Development of an Algorithm to Denoise HFNO-generated Tracheal Sound

National Taiwan University

Status: Enrolling

Conditions: Endoscopy, Gastrointestinal; Sedation, Conscious

Treatments: Device: High flow nasal oxygen fist; Device: High flow nasal oxygen later

Study type: Interventional

Funder types: Other

Identifiers

NCT06218017

202310040DINC

Details and patient eligibility

About

This study aims to develop an algorithm to mitigate the noise generated by the high-flow nasal oxygen system for tracheal sound monitoring in deeply sedated patients.

Full description

During deep sedation without intubation, anesthesia respiratory care and monitoring are of utmost importance. Anesthesia can lead to adverse respiratory effects, increasing the risk of airway obstruction and respiratory depression. High-flow nasal oxygen (HFNO) systems are often used to mitigate these risks. Monitoring tracheal breathing sounds directly using amplification provides better detection accuracy. However, traditional auscultation instruments have limitations, including noise interference and lack of visual functionality. The Airmod Smart Respiratory Monitoring System, with FDA approval in the United States and Taiwan, addresses these issues. It offers respiratory sound recording, noise filtering, event logging, and respiratory rate analysis. Currently, no digital stethoscope on the market can filter out HFNO noise. Therefore, the current study is to develop an algorithm to mitigate the noise generated by the high-flow nasal oxygen system for tracheal sound monitoring in deeply sedated patients.

Enrollment: 60 estimated patients

Sex: All

Ages: 18+ years old

Volunteers: No Healthy Volunteers

Inclusion criteria

• Adult patients undergoing elective gastroinestinal endoscoy requiring deep sedation

Exclusion criteria

• Histories of sleep apnea, chronic obstructive pulmonary disease, or oropharyngeal tumor
• Patient who has a hisotry of previous intraoral surgery, tracheal surgery or pulmonary resection surgery

Trial design

Primary purpose: Other

Allocation: Randomized

Interventional model: Crossover Assignment

Masking: Double Blind

---

60 participants in 2 patient groups

HFNO first

Experimental group

Description:

This study follows a crossover design employing a randomized controlled methodology. Patients undergoing gastrointestinal endoscopy and receiving deep sedation were monitored using electronic tracheal sound auscultation. Within this group, patients first underwent a 10-minute session of high-flow nasal oxygen (50 L/min) followed by the utilization of the standard low-flow nasal oxygen (4 L/min). Electronic tracheal sound recordings were obtained during both the high-flow and low-flow nasal oxygen administrations. The aim is to develop an algorithm capable of mitigating the noise generated specifically by the high-flow nasal oxygen.

Treatment:

Device: High flow nasal oxygen fist

HFNO later

Active Comparator group

Description:

This study follows a crossover design employing a randomized controlled methodology. Patients undergoing gastrointestinal endoscopy and receiving deep sedation were monitored using electronic tracheal sound auscultation. Within this group, patients first underwent a 10-minute session of standard low-flow nasal oxygen (4 L/min) followed by the utilization of the high-flow nasal oxygen (50 L/min). Electronic tracheal sound recordings were obtained during both the high-flow and low-flow nasal oxygen administrations. The aim is to develop an algorithm capable of mitigating the noise generated specifically by the high-flow nasal oxygen

Treatment:

Device: High flow nasal oxygen later