Related Experiment Video
Updated: Sep 17, 2025
06:17
Murine Model of Central Venous Stenosis using Aortocaval Fistula with an Outflow Stenosis
Published on: July 11, 2019
7.5K
Method for Predict Stenosis of Arteriovenous Fistula Patients Based on Machine Learning
Yunghsin Chen1, Wei-Tse Hsu2, Christopher Chen1
1Above Care Inc., San Jose, California, USA.
Seminars in Dialysis
|July 2, 2025
Summary
Artificial intelligence can predict arteriovenous fistula (AVF) stenosis by analyzing sound waves, offering a non-invasive method for monitoring. This technology aids in early detection, potentially reducing complications and improving hemodialysis access survival.
Area of Science:
- Nephrology and Biomedical Engineering
- Application of Artificial Intelligence in Medical Diagnostics
Background:
- Arteriovenous fistula (AVF) is crucial for hemodialysis, but stenosis and thrombosis are common complications.
- Current physical examination methods for AVF stenosis lack consistent diagnostic accuracy.
- Analyzing AVF sound waves offers a potential non-invasive monitoring approach.
Purpose of the Study:
- To evaluate the efficacy of an AI algorithm in detecting arteriovenous fistula (AVF) stenosis using digitalized sound wave analysis.
- To determine the sensitivity and specificity of AI in classifying AVF stenosis severity.
Main Methods:
- Machine learning algorithm trained on digitally recorded AVF sounds from 199 hemodialysis patients.
- Stenosis severity classified into significant (>70%) or non-significant based on angiography.
- Algorithm tested on 30% of the recorded data.
Main Results:
- The AI model achieved a 94.1% sensitivity and 81.7% specificity in classifying AVF stenosis.
- Successfully differentiated between significant and insignificant stenosis based on acoustic signals.
- 199 patients were enrolled, with 96 having significant stenosis.
Conclusions:
- AI-powered analysis of AVF digital sound waves is a viable and non-invasive method for predicting stenosis.
- This technique facilitates early detection and management of AVF stenosis, preventing access loss.
- Potential for developing remote monitoring systems for AVF stenosis, crucial in pandemic situations.
