1.A machine learning-based depression recognition model integrating spirit-expression features from traditional Chinese medicine
Minghui YAO ; Rongrong ZHU ; Peng QIAN ; Huilin LIU ; Xirong SUN ; Limin GAO ; Fufeng LI
Digital Chinese Medicine 2026;9(1):68-79
Objective:
To develop a depression recognition model by integrating the spirit-expression diagnostic framework of traditional Chinese medicine (TCM) with machine learning algorithms. The proposed model seeks to establish a TCM-informed tool for early depression screening, thereby bridging traditional diagnostic principles with modern computational approaches.
Methods:
The study included patients with depression who visited the Shanghai Pudong New Area Mental Health Center from October 1, 2022 to October 1, 2023, as well as students and teachers from Shanghai University of Traditional Chinese Medicine during the same period as the healthy control group. Videos of 3 – 10 s were captured using a Xiaomi Pad 5, and the TCM spirit and expressions were determined by TCM experts (at least 3 out of 5 experts agreed to determine the category of TCM spirit and expressions). Basic information, facial images, and interview information were collected through a portable TCM intelligent analysis and diagnosis device, and facial diagnosis features were extracted using the Open CV computer vision library technology. Statistical analysis methods such as parametric and non-parametric tests were used to analyze the baseline data, TCM spirit and expression features, and facial diagnosis feature parameters of the two groups, to compare the differences in TCM spirit and expression and facial features. Five machine learning algorithms, including extreme gradient boosting (XGBoost), decision tree (DT), Bernoulli naive Bayes (BernoulliNB), support vector machine (SVM), and k-nearest neighbor (KNN) classification, were used to construct a depression recognition model based on the fusion of TCM spirit and expression features. The performance of the model was evaluated using metrics such as accuracy, precision, and the area under the receiver operating characteristic (ROC) curve (AUC). The model results were explained using the Shapley Additive exPlanations (SHAP).
Results:
A total of 93 depression patients and 87 healthy individuals were ultimately included in this study. There was no statistically significant difference in the baseline characteristics between the two groups (P > 0.05). The differences in the characteristics of the spirit and expressions in TCM and facial features between the two groups were shown as follows. (i) Quantispirit facial analysis revealed that depression patients exhibited significantly reduced facial spirit and luminance compared with healthy controls (P < 0.05), with characteristic features such as sad expressions, facial erythema, and changes in the lip color ranging from erythematous to cyanotic. (ii) Depressed patients exhibited significantly lower values in facial complexion L, lip L, and a values, and gloss index, but higher values in facial complexion a and b, lip b, low gloss index, and matte index (all P < 0.05). (iii) The results of multiple models show that the XGBoost-based depression recognition model, integrating the TCM “spirit-expression” diagnostic framework, achieved an accuracy of 98.61% and significantly outperformed four benchmark algorithms—DT, BernoulliNB, SVM, and KNN (P < 0.01). (iv) The SHAP visualization results show that in the recognition model constructed by the XGBoost algorithm, the complexion b value, categories of facial spirit, high gloss index, low gloss index, categories of facial expression and texture features have significant contribution to the model.
Conclusion
This study demonstrates that integrating TCM spirit-expression diagnostic features with machine learning enables the construction of a high-precision depression detection model, offering a novel paradigm for objective depression diagnosis.
