1.Comparison of small-sample multi-class machine learning models for plasma concentration prediction of valproic acid
Xi CHEN ; Shen’ao YUAN ; Hailing YUAN ; Jie ZHAO ; Peng CHEN ; Chunyan TIAN ; Yi SU ; Yunsong ZHANG ; Yu ZHANG
China Pharmacy 2025;36(11):1399-1404
		                        		
		                        			
		                        			OBJECTIVE To construct three-class (insufficient, normal, excessive) and two-class (insufficient, normal) models for predicting plasma concentration of valproic acid (VPA), and compare the performance of these two models, with the aim of providing a reference for formulating clinical medication strategies. METHODS The clinical data of 480 patients who received VPA treatment and underwent blood concentration test at the Xi’an International Medical Center Hospital were collected from November 2022 to September 2024 (a total of 695 sets of data). In this study, predictive models were constructed for target variables of three-class and two-class models. Feature ranking and selection were carried out using XGBoost scores. Twelve different machine learning algorithms were used for training and validation, and the performance of the models was evaluated using three indexes: accuracy, F1 score, and the area under the working characteristic curve of the subject (AUC). RESULTS XGBoost feature importance scores revealed that in the three-class model, the importance ranking of kidney disease and electrolyte disorders was higher. However, in the two-class model, the importance ranking of these features significantly decreased, suggesting a close association with the excessive blood concentration of VPA. In the three-class model, Random Forest method performed best, with F1 score of 0.704 0 and AUC of 0.519 3 on the test set; while in the two-class model, CatBoost method performed optimally, with F1 score of 0.785 7 and AUC of 0.819 5 on the test set. CONCLUSIONS The constructed three-class model has the ability to predict excessive VPA blood concentration, but its prediction and model generalization abilities are poor; the constructed two-class model can only perform classification prediction for insufficient and normal blood concentration cases, but its model performance is stronger.
		                        		
		                        		
		                        		
		                        	
2.Noninvasive Diagnostic Technique for Nonalcoholic Fatty Liver Disease Based on Features of Tongue Images.
Rong-Rui WANG ; Jia-Liang CHEN ; Shao-Jie DUAN ; Ying-Xi LU ; Ping CHEN ; Yuan-Chen ZHOU ; Shu-Kun YAO
Chinese journal of integrative medicine 2024;30(3):203-212
		                        		
		                        			OBJECTIVE:
		                        			To investigate a new noninvasive diagnostic model for nonalcoholic fatty liver disease (NAFLD) based on features of tongue images.
		                        		
		                        			METHODS:
		                        			Healthy controls and volunteers confirmed to have NAFLD by liver ultrasound were recruited from China-Japan Friendship Hospital between September 2018 and May 2019, then the anthropometric indexes and sampled tongue images were measured. The tongue images were labeled by features, based on a brief protocol, without knowing any other clinical data, after a series of corrections and data cleaning. The algorithm was trained on images using labels and several anthropometric indexes for inputs, utilizing machine learning technology. Finally, a logistic regression algorithm and a decision tree model were constructed as 2 diagnostic models for NAFLD.
		                        		
		                        			RESULTS:
		                        			A total of 720 subjects were enrolled in this study, including 432 patients with NAFLD and 288 healthy volunteers. Of them, 482 were randomly allocated into the training set and 238 into the validation set. The diagnostic model based on logistic regression exhibited excellent performance: in validation set, it achieved an accuracy of 86.98%, sensitivity of 91.43%, and specificity of 80.61%; with an area under the curve (AUC) of 0.93 [95% confidence interval (CI) 0.68-0.98]. The decision tree model achieved an accuracy of 81.09%, sensitivity of 91.43%, and specificity of 66.33%; with an AUC of 0.89 (95% CI 0.66-0.92) in validation set.
		                        		
		                        			CONCLUSIONS
		                        			The features of tongue images were associated with NAFLD. Both the 2 diagnostic models, which would be convenient, noninvasive, lightweight, rapid, and inexpensive technical references for early screening, can accurately distinguish NAFLD and are worth further study.
		                        		
