基于机器学习的胃肠道疾病舌诊模型构建

doi:10.6040/j.issn.1671-7554.0.2023.0435

摘要/Abstract

摘要： 目的构建基于机器学习的胃肠道疾病舌诊模型,以寻求更加方便、经济的方式实现对常见胃肠道疾病的非侵入性诊断。方法前瞻性收集接受电子内镜检查的948名受试者的舌象图片,经过质量筛选,最终获得符合应用标准的3 140张图片构成本研究使用的舌象数据集。对原始舌象数据进行预处理、特征提取与模式识别,在传统机器学习方法的基础之上,提出一种从特征融合和决策融合两个方面实现信息融合的方法,以此构建以舌象特征为输入的胃肠道疾病舌诊模型。结果本研究构建的基于舌象的信息融合诊断模型的曲线下面积(area under the curve, AUC)为0.808,高于单一手工特征(AUC=0.769)和深度特征(AUC=0.779)模型;使用BSFCM混合采样方法进行样本增强提高了该模型对幽门螺杆菌(Helicobacter pylori, H.pylori)感染(AUC=0.816)、胆汁反流(AUC=0.829)、反流性食管炎(AUC=0.800)、胃糜烂(AUC=0.833)和十二指肠糜烂(AUC=0.818)的分类性能。结论本研究构建的基于机器学习的智能舌诊模型对多种胃肠道疾病具有较高的区分度,或为胃肠道疾病的诊断与筛查提供一种新的、有价值的思路与方法。

关键词: 人工智能, 舌象, 胃肠道疾病, 机器学习, 舌诊模型

Abstract: Objective To construct a machine learning(ML)-based tongue diagnostic model for the diagnosis of gastrointestinal diseases so as to realize the non-invasive auxiliary diagnosis of common gastrointestinal diseases in a more convenient and faster way. Methods Tongue images of 948 subjects who underwent electronic endoscopy were prospectively collected. After quality screening, 3,140 images that met the application criteria were finally obtained to constitute the tongue image data set, which underwent preprocessing, feature extraction and pattern recognition. On the basis of traditional machine learning methods, a method to realize information fusion in terms of feature fusion and decision fusion was proposed, and a tongue diagnosis model of gastrointestinal diseases was constructed. Results The area under the curve(AUC)of the model was 0.808, which was higher than that of the single handcrafted feature(AUC=0.769)and deep feature(AUC=0.779)models. Sample enhancement using the BSFCM hybrid sampling method improved the models performance for Helicobacter pylori(H. pylori) infection(AUC=0.816), bile reflux(AUC=0.829), reflux esophagitis(AUC=0.800), gastric erosion(AUC=0.833)and duodenal erosion(AUC=0.818). Conclusion The intelligent tongue diagnostic model based on ML constructed in this study shows a high degree of differentiation for a variety of gastrointestinal diseases, and may provide a new and valuable idea and method for the diagnosis and screening of gastrointestinal diseases.

Key words: Artificial intelligence, Tongue image, Gastrointestinal diseases, Machine learning, Tongue diagnosis model

中图分类号:

R241.25

张景慧,王娟,赵玉洁,段淼,刘毅然,林敏娟,谯旭,李真,左秀丽. 基于机器学习的胃肠道疾病舌诊模型构建[J]. 山东大学学报 (医学版), 2024, 62(1): 38-47.

ZHANG Jinghui, WANG Juan, ZHAO Yujie, DUAN Miao, LIU Yiran, LIN Minjuan, QIAO Xu, LI Zhen, ZUO Xiuli. Construction of a machine learning-based tongue diagnosis model for gastrointestinal diseases[J]. Journal of Shandong University (Health Sciences), 2024, 62(1): 38-47.

