Development of a tongue image-based machine learning tool for the diagnosis of gastric cancer: a prospective multicentre clinical cohort study

Li, Yuan; Yang, Lin; Zhang, Shichuan; Xu, Zhiyuan; Qin, Jiang‐Jiang; Shi, Yunfu; Yu, Pengcheng; Wang, Yi; Bao, Zhehan; Xia, Yuhang; Sun, Juan; He, Weiyang; Chen, Tianhui; Chen, Xiaolei; Hu, Can; Zhang, Yunlong; Changwu, Dong; Zhao, Ping; Wang, Yanan; Jiang, Nan; Lv, Bin; Xue, Yingwei; Jiao, Baoping; Gao, Hongyu; Chai, Kequn; Li, Jun; Wang, Hao; Wang, Xibo; Guan, Xiaoqing; Liu, Xu; Zhao, Gang; Zheng, Zhichao; Yan, Jie; Yu, Haiyue; Yao, Xueqing; Ye, Zaisheng; You, Hua; Yu, Bin; Cheng, Xi; Zhao, Pei-Zheng; Wang, Liang; Zeng, Wenting; Tian, Yazhou; Chen, Ming; You, You; Yuan, Guihong; Gao, Xiaole; Xu, Jingli; Xu, Handong; Du, Lingbin; Zhang, Shengjie; Fu, Huanying; Cheng, Xiangdong

doi:10.1016/j.eclinm.2023.101834

Cited by 43 publications

(25 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Xu et al developed a multi-task joint learning model to segment and classify tongue images using deep neural networks, which can optimally extract tongue image features [29]. Li et al pioneered the use of both tongue images and the tongue-coating microbiome as diagnostic tools for gastric cancer [22]. Similarly, Ma et al constructed the first deep learning model for screening precancerous lesions of gastric cancer based on tongue images [21].…”

Section: Discussionmentioning

confidence: 99%

“…Ma et al constructed the deep learning model for screening precancerous lesions of gastric cancer based on tongue images [21]. Li et al found that both tongue images and the tongue-coating microbiome can be used as tools for the diagnosis of gastric cancer [22]. However, the value of these features in pre-endoscopic PLGC risk screening remains uncertain, and integrating these features based on AI to achieve a more refined PLGC risk screening still poses significant challenges.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Pre-endoscopic screening of precancerous lesions in gastric cancer using deep learning

Wang,

Zhang,

Zhang

et al. 2024

Preprint

View full text Add to dashboard Cite

Objective: Given the high cost of endoscopy in gastric cancer (GC) screening, there is an urgent need to explore cost-effective methods for the large-scale prediction of precancerous lesions of gastric cancer (PLGC). We aim to construct a hierarchical artificial intelligence-based multimodal non-invasive method for pre-endoscopic risk screening, to provide tailored recommendations for endoscopy. Design: From December 2022 to December 2023, a large-scale screening study was conducted in Fujian, China. Based on traditional Chinese medicine theory, we simultaneously collected tongue images and inquiry information from 1034 participants, considering the potential of these data for PLGC screening. Then, we introduced inquiry information for the first time, forming a multimodality artificial intelligence model to integrate tongue images and inquiry information for pre-endoscopic screening. Moreover, we validated this approach in another independent external validation cohort, comprising 143 participants from the China-Japan Friendship Hospital. Results: A multimodality artificial intelligence-assisted pre-endoscopic screening model based on tongue images and inquiry information (AITonguequiry) was constructed, adopting a hierarchical prediction strategy, achieving tailored endoscopic recommendations. Validation analysis revealed that the area under the curve (AUC) values of AITonguequiry were 0.74 for PLGC (95% confidence interval (CI) 0.71 to 0.76, p < 0.05) and 0.82 for high-risk PLGC (95% CI 0.82 to 0.83, p < 0.05), which were significantly and robustly better than those of the independent use of either tongue images or inquiry information alone, and also demonstrated superior performance compared to existing screening methodologies. In the independent external verification, the AUC values were 0.69 for PLGC and 0.76 for high-risk PLGC. Conclusion: Our AITonguequiry artificial intelligence model, for the first time, incorporates inquiry information and tongue images, leading to a higher precision and finer-grained pre-endoscopic screening of PLGC. This enhances patient screening efficiency and alleviates patient burden.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Pre-endoscopic screening of precancerous lesions in gastric cancer using deep learning

Wang,

Zhang,

Zhang

et al. 2024

Preprint

View full text Add to dashboard Cite

show abstract

“…Observing the change of tongue coating is an important part of traditional Chinese medicine diagnosis. Microbial changes in tongue coating have been observed in patients with GC (30,44).…”