2.Regulatory effects of G3BP2 on activation, proliferation, and migratory capacity in hepatic stellate cells
Qiqi DONG ; Wenjie SUN ; Minghui LI ; Jingjing YANG ; Renpeng ZHOU ; Wei HU ; Chao LU
Acta Universitatis Medicinalis Anhui 2026;61(3):501-508
ObjectiveTo investigate the role of Ras-GTPase-activating protein SH3 domain-binding protein 2 (G3BP2) in regulating the activation, proliferation, and migration of hepatic stellate cells (HSCs). MethodsThe mouse HSCs (JS-1 cell line) were treated with 5 μg/L transforming growth factor-beta 1(TGF-β1) for 24 hours to establish an HSC activation and proliferation model. A G3BP2 knockdown system was constructed using siRNA interference technology. The experiment was divided into four groups: Control, TGF-β1 treatment, TGF-β1+si-NC, and TGF-β1+ G3BP2-siRNA. The expression levels of key fibrosis indicators, including type I collagen (Collagen I), α-smooth muscle actin (α-SMA), and G3BP2, were detected by Western blot and RT-qPCR. Cell proliferation activity was assessed using the CCK-8 proliferation assay kit and EdU fluorescence labeling technology. Cell migration ability was analyzed by scratch wound healing assay and Transwell migration assay. The formation level of stress granules was quantified by immunofluorescence microscopy to investigate the effects of G3BP2 on stress granule formation in activated HSCs. ResultsStimulation with TGF-β1 upregulated the expression of G3BP2 in JS-1 cells (RT-qPCR: P0.000 1; Western blot: P0.000 1), while a downward trend in its expression was observed in the G3BP2‑silenced group (RT-qPCR: P0.01; Western blot: P0.000 1). Compared with the control group, the TGF-β1 group exhibited increased protein expression levels of α-SMA and Collagen I (RT-qPCR: both P0.01; Western blot: P0.01 and P0.05, respectively), concomitant with an increased number of stress granules and enhanced cell proliferation and migration capacity (all P0.001). The experimental results demonstrated that G3BP2 knockout effectively reversed the aforementioned phenotypes, with the G3BP2-silenced group showing reduced expression of fibrotic markers (all P0.01), decreased stress granule formation (P0.01), and reduced cell proliferation and migration capacity (all P0.05), compared to the negative control group. ConclusionG3BP2 enhances the activation, proliferation, and migration of HSCs by promoting the formation of stress granules, thereby accelerating the pathological progression of liver fibrosis. This suggests that stress granules may serve as important regulators in controlling the activation, proliferation, and migration of HSCs.
3.Characteristics of the Five Tones of Parkinson's Disease Patients Based on the Theory of "Five-Viscera Phonology":A Case-Control Study
Wenxue YANG ; Xuelin WANG ; Min LI ; Shaodan LI ; Minghui YANG
Journal of Traditional Chinese Medicine 2025;66(2):165-169
ObjectiveTo study the characteristics of the five tones of Parkinson's disease (PD) patients based on the theory of "five-viscera phonology", and provide references for the syndrome differentiation and treatment of PD. MethodsA total of 272 cases of PD patients were collected as the PD group, and 240 individuals, including patient family members and hospital staff, were recruited as the control group. The 25-tone analyzer was used to collect the five-tone characteristics of both PD patients and control group participants. The participants were then stratified into three age groups, 41~55, 56~70, and 71~85 years old, and categorized by gender (male and female) for analysis. The frequency and composition ratio of the five tones were analyzed for both groups across the different age ranges and genders. Additionally, the average voice frequency of each participant was calculated to compare differences between groups, stratified by age range and gender. ResultsIn the 41~55 and 56~70 age groups, male participants in the PD group and the control group exhibited the highest frequency of Yu (羽) tone, with the PD group showing a significantly higher composition ratio of Yu tone compared to the control group (P<0.05); for males in the 56~70 age group, the composition ratios of Shang (商) and Zhi (徵) tone in the PD group were lower than those in the control group (P<0.05). For males in the 71~85 age group, both the PD group and the control group had the highest frequency of Yu tone, but there was no statistically significant difference in the composition ratios of the five tones between groups (P>0.05). For female participants in the PD group across all age groups, Yu tone was the most frequent, whereas for the control group, Jue (角) tone was the most frequent in all age groups, and the composition ratio of Yu tone in the PD group was significantly higher than that in the control group across all age groups (P<0.05); in the 56~70 age group, the composition ratio of Jue tone was lower in the PD group compared to that in the control group (P<0.05). Regarding voice frequency, males in the PD group aged 41~55 and 56~70 had higher voice frequency than those in the control group of the same age range, and similarly, females in the PD group aged 56~70 and 71~85 had higher voice frequency than their counterparts in the control group (P<0.05). ConclusionPD patients have a voice with a higher frequency and an increased proportion of Yu tone in their five-tone distribution. According to the theory of five-viscera phonology, PD patients may have disease mechanism of kidney essence deficiency.