		                        		
		                        		
		                        			Humans
		                        			;
		                        		
		                        			Non-alcoholic Fatty Liver Disease/diagnostic imaging*
		                        			;
		                        		
		                        			Ultrasonography
		                        			;
		                        		
		                        			Anthropometry
		                        			;
		                        		
		                        			Algorithms
		                        			;
		                        		
		                        			China
		                        			
		                        		
		                        	
3.Effect analysis of combined Tiaojingzhuyun mixture and regulating menstration and promoting pregnancy by thirteen acupuncture in treating infertility caused by Follicle dysplasia
Xi CHEN ; Xinni WANG ; Yuan QI ; Shuxiang HU ; Jin LI
The Journal of Practical Medicine 2024;40(3):412-416
		                        		
		                        			
		                        			Objective To explore the clinical effects of Tiaojing Zhuyun mixture combined with regulating menstration and promoting pregnancy by thirteen acupuncture in the treatment of Declining Ovarian Function(DOR)combined with follicular dysplasia infertility.Methods From January 1,2022 to January 1,2023,60 cases of infertility patients with DOR follicular dysplasia were treated in our hospital,and the above cases were randomly divided into experimental group:Tiaojing Zhuyun mixture+ regulating menstration and promoting pregnancy by thirteen acupuncture + Clomiphene group;Control group:Clomiphene group,30 cases each.The improvement of FSH,maximum follicle size and ovulation,uterine artery blood flow index,endometrial thickness and pregnancy were observed after treatment.Results The improvement of follicle stimulating hormone,maximum follicle size,endometrial thickness and uterine artery blood flow indexes in Tiaojing Zhuyun mixture+ regulating menstration and promoting pregnancy by thirteen acupuncture + Clomiphene group were better than clomiphene group.The difference between groups was statistically significant(P<0.05).The ovulation rate was 66.7%and the pregnancy rate was 40%.The ovulation rate was 40%higher than the control group,and the pregnancy rate was 20%.The difference between groups was statistically significant(P<0.05).Conclusion In the patients with DOR follicular dysplasia infertility,Tiaojing Zhuyun mixture combined with regulating menstration and promoting pregnancy by thirteen acupuncture can improve the level of follicle stimulating hormone,promote follicular maturation and ovula-tion,reduce the resistance of uterine artery blood flow,increase the thickness of endometrial,promote pregnancy,and have fewer adverse reactions.
		                        		
		                        		
		                        		
		                        	
4.OCT analysis of in-stent neointima over 5 years post-DES implantation
Yan HAN ; Xiaohang YUAN ; Mengting JIANG ; Huanhuan FENG ; Xi ZHANG ; Yingqian ZHANG ; Jing JING ; Yundai CHEN ; Lei GAO
Chinese Journal of Cardiology 2024;52(2):150-157
		                        		
		                        			
		                        			Objective:To investigate the characteristics of neointimal hyperplasia (NIH) in patients with in-stent restenosis (ISR) over 5 years post-drug-eluting stent (DES) implantation based on optical coherence tomography (OCT).Methods:In this cross-sectional study, patients with DES-ISR who underwent OCT examination at PLA General Hospital between March 2010 and March 2022 were retrospectively included. All patients were divided into≤5 years DES-ISR group and>5 years DES-ISR group according to the time interval after DES implantation. Quantitative and qualitative analyses were conducted on OCT images to compare the clinical data and lesion characteristics of two patient groups. Furthermore, the independent clinical predictive factors of in-stent neoatherosclerosis (ISNA) were analyzed by multivariable logistic regression.Results:A total of 230 DES-ISR patients with 249 lesions were included, with an age of (63.1±10.4) years and 188 males (81.7%). The median interval after DES implantation was 6 (2, 9) years. There were 117 patients (122 ISR lesions) in the≤5 years DES-ISR group, and 113 patients (127 ISR lesions) in the>5 years DES-ISR group. Compared with≤5 years DES-ISR,>5 years DES-ISR showed more heterogeneous patterns (65.4% (83/127) vs. 48.4% (59/122), P=0.007), diffuse patterns (46.5% (59/127) vs. 31.2% (38/122), P=0.013), macrophage accumulations (44.1% (56/127) vs. 31.2% (38/122), P=0.035) in NIH and higher prevalence of ISNA (83.5% (106/127) vs. 72.1% (88/122), P=0.031). According to multivariable logistic regression, the independent predictive factor for ISNA was female ( OR=0.44, 95% CI 0.21-0.90, P=0.026). Female ( OR=0.48, 95% CI 0.23-0.99, P=0.046) and low-density lipoprotein cholesterol level ( OR=1.62, 95% CI 1.01-2.59, P=0.046) were independent predictive factors, respectively, for lipid ISNA. Calcified ISNA was independently associated with time interval of post-DES implantation ( OR=1.18, 95% CI 1.07-1.29, P=0.001). Conclusion:DES-ISR patients with a time interval of>5 years after stent implantation have a higher prevalence of ISNA and more complex lesions. Gender, the level of low-density lipoprotein cholesterol, and the time interval post-DES implantation are independently correlated with ISNA, lipid ISNA, and calcified ISNA.
		                        		