Section: Discussionmentioning

confidence: 99%

The oral microbiota disorder in the GC patients revealed by 2bRAD-M

Sun

et al. 2023

Preprint

View full text Add to dashboard Cite

Background Microbiota alterations are linked with gastric cancer (GC). However, the relationship between the oral microbiota (especially oral fungi) and GC is not known. In this study, we aimed to apply 2bRAD-M to characterize the oral microbiota in GC. Methods We performed 2bRAD-M analysis in saliva and tongue coating of GC patients and healthy controls. We carried out the diversity, relative abundance, and composition analyses of saliva and tongue coating bacteria and fungi of the two groups. In addition, indicator analysis, the Gini index, and the mean decrease accuracy were used to find GC oral fungal indicator. Results In this study, fungi imbalance of saliva and tongue coating were observed in GC group. At the species level, enriched salivary and tongue coating Malassezia globosa (M. globosa) and decreased Saccharomyces cerevisiae (S. cerevisiae) were observed in the GC group. Random forest analysis indicated that salivary and tongue coating M. globosa could serve as a biomarker to diagnose gastric cancer. The Gini index and mean decrease in accuracy of saliva and tongue coating M. globosa are the largest. In addition, Saliva and tongue coating M. globosa classified GC from the control with an area under the receiver operating curve (AUC) of 0.976 and 0.846, respectively. Further ecological analysis revealed the correlations between oral bacterial and fungi. Conclusion For the first time, our data suggested that changes in oral fungi between GC and control may help deepen our understanding of the complex spectrum of the different microbiotas involved in the GC development. Although the cohort size is small, this study is the first to use 2bRAD-M to reveal that oral M. globosa can be a fungal biomarker for detecting GC.

show abstract

“…Tongue diagnosis is a method for comprehensively evaluating the functional state of the body based on an assessment of the tongue. According to studies, tongue images are more accurate than blood biomarkers in detecting gastric cancer, and tongue diagnosis can be used as a stable method for gastric cancer diagnosis ( Yuan et al, 2023 ). The condition of the gastric mucosa can be predicted by tongue images in patients with chronic gastritis ( Shang et al, 2022 ).…”

Section: Introductionmentioning

confidence: 99%

A lung cancer risk warning model based on tongue images

et al. 2023

View full text Add to dashboard Cite

Objective: To investigate the tongue image features of patients with lung cancer and benign pulmonary nodules and to construct a lung cancer risk warning model using machine learning methods.Methods: From July 2020 to March 2022, we collected 862 participants including 263 patients with lung cancer, 292 patients with benign pulmonary nodules, and 307 healthy subjects. The TFDA-1 digital tongue diagnosis instrument was used to capture tongue images, using feature extraction technology to obtain the index of the tongue images. The statistical characteristics and correlations of the tongue index were analyzed, and six machine learning algorithms were used to build prediction models of lung cancer based on different data sets.Results: Patients with benign pulmonary nodules had different statistical characteristics and correlations of tongue image data than patients with lung cancer. Among the models based on tongue image data, the random forest prediction model performed the best, with a model accuracy of 0.679 ± 0.048 and an AUC of 0.752 ± 0.051. The accuracy for the logistic regression, decision tree, SVM, random forest, neural network, and naïve bayes models based on both the baseline and tongue image data were 0.760 ± 0.021, 0.764 ± 0.043, 0.774 ± 0.029, 0.770 ± 0.050, 0.762 ± 0.059, and 0.709 ± 0.052, respectively, while the corresponding AUCs were 0.808 ± 0.031, 0.764 ± 0.033, 0.755 ± 0.027, 0.804 ± 0.029, 0.777 ± 0.044, and 0.795 ± 0.039, respectively.Conclusion: The tongue diagnosis data under the guidance of traditional Chinese medicine diagnostic theory was useful. The performance of models built on tongue image and baseline data was superior to that of the models built using only the tongue image data or the baseline data. Adding objective tongue image data to baseline data can significantly improve the efficacy of lung cancer prediction models.

show abstract

Development of a tongue image-based machine learning tool for the diagnosis of gastric cancer: a prospective multicentre clinical cohort study

Cited by 43 publications

References 45 publications

Pre-endoscopic screening of precancerous lesions in gastric cancer using deep learning

Pre-endoscopic screening of precancerous lesions in gastric cancer using deep learning

The oral microbiota disorder in the GC patients revealed by 2bRAD-M

A lung cancer risk warning model based on tongue images

Contact Info

Product

Resources

About