电子鼻
乳果糖
气体分析呼吸
小肠细菌生长过度
医学
胃肠病学
呼吸试验
生物医学工程
鼻子
内科学
气体分析
诊断准确性
自动化方法
作者
Xinxin He,Ping Guo,Yinhua Hu,Xuyang An,Shuai Liang,Tiezhu Liu,Tong Zhang,Guohua Cai,Yanjie Chu,Jia Zhang
出处
期刊:ACS Sensors
[American Chemical Society]
日期:2026-04-13
卷期号:11 (4): 3442-3454
标识
DOI:10.1021/acssensors.6c00086
摘要
Exhaled human breath contains disease-related biomarkers, enabling non-invasive medical diagnostics. Small intestinal bacterial overgrowth (SIBO) is traditionally diagnosed through hydrogen measurement, a method that offers limited specificity and sensitivity. Herein, we introduce a simplified Pt/Pd-SnO 2 electronic composed of using only two sensing materials to achieve multi-gas detection. By simultaneously monitoring H 2, CO, ethanol, and acetone, the system provides a more informative diagnostic profile than conventional single-marker tests. Furthermore, the electronic nose employs dual operating temperatures (135 and 190 °C) to enhance gas discrimination. Machine learning-assisted signal processing further optimizes the system’s ability to differentiate gases, enabling more accurate detection in clinical settings. It achieves a 93% classification accuracy and a mean squared error of 0.78 for gas concentration estimation. Importantly, the system was validated using real clinical breath samples collected after lactulose ingestion, where SIBO patients exhibited markedly elevated responses (7.62−18.77) compared to healthy individuals (2.45−4.91). This clinically validated, multi-gas platform offers improved accuracy and practical potential for SIBO diagnosis.
科研通智能强力驱动
Strongly Powered by AbleSci AI