A Deep Learning Approach to Identify Patients With Full Stomach on Ultrasonography
- Conditions
- Gastic Antrum
- Interventions
- Dietary Supplement: Intervention oral supplement
- Registration Number
- NCT05846607
- Lead Sponsor
- Huashan Hospital
- Brief Summary
Preoperative gastric ultrasonography is a newly developed tool used to evaluate gastric content and volume in assessing perioperative aspiration risk and guide anaesthetic management. And then build up effective clinical predictive models for identification of full stomach, which can predict the high aspiration risk.
- Detailed Description
Aspiration of gastric contents can be a serious anesthetic related complication. Preoperative fasting was a common practice to decrease perioperative aspiration risk. However，one of most important prescription of enhanced recovery after surgery protocols is the reduction of preoperative fasting time in opposition to the traditional recommendation of overnight fast. Gastric antral sonography prior to anesthesia may have a role in identifying patients at risk of aspiration. The aim of this study is to construct models using deep learning for identification of full stomach, which can predict the aspiration risk.
Recruitment & Eligibility
- Status
- RECRUITING
- Sex
- All
- Target Recruitment
- 800
Patients undergoing surgery Age: 18~85 yeas ASA 1~3
-Diabetes mellitus Upper gastrointestinal pathology such as hiatus hernia, oesophageal cancer Prior surgery to upper GI On medication that may affect gastric emptying time Pregnancy
Study & Design
- Study Type
- INTERVENTIONAL
- Study Design
- SINGLE_GROUP
- Arm && Interventions
Group Intervention Description experiment Intervention oral supplement The data will be collected in two phases. Firstly, after a fasting period of eight hours, the participants received ultrasound scan by expert doctor with the objective of measuring the fasting gastric antrum area in supine and right lateral decubitus(RLD)position(phase 1).Then the expert made the grading according to the ultrasound image. In phase 2, the participant will ingest 0 to 400ml of the study drink described below randomly. The individual will be immediately positioned in the supine position and another ultrasound scan will be carried out. Again in the RLD position.
- Primary Outcome Measures
Name Time Method Diagnostic accuracy of the model 1 year
- Secondary Outcome Measures
Name Time Method
Trial Locations
- Locations (1)
Fudan University affiliated Huashan Hospital
🇨🇳Shanghai, China