参考文献

[1] Wang YC, Huang YT, Chase RC, et al. Global burden of digestive diseases: a systematic analysis of the global burden of diseases study, 1990 to 2019[J]. Gastroenterology, 2023, 165(3): 773-783.
[2] Wang WY, Zhou H, Wang YF, et al. Current policies and measures on the development of traditional Chinese medicine in China[J].Pharmacol Res,2021,163:105187. doi:10.1016/j.phrs.2020.105187.
[3] 王庆盛, 高慧, 许朝霞, 等. 冠心病及其不同合并病患者的舌诊参数特征分析[J]. 中华中医药杂志, 2022, 37(3): 1316-1320. WANG Qingsheng, GAO Hui, XU Zhaoxia, et al. Analysis of tongue diagnostic parameters in patients with coronary heart disease and its different complications[J]. China Journal of Traditional Chinese Medicine and Pharmacy, 2022, 37(3): 1316-1320.
[4] 于然, 娄彦妮, 梁婉娴, 等. 基于食管癌高发区人群筛查探索反流性食管炎及Barrett食管的舌象转化规律[J]. 中医杂志, 2021, 62(9): 782-788. YU Ran, LOU Yanni, LIANG Wanxian, et al. The change characteristics of tongue manifestation of reflux esophagitis and barretts esophagus based on an esophageal carcinoma screening study in high-risk areas of China[J]. Journal of Traditional Chinese Medicine, 2021, 62(9): 782-788.
[5] Wu TC, Lu CN, Hu WL, et al. Tongue diagnosis indices for gastroesophageal reflux disease: a cross-sectional, case-controlled observational study[J]. Medicine, 2020, 99(29): e20471. doi:10.1097/MD.0000000000020471.
[6] Hou B, Zeng Y, Ling H, et al. Correlation between helicobacter pylori infection and tongue manifestations: a meta-analysis[J]. Digital Chinese Medicine, 2018, 1(2): 155-163.
[7] Haug CJ, Drazen JM. Artificial intelligence and machine learning in clinical medicine, 2023[J]. N Engl J Med, 2023, 388(13): 1201-1208.
[8] Litjens G, Kooi T, Bejnordi BE, et al. A survey on deep learning in medical image analysis[J]. Med Image Anal, 2017, 42: 60-88. doi:10.1016/j.media.2017.07.005.
[9] Fan SY, Chen B, Zhang XR, et al. Machine learning algorithms in classifying TCM tongue features in diabetes mellitus and symptoms of gastric disease[J]. Eur J Integr Med, 2021, 43: 101288. doi:10.1016/j.eujim.2021.101288.
[10] Bi WL, Hosny A, Schabath MB, et al. Artificial intelligence in cancer imaging: clinical challenges and applications[J]. CA Cancer J Clin, 2019, 69(2): 127-157.
[11] Sun TG, Mao L, Chai ZK, et al. Predicting the proliferation of tongue cancer with artificial intelligence in contrast-enhanced CT[J]. Front Oncol, 2022,12: 841262. doi:10.3389/fonc.2022.841262.
[12] Yuan L, Yang L, Zhang SC, et al. Development of a tongue image-based machine learning tool for the diagnosis of gastric cancer: a prospective multicentre clinical cohort study[J]. E Clinical Medicine, 2023, 57: 101834. doi:10.1016/j.eclinm.2023.101834.
[13] Li MY, Zhu DJ, Xu W, et al. Application of U-net with global convolution network module in computer-aided tongue diagnosis[J]. J Healthc Eng, 2021: 5853128. doi:10.1155/2021/5853128.
[14] Ma CZ, Zhang P, Du SY, et al. Construction of tongue image-based machine learning model for screening patients with gastric precancerous lesions[J]. J Pers Med, 2023, 13(2): 271. doi:10.3390/jpm13020271.
[15] Wang XZ, Luo SY, Tian GH, et al. Deep learning based tongue prickles detection in traditional Chinese medicine[J]. Evid Based Complement Alternat Med, 2022: 5899975. doi:10.1155/2022/5899975.
[16] 姜楠, 袁莉, 汪莉, 等. 初诊胃癌患者舌象特征的客观化研究[J]. 中华中医药杂志, 2023, 38(1): 427-433. JIANG Nan, YUAN Li, WANG Li, et al. Objective study on tongue features of newly diagnosed gastric cancer patients[J]. China Journal of Traditional Chinese Medicine and Pharmacy, 2023, 38(1): 427-433.
[17] Jiang T, Guo XJ, Tu LP, et al. Application of computer tongue image analysis technology in the diagnosis of NAFLD[J]. Comput Biol Med, 2021,135:104622. doi:10.1016/j.compbiomed.2021.104622.