4.Lung transplantation and post-transplantation changes in patients with pneumoconiosis: a clinicopathological analysis of 28 cases
Sisi WANG ; Minghui ZHENG ; Ying WANG ; Junjie ZENG ; Li WEI ; Zeheng MA ; Fangfang FU ; Qiuyu LIU
Chinese Journal of Pathology 2025;54(12):1304-1310
Objective:To investigate the clinicopathological characteristics of lung transplantation and post-transplantation changes in patients with pneumoconiosis.Methods:A retrospective study was conducted to analyze the clinical and pathological data of 28 patients with pulmonary silicosis who underwent lung transplantation and were managed at the Department of Internal Medicine, Henan Provincial People′s Hospital, Zhengzhou, China from January 2015 to December 2024. Among them, 8 patients underwent lung biopsy 6-20 months after transplantation to evaluate the histopathological changes of the recipient and the donor lungs post-transplantation. The expression of relevant indicators was examined using immunohistochemical EnVision staining, while presence of microorganisms was assessed using histochemical special staining. The patients were all followed up.Results:Among the 28 patients with pneumoconiosis who underwent lung transplantation, 26 were male and 2 were female, with a male-to-female ratio of 13∶1. Their ages ranged from 23 to 68 years, median 50.0 (46.0, 53.5) years. They were diagnosed with pneumoconiosis at local occupational disease prevention and control centers for 3 to 15 years (mean, 9.65 years), including 13 left single lung transplants and 15 right single lung transplants. Gross examination showed fleshy nodules with irregular cystic cavities at the periphery. The cut surfaces exhibited gray-brown color and firm texture. Microscopically, most alveolar structures of the lung were obliterated, with nodular or diffuse proliferation of collagen fibers accompanied by hyaline degeneration. Focal massive carbon dust deposition and massive silicotic fibrosis were observed, surrounded by lung parenchyma with emphysematous changes and localized bullae formation. Seven patients underwent re-biopsy after transplantation that showed extensive infiltration of inflammatory cells. In 4 cases, microscopy revealed complete coagulative necrosis, with negative acid-fast staining and TB-DNA results. Of the 4 cases, 3 cases exhibited Aspergillus infection confirmed by Grocott′s methenamine silver and PAS stains, while 2 cases showed chronic bronchitis with squamous metaplasia. Follow-up revealed that 8 patients died of acute respiratory failure due to severe infection, while the remaining 20 demonstrated significant postoperative improvement in lung function.Conclusions:For patients with advanced pulmonary dust deposition disease who undergo lung transplantation, it is necessary to conduct standardized sampling and pathological assessment of the recipient lungs. In the early post-transplant period, the complications of re-biopsy tissues are mainly fungal infections. The combination of morphological manifestations and immunohistochemical detection is helpful to distinguish infection from rejection reactions. At the same time, it is essential to integrate clinical information and laboratory results to provide post-transplantation pathological assessment for individualized treatment.
