Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2025 May 29;20(1):78.
doi: 10.1186/s13020-025-01118-w.

Clinical study of intelligent tongue diagnosis and oral microbiome for classifying TCM syndromes in MASLD

Jialin Deng  1 , Shixuan Dai  1 , Shi Liu  1 , Liping Tu  1 , Ji Cui  1 , Xiaojuan Hu  1 , Xipeng Qiu  2 , Hao Lu  3 , Tao Jiang  4 , Jiatuo Xu  5
Affiliations

Clinical study of intelligent tongue diagnosis and oral microbiome for classifying TCM syndromes in MASLD

Jialin Deng et al. Chin Med. .

Abstract

Background: This study aimed to analyze the tongue image features and oral microbial markers in different TCM syndromes related to metabolic dysfunction-associated steatotic liver disease (MASLD).

Methods: This study involved 34 healthy volunteers and 66 MASLD patients [36 with Dampness-Heat (DH) and 30 with Qi-Deficiency (QD) syndrome]. Oral microbiome analysis was conducted through 16S rRNA sequencing. Tongue image feature extraction used the Uncertainty Augmented Context Attention Network (UACANet), while syndrome classification was performed using five different machine learning methods based on tongue features and oral microbiota.

Results: Significant differences in tongue color, coating, and oral microbiota were noted between DH band QD syndromes in MASLD patients. DH patients exhibited a red-crimson tongue color with a greasy coating and enriched Streptococcus and Rothia on the tongue. In contrast, QD patients displayed a pale tongue with higher abundances of Neisseria, Fusobacterium, Porphyromonas and Haemophilus. Combining tongue image characteristics with oral microbiota differentiated DH and QD syndromes with an AUC of 0.939 and an accuracy of 85%.

Conclusion: This study suggests that tongue characteristics are related to microbial metabolism, and different MASLD syndromes possess distinct biomarkers, supporting syndrome classification.

Keywords: Machine learning; Metabolic dysfunction-associated steatotic liver disease (MASLD); Microbial metabolism; TCM syndromes; Tongue diagnosis.

PubMed Disclaimer

Conflict of interest statement

Declarations. Ethics approval and consent to participate: This study has been approved by the ethics committee of Shuguang Hospital Affiliated to Shanghai University of TCM. The ethics committee number is 2020-916-125-01, and the clinical trial registration number is ChiCTR2100043546. Consent for publication: Not applicable. Competing interests: The authors declare that they have no competing interests.

Figures

Fig. 1
Fig. 1
Display of TFDA-1 tongue diagnostic instrument. A is the device diagram of TFDA-1, B is the shooting interface picture
Fig. 2
Fig. 2
Intelligent tongue image analysis based on UACANet
Fig. 3
Fig. 3
Program flowchart
Fig. 4
Fig. 4
Specific tongue image features
Fig. 5
Fig. 5
Differences in the relative abundances of oral microbiota between the Dampness-Heat syndrome and Qi-Deficiency syndrome groups. (A) The α diversity of oral microbiota; (B) PCoA of the oral microbiota; (C) Venn diagrams showing the unique and shared species between two groups; (D) Stacked bar plots showing the relative abundances of oral microbiota at the phylum level in participants; (E) Stacked bar plots showing the relative abundances of oral microbiota at the genus level in participants; (F) At the genus level, the flora with statistically significant differences between the two groups (*P < 0.05, **P < 0.01); (G) LEfSe analysis of oral signature microbiota in two groups
Fig. 6
Fig. 6
Heat map of the correlation between oral microbiota and tongue image parameters
Fig. 7
Fig. 7
To construct a syndrome classification model of Dampness-Heat syndrome and Qi-Deficiency syndrome groups. (A) Tongue image features were used to construct a syndrome classification model of the Receiver Operating Characteristic curve; (B)Tongue features combined with oral flora were used to construct a syndrome classification model of the Receiver Operating Characteristic curve; (C) SHAP Summary Plot
See this image and copyright information in PMC

References

    1. Alisi A, McCaughan G, Grønbæk H. Role of gut microbiota and immune cells in metabolic-associated fatty liver disease: clinical impact. Hepatol Int. 2024. 10.1007/s12072-024-10674-6. - PubMed
    1. Quetglas-Llabrés MM, Monserrat-Mesquida M, Bouzas C, García S, Mateos D, Casares M, et al. Effects of a two year lifestyle intervention on intrahepatic fat reduction and renal health: mitigation of inflammation and oxidative stress, a randomized trial. Antioxid Basel Switz. 2024;13:754. 10.3390/antiox13070754. - PMC - PubMed
    1. Beygi M, Ahi S, Zolghadri S, Stanek A. Management of metabolic-associated fatty liver disease/metabolic dysfunction-associated steatotic liver disease: from medication therapy to nutritional interventions. Nutrients. 2024;16:2220. 10.3390/nu16142220. - PMC - PubMed
    1. Zong X, Zhang H, Yang T. Analysis of traditional Chinese medicine syndrome types and frequency changes of CD8+ and CD25+ T cells in metabolic-related fatty liver disease. Cell Mol Biol Noisy Gd Fr. 2023;69:126–31. 10.14715/cmb/2023.69.15.22. - PubMed
    1. Thirunavukkarasu U, Umapathy S, Ravi V, Alahmadi TJ. Tongue image fusion and analysis of thermal and visible images in diabetes mellitus using machine learning techniques. Sci Rep. 2024;14:14571. 10.1038/s41598-024-64150-0. - PMC - PubMed

LinkOut - more resources