Characteristics of Oral Breath Odor Map of Chronic Atrophic Gastritis Patients with Dampness-Heat Syndrome： A Cross-Sectional Study
10.13288/j.11-2166/r.2024.16.010
- VernacularTitle:慢性萎缩性胃炎湿热证患者口腔呼气气味图谱特征的横断面研究
- Author:
Xuejuan LIN
1
;
Yanyu HUANG
1
;
Long ZHU
1
;
Donglin CAO
2
;
Shanshan DING
1
;
Xinghui LI
1
;
Yingying HU
1
;
Meng LAN
1
;
Weirong HUANG
3
Author Information
1. Research Base of Traditional Chinese Medicine Syndrome of Fujian University of Traditional Chinese Medicine，Fuzhou，350122
2. Department of Artificial Intelligence，School of Informatics，Xiamen University
3. Jinjiang Hospital of Traditional Chinese Medicine Affiliated to Fujian University of Traditional Chinese Medicine
- Publication Type:Journal Article
- Keywords:
chronic gastritis;
chronic atrophic gastritis;
dampness-heat syndrome;
oral breath;
odor map;
electronic nose
- From:
Journal of Traditional Chinese Medicine
2024;65(16):1687-1694
- CountryChina
- Language:Chinese
-
Abstract:
ObjectiveTo explore the recognition of oral breath odor map of chronic atrophic gastritis （CAG） patients with dampness-heat syndrome by electronic nose technique. MethodsPatients with chronic gastritis were recruited， including 60 cases in CAG group of dampness-heat syndrome， 50 cases in CAG group of non-dampness-heat syndrome， 60 cases in chronic non-atrophic gastritis （CNAG） group of dampness-heat syndrome， 50 cases in CNAG group of non-dampness-heat syndrome， and 30 cases of healthy volunteers were selected to set up the health control group. Ten cases in the CAG dampness-heat group and 50 cases in the CAG non-dampness-heat group were selected to form the CAG group， and 10 cases in CNAG dampness-heat group and 50 cases in CNAG non-dampness-heat group were selected to form the CNAG group. In addition to the health control group， the remaining patients were tested for Helicobacter pylori （Hp）； the electronic nose （GISXM-MQWA01） was used to collect the oral breath odor of all the participants to draw the mapping， and amplitudes and slopes of each curve （including curves A， B， C， D， E， F， G， H， I， J） of the oral odor mapping of health control group， CAG group， CNAG group， CAG dampness-heat group， CAG non-dampness-heat group， and CNAG dampness-heat group was compared. The modified transformer model was used to classify the odor mapping characteristics， and the confusion matrix method was used to evaluate the classification model， with metrics including accuracy and area under ROC curve （AUC）. ResultsThe Hp positivity rate in CAG dampness-heat group was 80.00% （48/60）， CAG non-dampness-heat group was 62.00% （31/50）， CNAG dampness-heat group was 46.67% （28/60）， and CNAG non-dampness-heat group was 42.00% （21/50）； the difference in Hp positivity rate between CAG dampness-heat group and CAG non-dampness-heat group was statistically significant （P＜0.05）. The amplitudes of response curves A， B， C， D， F， G， and I， and slopes of A and F in the odor mapping of the CAG group were lower than those in health control group， while the amplitude and slope of curve E were higher than those in the health control group and CNAG group （P＜0.05 or P＜0.01）； The amplitude of the response curves A， B， C， D， F， G， and I， and slopes of A， D， and F in the CNAG group were lower than those in the health control group （P＜0.05 or P＜0.01）. The amplitude of response curve D and slope of response curve J in the odor mapping of the CAG dampness-heat group were higher than those in CNAG dampness-heat group， the amplitude of curve F was lower than that in CAG non-dampness-heat group， and the amplitude of curve H and slopes of curve A， H， and J were higher than those in CAG non-dampness-heat group （P＜0.05）. The recognition accuracy of CAG group and health control group reached 77.78%， AUC = 0.88； the recognition accuracy of CAG group and CNAG group was 69.44%， AUC = 0.61； the recognition accuracy of CAG dampness-heat group and CAG non-dampness-heat group reached 75.8%， AUC=0.70. ConclusionElectronic nose technology can make a more accurate identification of the oral breath odor in CAG patients with dampness-heat syndrome， with a decrease in the amplitude of the curve F and an increase in the amplitude of the curve H and in the slopes of the curves A， H， and J may as the characteristics of their odor mapping.