		                        		
		                        		
		                        	
5.Dynamic monitoring and esthetic evaluation of dimensional changes in the peri-implant soft tissue contour after the immediate implant placement and provisionalization with the modified socket-shield technique in the esthetic zone
Xiaofeng TANG ; Qingqing WU ; Xi CHEN ; Yuan WANG ; Gang FU
Chinese Journal of Stomatology 2024;59(6):551-558
		                        		
		                        			
		                        			Objective:To accurately measure the dynamic changes of peri-implant soft tissue within one year after the immediate implant placement and provisionalization with the modified socket-shield technique (MSST) in the esthetic zone, and to provide a basis for evaluating the effect of the modified socket-shield technique on the maintenance of peri-implant soft tissue.Methods:A total of 22 patients (22 implants) were prospectively included 1 year after completion of immediate implant placement and provisionalization (IIPP) within MSST in the esthetic zone from January 2022 to January 2024 at the Department of Oral Implantology in the Stomatological Hospital of Chongqing Medical University. The intraoral optical models of patients were obtained by an intraoral scanner system preoperatively and at 3, 6, and 12 months postoperatively, respectively. The standard tessellation language files of intraoral optical models at multiple time points were imported to Geomagic Studio 2013 to be superimposed and aligned for analyzing the peri-implant soft tissue contour on the labial side of the implant site at multiple levels. The amount of gingival margin recession, gingival papilla change, and thickness change of the labial side of the soft tissues at each postoperative point in time were measured at each postoperative time point, as well as evaluating the esthetic effect by the pink esthetic score (PES).Results:The patients were (40±13) years old (21-75 years), including 9 males and 13 females. No adverse events occurred in all the implants during the 12-month follow-up period. The recession level of the gingival margin of the implant site (GL) was 0.08 (0.07) mm, the recession level of the mesial papilla (ML) was 0.19 (0.25) mm, and the recession level of the distal papilla (DL) was 0.19 (0.10) mm. The average collapse thickness of the soft tissue contour on the labial side of the implant (ΔD) was (0.39±0.09) mm, mainly occurring within 2 mm of the root of the gingival margin. The height of the alveolar bone was reduced by (0.17±0.08) mm. The thickness of the labial alveolar bone at 1, 3, and 5 mm root side of the implant shoulder was reduced by (0.13±0.08), (0.12±0.10) and 0.04 (0.17) mm, respectively. The postoperative pink esthetic score was 13.00 (2.25) points at 12 months, which suggested that all implant sites achieved ideal esthetic results.Conclusions:The labial soft tissue contour at implant sites shows minimal change following immediate implant placement and provisionalization using the modified socket-shield technique for 1 year in the esthetic zone.
		                        		
		                        		
		                        		
		                        	
6.TCM Guidelines for Diagnosis and Treatment of Chronic Cough in Children
Xi MING ; Liqun WU ; Ziwei WANG ; Bo WANG ; Jialin ZHENG ; Jingwei HUO ; Mei HAN ; Xiaochun FENG ; Baoqing ZHANG ; Xia ZHAO ; Mengqing WANG ; Zheng XUE ; Ke CHANG ; Youpeng WANG ; Yanhong QIN ; Bin YUAN ; Hua CHEN ; Lining WANG ; Xianqing REN ; Hua XU ; Liping SUN ; Zhenqi WU ; Yun ZHAO ; Xinmin LI ; Min LI ; Jian CHEN ; Junhong WANG ; Yonghong JIANG ; Yongbin YAN ; Hengmiao GAO ; Hongmin FU ; Yongkun HUANG ; Jinghui YANG ; Zhu CHEN ; Lei XIONG
Journal of Nanjing University of Traditional Chinese Medicine 2024;40(7):722-732
		                        		
		                        			
		                        			Following the principles of evidence-based medicine,in accordance with the structure and drafting rules of standardized documents,based on literature research,according to the characteristics of chronic cough in children and issues that need to form a consensus,the TCM Guidelines for Diagnosis and Treatment of Chronic Cough in Children was formulated based on the Delphi method,expert discussion meetings,and public solicitation of opinions.The guideline includes scope of application,terms and definitions,eti-ology and diagnosis,auxiliary examination,treatment,prevention and care.The aim is to clarify the optimal treatment plan of Chinese medicine in the diagnosis and treatment of this disease,and to provide guidance for improving the clinical diagnosis and treatment of chronic cough in children with Chinese medicine.
		                        		