[18] Li J, Chen QG, Hu XJ, et al. Establishment of noninvasive diabetes risk prediction model based on tongue features and machine learning techniques[J]. Int J Med Inform, 2021, 149: 104429. doi:10.1016/j.ijmedinf.2021.104429.
[19] Olaf R, Philipp F, Thomas B. U-Net: convolutional networks for biomedical image segmentation[C]. Medical Image Computing and Computer-Assisted Intervention-MICCAI 2015: 18th International Conference, Munich, Germany, October 5-9, 2015, Proceedings, Part III 18, 2015: 234-241.
[20] Mohanaiah P, Sathyanarayana P, GuruKumar L, et al. Image texture feature extraction using GLCM approach[J]. In J Sci Res Publ, 2013, 3(5): 1-5.
[21] Kaplan K, Kaya Y, Kuncan M, et al. Brain tumor classification using modified local binary patterns(LBP)feature extraction methods[J]. Med Hypotheses, 2020, 139: 109696. doi:10.1016/j.mehy.2020.109696.
[22] Arivazhagan S, Ganesan L, Priyal SP. Texture classification using Gabor wavelets based rotation invariant features[J]. Pattern Recognit Lett, 2006, 27(16): 1976-1982.
[23] Hu MK. Visual pattern recognition by moment invariants[J]. IRE Trans Inf Theory, 1962, 8(2): 179-187.
[24] Liu YR, Qiao X, Gao R. Plankton classification on imbalanced dataset via hybrid resample method with LightBGM[C]. 2021 6th International Conference on Image, Vision and Computing(ICIVC). July 23-25, 2021, Qingdao, China. IEEE,2021:191-195. doi:10.1109/ICIVC52351.2021.9526988.
[25] Džeroski S, Ženko B. Is combining classifiers with stacking better than selecting the best one?[J]. Mach Learn, 2004, 54(3): 255-273.
[26] Malfertheiner P, Camargo MC, El-Omar E, et al. Helicobacter pylori infection[J]. Nat Rev Dis Primers, 2023, 9: 19. doi:10.1038/s41572-023-00431-8.
[27] Zhang LY, Zhang J, Li D, et al. Bile reflux is an independent risk factor for precancerous gastric lesions and gastric cancer: an observational cross-sectional study[J]. J Dig Dis, 2021, 22(5): 282-290.
[28] Gore JC. Artificial intelligence in medical imaging[J]. Magn Reson Imaging, 2020, 68:1-4. doi:10.1016/j.mri.2019.12.006.
[29] You YJ, Lai X, Pan Y, et al. Artificial intelligence in cancer target identification and drug discovery[J]. Signal Transduct Target Ther, 2022, 7(1): 156. doi:10.1038/s41392-022-00994-0.
[30] Lin S, Li ZG, Fu BW, et al. Feasibility of using deep learning to detect coronary artery disease based on facial photo[J]. Eur Heart J, 2020, 41(46): 4400-4411.
[31] Liang B, Li R, Lu J, et al. Tongue diagnostic parameters-based diagnostic signature in coronary artery disease patients with clopidogrel resistance after percutaneous coronary intervention[J]. Explore, 2023, 19(4): 528-535.
[32] Zhang NN, Jiang ZX, Li JX, et al. Multiple color representation and fusion for diabetes mellitus diagnosis based on back tongue images[J]. Comput Biol Med, 2023, 155: 106652. doi:10.1016/j.compbiomed.2023.106652.
[33] Zhang Q, Wen J, Zhou JH, et al. Missing-view completion for fatty liver disease detection[J]. Comput Biol Med, 2022, 150: 106097. doi:10.1016/j.compbiomed.2022.106097.
[34] Shi YL, Guo DD, Chun Y, et al. A lung cancer risk warning model based on tongue images[J]. Front Physiol, 2023, 14: 1154294. doi:10.3389/fphys.2023.1154294.
[35] Ma JJ, Wen GH, Wang CJ, et al. Complexity perception classification method for tongue constitution recognition[J]. Artif Intell Med, 2019, 96: 123-133. doi:10.1016/j.artmed.2019.03.008.
[36] He KM, Zhang XY, Ren SQ, et al. Deep residual learning for image recognition[C]. 2016 IEEE Conference on Computer Vision and Pattern Recognition(CVPR). June 27-30, 2016, Las Vegas, NV, USA. IEEE, 2016: 770-778. doi:10.1109/CVPR.2016.90.
[37] Hosny A, Parmar C, Quackenbush J, et al. Artificial intelligence in radiology[J]. Nat Rev Cancer, 2018, 18(8): 500-510.

多维度评价

Viewed

Full text

Abstract

Cited

Shared

Discussed