Identification of Interscalene Brachial Plexus on Ultrasonography Using a Deep Neural Network

Not Applicable

Completed

• Conditions: Ultrasound Therapy; Complications
• Interventions: Procedure: ultrasound examination

• Registration Number: NCT04183972
• Lead Sponsor: Huashan Hospital

Brief Summary

The purpose of the study is to develop and validate an algorithm based on deep neural networks (DNNs) to identify interscalene brachial plexus on ultrasonography automatically.

Detailed Description

The investigators plan to develop a deep learning-based network to automatically identify interscalene brachial nerves on ultrasound images. The trained model will be validated on an independent dataset. The performance of the network will also be compared against practicing anesthesiologists.

Study Timeline

• Study First Submitted: 2019-11-19
• Study First Submitted QC: 2019-11-29
• Study Start: 2019-12-01
• Study First Posted: 2019-12-03
• Primary Completion: 2020-09-30
• Study Completion: 2020-10-31
• Status Verified: 2021-06
• Last Update Submitted: 2021-06-28
• Last Update Posted: 2021-06-30

Recruitment & Eligibility

• Status: COMPLETED
• Sex: All
• Target Recruitment: 1126

Inclusion Criteria

• ASA physical status class I or II
• scheduled for elective surgery

Exclusion Criteria

• skin lesion or infection of neck
• any known peripheral neuropathy
• brachial nerve plexus injury
• previous injury or operation on neck
• pregnancy
• allergic to ultrasound gel

Study & Design

• Study Type: INTERVENTIONAL
• Study Design: SINGLE_GROUP

Arm && Interventions

Group	Intervention	Description
Image collecting Group	ultrasound examination	An computer algorithm will be developed and evaluated by these image data.

Primary Outcome Measures

Name	Time	Method
The distance of the lateral midpoints of the nerve sheath contours	immediately after the procedure	between model predictions and the ground truth; between nonexpert anesthesiologist predictions and the ground truth

Secondary Outcome Measures

Name	Time	Method
Accuracy, Sensitivity and specificity	immediately after the procedure	Accuracy, Sensitivity and specificity of the network and nonexpert anesthesiologists
The percentage of the intersection over union	immediately after the procedure	between model predictions and the ground truth; between nonexpert anesthesiologist predictions and the ground truth

Trial Locations

Locations (1)

Huashan Hospital; 🇨🇳 Shanghai, Shanghai, China