5.From"insufficiency of ZhiYi"to anxiety onset:a preliminary construction of the emotion-pathogenesis hypothesis based on body-spirit integration theory
Mingzhou GAO ; Minghui HU ; Hongwei DONG ; You LI ; Yue ZHAO ; Xinyu WANG ; Zifa LI ; Xiwen GENG ; Sheng WEI ; Hao ZHANG
Acta Laboratorium Animalis Scientia Sinica 2025;33(9):1320-1328
Anxiety is a major emotional disorder manifested in the individual's expectation of future threats.The incidence rate of anxiety is about 7.3%,with the highest lifetime prevalence rate among mental health conditions.The mechanism of anxiety overlaps with depression,and anxiety is a typical symptom of various mental diseases or emotional disorders in traditional Chinese medicine.The high rates of comorbidity and disability pose serious threats to people's health.Animal models are important tools for studying anxiety and are of great use for deciphering the pathogenesis of anxiety and for developing drugs.The traditional paradigm of stress-induced anxiety,however,is relatively limited.Based on traditional theory combined with clinical and animal experimental data,we propose a new hypothesis of"insufficiency of ZhiYi'causing anxiety,defined as"an anxiety state induced by the inability of an individual to meet their own needs,limited or lacking after multiple attempts,rendered hindered and powerless by an inability to meet their desires".This hypothesis is more in line with the typical manifestations of despair,lack of pleasure,and social withdrawal in clinical patients,and is supported by traditional theory and experimental data showing"hunger but unable to eat,food but unable to obtain,and gain but not full".Based on this,the established modeling paradigm is easy to apply,with good repeatability and low cost,and can be used to establish anxiety models in rats and mice,to provide a theoretical and model basis for the development and pharmacological evaluation of anti-anxiety drugs.
6.Association between A Body Shape Index (ABSI) and sarcopenia among older adults in China: evidence from the CHARLS
Tong LI ; Minghui DU ; You LV ; Jia CHONG ; Yuwei LI ; Fang WANG ; Yan WANG
Chinese Journal of Clinical Nutrition 2025;33(4):275-281,289
Objective:To explore the association between A Body Shape Index (ABSI) and sarcopenia in Chinese older adults.Methods:Totally 1 728 participants in waves 1 and 3 of the China Health and Retirement Longitudinal Survey (CHARLS) were enrolled in this cohort-based cross-sectional analysis. ABSI was calculated by using anthropometric measurements from CHARLS. According to the Asian Working Group for Sarcopenia (AWGS) 2019 consensus update on sarcopenia, low muscle strength and low muscle mass were used as the diagnostic criteria for sarcopenia. Logistic regression and restricted cubic spline (RCS) were used to assess the relationship between ABSI and sarcopenia, and subgroup and interaction analyses were conducted to evaluate the influence of other factors on this association.Results:High ABSI showed a significant association with the risk of sarcopenia among older adults in China, and RCS analysis demonstrated a positive linear correlation between ABSI and sarcopenia. Additionally, the relationship of ABSI with sarcopenia was significantly modified by educational level and depressive status.Conclusion:ABSI, a novel anthropometric measure, serves as an effective threshold for assessing sarcopenia risk in the Chinese older population, providing a basis for developing clinical interventions and public health strategies for the prevention and early identification of sarcopenia.
7.Analysis of risk factors of bleeding caused by thrombus head shedding in patients with esophageal varices during endoscopic diagnosis and treatment and construction of nomogram prediction model
Minghui WANG ; Xiaofeng LIU ; Qun LI ; Jing WANG ; Wenbo LI
China Journal of Endoscopy 2025;31(7):25-30
Objective To investigate the risk factors of thrombus head shedding and bleeding during endoscopic treatment of patients with esophageal varices veins,and build a risk prediction model.Methods 209 cases of esophageal varices with thrombus head found by endoscopy from February 2009 to July 2024 were retrospectively analyzed,and divided into non-shedding group(n=186)and shedding group(n=23)according to whether thrombus head shedding bleeding occurred during endoscopic treatment.Clinical and endoscopic data of the two groups were compared.Statistically significant factors were included in multivariate Logistic regression analysis and their independent risk factors were explored.A nomogram risk prediction model was constructed by R software,and its prediction efficiency was evaluated.Results Multivariate Logistic regression analysis showed that red thrombus head((O^R)=6.231,95%CI:1.748~22.208),thrombus head diameter≥3 mm((O^R)=4.355,95%CI:1.341~14.144),conical thrombus head((O^R)=8.555,95%CI:2.427~30.154),and the presence of hematoma in the stomach((O^R)=7.079,95%CI:1.665~30.103)were risk factors for bleeding after thrombus head shedding during endoscopic diagnosis and treatment.A nomogram prediction model was constructed with a sensitivity of 0.870(95%CI:0.732~1.000)and a specificity of 0.887(95%CI:0.842~0.933).Conclusion The thrombus head being red,conical in shape,with a diameter of≥3 mm and the presence of hematoma in the stomach are independent risk factors for thrombus head shedding and bleeding during endoscopic diagnosis and treatment.Timely intervention should be made for the above factors to benefit patients.
