基于舌象高光谱技术构建慢性胃炎患者脾胃湿热证诊断模型OA
Developing a diagnostic model for spleen-stomach damp-heat syndrome in chronic gastritis patients using hyperspectral tongue imaging
目的 通过舌象高光谱数据构建慢性胃炎患者脾胃湿热证诊断模型,探究舌象高光谱在慢性胃炎患者脾胃湿热证诊断中的价值.方法 采用图-谱数据采集设备样机 V3.0,对 2024年 9月 4日至 2025年 10月29日在安徽中医药大学第二附属医院名医堂就诊的慢性胃炎患者进行智能化舌面信息采集.将患者分为脾胃湿热证组和非脾胃湿热证组,按 7:3 比例随机划分为训练集和内部验证集.对舌象图像进行预处理和分割后,提取特定波长光谱特征.使用最小绝对收缩和选择算法(LASSO)对舌象高光谱特征进行降维筛选,基于筛选出的特征构建诊断模型.采用 Pearson 相关分析评估诊断模型与临床资料(包括舌象、脉象、症状、胃镜检查结果及生活习惯)的相关性.采用受试者工作特征(ROC)曲线下面积(AUC)、灵敏度、特异度、准确度、阳性预测值、阴性预测值评价模型诊断效能;计算内部验证集五折交叉验证平均 AUC,绘制 500次 bootstrap 校准曲线和 500次 bootstrap 决策曲线分析(DCA),以评价模型校准度与临床净获益.结果 研究共纳入 383例慢性胃炎患者,分为脾胃湿热证组(219例)和非脾胃湿热证组(164例).LASSO 算法筛选出 9 个舌象高光谱特征并将其纳入诊断模型.该模型与 13 个临床变量呈显著相关(P<0.05):诊断模型与黄腻苔呈正相关,与薄白苔、黄苔、少苔/无苔呈负相关(P<0.001);与滑数脉呈正相关,与细数脉、弦脉呈负相关(P<0.001);与男性、饮酒、吸烟、慢性萎缩性胃炎、身体困重、大便黏滞呈正相关(P<0.05、P<0.01或 P<0.001).内部验证集模型 AUC为 0.777,准确度为 0.704 3,灵敏度为0.741 9,特异度为 0.660 4,阳性预测值为 0.718 8,阴性预测值为 0.686 3.五折交叉验证平均 AUC为 0.764.Hosmer-Lemeshow 拟合优度检验 χ2=5.423 8,P=0.711 5.500次 bootstrap DCA结果表明,在 0.16 至0.80的高风险阈值范围内,此模型比完全治疗或不治疗提供了更大的净收益.结论 本研究基于舌象高光谱数据构建了慢性胃炎脾胃湿热证诊断模型,该模型预测效能良好,并与脾胃湿热证的中医表征具有较好的一致性,为该证型的客观诊断提供了新依据.
Objective To construct a diagnostic model for spleen-stomach damp-heat syndrome in pa-tients with chronic gastritis(CG)based on hyperspectral tongue imaging data,and to investi-gate the diagnostic value of hyperspectral tongue imaging in identifying spleen-stomach damp-heat syndrome in CG patients. Methods Using the prototype V3.0 image-spectral data collection device,intelligent tongue surface observations were conducted on CG patients at The Mingyi Hall of The Second Affili-ated Hospital of Anhui University of Chinese Medicine between September 4,2024 and Octo-ber 29,2025.Patients were categorized into spleen-stomach damp-heat syndrome and non-spleen-stomach damp-heat syndrome groups,and randomly assigned to a training set and an internal validation set at a ratio of 7:3.Following preprocessing and segmentation of the tongue images,wavelength-specific spectral features were extracted.The least absolute shrinkage and selection operator(LASSO)algorithm was applied for dimensionality reduc-tion and feature selection of hyperspectral tongue features,and a diagnostic model was con-structed based on the selected features.Pearson correlation analysis was performed to assess the correlation between the diagnostic model and clinical data,including tongue and pulse manifestations,symptoms,gastroscopic findings,and lifestyle habits.The diagnostic perfor-mance of the model was evaluated using the area under the receiver operating characteristic(ROC)curve(AUC),sensitivity,specificity,accuracy,positive predictive value(PPV),and neg-ative predictive value(NPV).Five-fold cross-validation was performed on the internal valida-tion set to calculate the mean AUC.Bootstrap calibration curves and decision curve analysis(DCA)based on 500 resamples were performed to evaluate model calibration and clinical net benefit. Results A total of 383 CG patients were enrolled and classified into the spleen-stomach damp-heat syndrome group(n=219)and the non-spleen-stomach damp-heat syndrome group(n=164).The LASSO algorithm identified nine hyperspectral tongue features for inclu-sion in the diagnostic model.The diagnostic model was significantly correlated with 13 clini-cal variables(P<0.05).It was positively correlated with yellow greasy coating,and negatively correlated with thin white coating,yellow coating,and scanty or absent coating(P<0.001).It was positively correlated with slippery and rapid pulse,and negatively correlated with thready and rapid pulse and wiry pulse(P<0.001).Diagnostic model was positively correlated with male sex,alcohol consumption,smoking,chronic atrophic gastritis(CAG)diagnosed by gas-troscopy,heaviness of the body,and sticky stools(P<0.05,P<0.01,or P<0.001).Model vali-dation yielded an AUC of 0.777,an accuracy of 0.704 3,a sensitivity of 0.741 9,a specificity of 0.660 4,a PPV of 0.718 8,and an NPV of 0.686 3 in the internal validation set.Five-fold cross-validation produced a mean AUC of 0.764.The Hosmer-Lemeshow goodness-of-fit test indi-cated that χ2=5.423 8 and P=0.711 5.Bootstrap DCA based on 500 resamples demonstrated,within the high-risk threshold range of 0.16-0.80,the model provided greater net benefit than either full therapy or no treatment. Conclusion This study developed a diagnostic model for spleen-stomach damp-heat syn-drome among CG patients based on tongue hyperspectral data which showed promising pre-dictive performance,and the model was correlated with the traditional Chinese medicine(TCM)characteristics of spleen-stomach damp-heat syndrome,providing new evidence for the objective diagnosis of this syndrome.
姜楠;徐慧;程斌;赵建根;董昌武
安徽中医药大学中医学院,安徽 合肥 230031,中国安徽中医药大学中医学院,安徽 合肥 230031,中国安徽中医药大学中医学院,安徽 合肥 230031,中国安徽中医药大学中医学院,安徽 合肥 230031,中国安徽中医药大学中医学院,安徽 合肥 230031,中国||安徽中医药大学第二附属医院名医堂,安徽 合肥 230061,中国
慢性胃炎脾胃湿热证舌象高光谱诊断模型
Chronic gastritisSpleen-stomach damp-heat syndromeTongue imageHyperspectralDiagnostic model
《数字中医药(英文)》 2026 (3)
387-396,10
Anhui Province Science and Technology Major Project(202303a07020008),Anhui Province Clinical and Transla-tional Research Special Project(202304295107020107),and Open Competition Project of Anhui University of Chinese Medicine(2024CXMMTCM002).
评论