists (ASA) updated its guidelines in 2022, emphasizing the importance of comprehensive respiratory assessment in the guidelines.

Various risk factors have been proposed in past literature for discussion, and corresponding to these risk factors, there is currently no single factor that can predict difficult intubation completely. Existing investigations into difficult intubation factors mostly focus on high-risk populations, including patients with morbid obesity, where significant differences have been identified but not developed into predictive models.

With the rapid development of AI-related technologies in recent years, numerous image-related AI frameworks have been proposed. In recent years, attempts have been made to combine various clinical risk factors using machine learning methods to create automated prediction models for difficult intubation. However, their effectiveness has not met expectations, reflecting the significant clinical problem of difficulty in prediction that remains unresolved.

This study is an observational study aimed at analyzing and establishing patient image data, refining various data engineering techniques, and optimizing existing prediction model frameworks to enhance their medical value. Additionally, the focus of this project will be on establishing more prediction models to improve existing clinical decision support systems.

Enrollment: 30,000 estimated patients

Sex: All

Ages: 18 to 85 years old

Volunteers: Accepts Healthy Volunteers

Inclusion criteria

• Patients undergoing general anesthesia
• Patients who can undergo pre-anesthetic consultation and airway examination.

Exclusion criteria

• Patients unable to undergo pre-anesthetic consultation and airway examination.
• Patients requiring emergency surgery.
• Vulnerable populations.

Trial design

30,000 participants in 2 patient groups

normal intubation

Description:

during general anesthesia, normal intubation without any difficult airway or difficult intubation were recorded in the note.

Treatment:

Procedure: intubation for general anesthesia

difficult airway or difficult intubation

Description:

during general anesthesia, any type of difficult airway or difficult intubation were recorded in the note.

Treatment:

Procedure: intubation for general anesthesia