8.Development of a clinical prediction model for cervical instability in young and middle-aged adults based on machine learning
Jing LI ; Guangqi LU ; Minghui ZHUANG ; Ying CUI ; Zhangjingze YU ; Xinyue SUN ; Mingming MA ; Liguo ZHU ; Jie YU
Chinese Journal of Tissue Engineering Research 2025;29(33):7203-7210
BACKGROUND:Cervical instability is a common orthopedic disease in young and middle-aged people,and is the early manifestation of cervical spondylosis,which has a great impact on the quality of life of patients.Therefore,early diagnosis of cervical instability to implement early intervention has positive clinical and social significance.OBJECTIVE:The clinical prediction model of cervical instability in young and middle-aged people was constructed based on machine learning to realize early screening of cervical instability in young and middle-aged people before X-ray examination.METHODS:From September 2022 to October 2023,155 young and middle-aged adults with cervical instability and 88 with non-cervical instability recruited through recruitment advertisements and spinal department outpatient of Wangjing Hospital,China Academy of Chinese Medical Sciences were selected as research subjects.The research subjects'general information,living and working habits,discomfort symptoms,visual analog scale score,Neck Disability Index,and 36-ltem Short Form Health Survey were collected on site based on questionnaires.The above information was used as predictive factors.After screening,six machine learning algorithms of Support Vector Machine,LightGBM,RandomForest,Logistic,AdaBoost,and XGBClassifier were used to train the model by ten-fold cross-validation method,and the clinical prediction model of cervical instability was constructed.Area under the curve was used as the main evaluation index.Univariate analysis was performed on the predictors,and SHAP method was used to rank the importance of the predictors.Correlation heat maps were used to show the degree of linear correlation between the predictors and the cervical instability.RESULTS AND CONCLUSION:(1)Among the six machine learning models,RandomForest model was chosen as the final prediction model,including nine predictors,such as age,body mass index,neck circumference/neck length,visual analog scale score,Neck Disability Index,bodily pain,general health,vitality,and mental health,area under the curve=0.725 4,and the calibration degree was good.It could be used as a reference tool for early screening of cervical instability in young and middle-aged people.(2)There were significant differences in age,visual analog scale score,Neck Disability Index,bodily pain,general health,and vitality between the two groups(P<0.05).(3)The order of importance of predictors was age,Neck Disability Index,visual analog scale score,general health,body mass index,vitality,bodily pain,neck circumference/neck length,mental health,among which age,visual analog scale score,Neck Disability Index were positively correlated with cervical instability,while general health,body mass index,vitality,bodily pain,neck circumference/neck length,and mental health were negatively correlated with cervical instability.
9.Study on the molecular mechanism of miR-886-5p targeting BAX to inhibit proliferation, migration and invasion of liver cancer cells
Shaopeng LIU ; Haichao LIU ; Hongxian YAN ; Minghui BAI ; Jixiang ZHANG ; Yingbo LI ; Chuang WANG ; Xiang ZOU
Chinese Journal of Hepatobiliary Surgery 2025;31(8):621-627
Objective:To investigate the molecular mechanism of miR-886-5p targeting BCL-2-associated X protein (BAX) to inhibit the proliferation, migration, and invasion of liver cancer cells.Methods:mRNA expression data of HCC patients were obtained from the Starbase database, including 370 liver cancer samples and 50 normal liver tissue samples adjacent to the cancer. Analyze the expression of miR-886-5p in the previously obtained data and investigate the relationship between miR-886-5p and BAX in liver cancer samples. After transfection of the corresponding plasmids into Huh7 and HepG2 cells, the following groups were established. Analyze the interaction between miR-886-5p and BAX in vitro, detect the protein expression by Western blotting, and verify the targeting relationship between the two by dual luciferase reporter gene assay.Results:Starbase database analysis found that the standardized expression level of miR-886-5p in 370 liver cancer samples was lower than that in normal liver tissue samples (0.12±0.07 vs. 0.73±0.27, t=-15.71, P<0.001), and the expression level of miR-886-5p was positively correlated with the expression level of BAX ( r=0.152, P=0.003). qRT-PCR analysis showed that the expression level of miR-886-5p in HL-7702 cells was higher than that in Huh7 (4.57±0.06 vs. 1.61±0.40, t=32.48) and HepG2 (4.57±0.06 vs. 1.03±0.13, t=143.9), and the expression level of BAX in HL-7702 cells was higher than that in Huh7 (4.01±0.12 vs. 1.28±0.09, t=82.20) and HepG2 (4.01±0.12 vs. 1.30±0.11, t=80.76), the differences were statistically significant (all P<0.001). The proliferation, migration, and invasion abilities of Huh7 and HepG2 cells decreased after transfection with miR-886-5p mimics, while the expression levels of BAX at the mRNA and protein levels increased. However, after inhibiting the expression of miR-886-5p, the above indicators of cells were the opposite, and the dif-ferences were statistically significant (all P<0.05). The viability, EdU positivity rate, cell migration rate, and number of transmembrane cells in the miR-886-5p+ BAX group were lower than those in the BAX group, and the relative expression levels of miR-886-5p, BAX mRNA, and BAX protein were higher than those in the BAX group. However, the above indicators in the Sponge+ BAX group showed opposite trends, and all differences were statistically significant (all P<0.05). There was a targeted binding site between miR-886-5p and BAX. Conclusion:Both miR-886-5p and BAX are downregulated in liver cancer, and miR-886-5p inhibits the proliferation, migration, and invasion of liver cancer cells by targeting BAX.
10.Clinical study of self-help inflatable balloon for refractory benign upper digestive tract stricture
Qianqian DONG ; Wenbo LI ; Qun LI ; Zengyi MA ; Wenming WU ; Minghui WANG ; Jing WANG ; Feifei FAN ; Kuilin SUN ; Xiaofeng LIU
China Journal of Endoscopy 2025;31(4):80-85
Objective To investigate the efficacy and safety of self-help inflatable balloon in the treatment of refractory benign upper digestive tract stricture.Methods From November 2022 to February 2024,10 patients with refractory benign upper digestive tract stricture underwent self-help inflatable balloon were retrospectively analyzed.Preoperative treatment,intraoperative and postoperative complications,dysphagia score before and after surgery,wearing time of external self-help inflatable balloon,and the clinical efficacy were evaluated.Results All the 10 patients received self-help inflatable balloon treatment.No complications such as bleeding and perforation occurred in all patients during the wearing of the balloon.The placement time of the balloon was 60~180 d,and the average wearing time was 104.9 d.Before surgery,nine cases had a stricture segment length less than 2 cm,and one case had a stricture segment length of 6 cm.The dysphagia score before surgery was(3.60±0.52),and the dysphagia score after removal of the self-help inflatable balloon was(0.60±0.52),and there was statistical significance in preoperative and postoperative comparison(P<0.05).The dysphagia score 3 months after removal of the external self-help inflatable balloon was(0.70±0.48),there was no significant difference in dysphagia score between the day after surgery and 3 months after surgery(P>0.05).The diameter of the stricture before and after operation was(5.09±2.02)mm and(10.35±0.73)mm respectively,and the difference was statistically significant(P<0.05).Sore throat,wing of nose pain and balloon displacement were the most common adverse events,with 30.0%(3/10),50.0%(5/10)and 70.0%(7/10),respectively.Conclusion It is safe,effective and feasible to treat refractory benign upper digestive tract stricture with self-help inflatable balloon in vitro.It is worthy for clinical application.

Result Analysis
Print
Save
E-mail