		                        		
		                        		
		                        	
7.Analysis of cardiovascular disease prevention indicators among residents with intra-urban migration in Central China
HUANG Tianshu ; TIAN Yuan ; ZHANG Xingyi ; LI Chenhui ; ZHAO Yun ; ZHAO Dongyuan ; CHEN Xianhua ; ZHU Mengyao ; JIAO Guanqi ; GUO Dongmin ; LI Xi ; CUI Jianlan
Journal of Preventive Medicine 2024;36(5):451-456
		                        		
		                        			Objective:
		                        			To investigate cardiovascular disease (CVD) prevention status among residents with intra-urban migration in Central China, so as to provide insights into targeted prevention and control of CVD.
		                        		
		                        			Methods:
		                        			Basic data of residents aged 35 to 75 years who participated in Early Screening and Comprehensive Intervention Project for CVD high-risk populations in Central China from September 2015 to August 2020 were collected. According to birth place, type of registered residence and current residence, residents were divided into four groups: local residents in old urban area, local residents in new urban area, other urban migrants and other rural migrants. The status of CVD primary and secondary prevention, were analysed by using a robust Poisson regression model.
		                        		
		                        			Results:
		                        			A total of 76 513 residents were recruited, including 29 420 males (38.45%) and 47 093 females (61.55%), and had a mean age of (56.36±9.84) years. There were 45 087 (58.93%) local residents in old urban area, 23 868 (31.19%) local residents in new urban area, 5 668 (7.41%) other urban migrants and 1 890 (2.47%) other rural migrants. After adjusting for variables such as age, gender and educational level, the results of robust Poisson regression analysis showed that compared with local residents in old urban area, local residents in new urban area had lower compliance rates of non- or moderate-drinking (RR=0.987, 95%CI: 0.975-1.000) and healthy diet (RR=0.535, 95%CI: 0.365-0.782), lower proportion of using aspirin as primary prevention in CVD high-risk population (RR=0.616, 95%CI: 0.511-0.741), lower awareness (RR=0.873, 95%CI: 0.782-0.974) and control rates (RR=0.730, 95%CI: 0.627-0.849) of hypertension; other urban migrants had higher compliance rate of non-smoking (RR=1.045, 95%CI: 1.017-1.075); other rural migrants had lower proportion of using aspirin as primary prevention in CVD high-risk population (RR=0.826, 95%CI: 0.707-0.966).
		                        		
		                        			Conclusion
		                        			The CVD primaryprevention among local residents in new urban area is relatively poor among four groups of residents in Central China, and key interventions are needed.
		                        		
		                        		
		                        		
		                        	
8.Analysis of epidemiological characteristics of risk factors for cardiovascular diseases and malignant tumors based on the Shanghai community elderly cohort
Ping LI ; Huiru JIANG ; Mengyue YE ; Yayu WANG ; Xiaoyu CHEN ; Ancai YUAN ; Wenjie XU ; Huimin DAI ; Xi CHEN ; Xiaoxiang YAN ; Shengxian TU ; Yuanqi ZHENG ; Wei ZHANG ; Jun PU
Journal of Shanghai Jiaotong University(Medical Science) 2024;44(5):617-625
		                        		
		                        			
		                        			Objective·To analyze the epidemiological characteristics of risk factors for cardiovascular diseases and malignant tumors based on the Shanghai community elderly cohort.Methods·The study subjects were selected from the Shanghai community elderly cohort established from February to August 2019,with a total of 17 948 people.The study subjects were divided into 4 groups according to self-reported presence or absence of tumors and/or cardiovascular diseases during the baseline survey:tumor-free and non-cardiovascular disease group,single cardiovascular disease group,single tumor group and tumor cardiovascular disease co-occurrence group.The differences among the four groups of subjects were collected and compared in terms of demographic characteristics and physiological indicators,daily living habits(smoking,drinking tea,drinking coffee,drinking carbonated drink,drinking alcohol,sedentary time,physical activity level and sleep quality),past medical history,psychological status(depression and anxiety)and dietary compliance.Results·Among the study subjects,60.1%of tumor patients were complicated with cardiovascular diseases.The differences among the four groups of subjects in age,gender,educational level,pre-retirement occupation,waist circumference,hip circumference and body mass index were statistically significant(all P<0.05).Compared with the tumor-free and non-cardiovascular disease group,the single cardiovascular disease group,single tumor group and tumor cardiovascular disease co-occurrence group all exhibited lower proportions of smoking and high physical activity levels(all P<0.05),and higher proportion of sedentary time exceeding 4 h/d and poor sleep quality(all P<0.05);the proportion of subjects with past medical histories including hyperlipidemia,peripheral vascular disease,endocrine system disease,respiratory system disease,urinary system disease and digestive system disease of the single cardiovascular disease group and the tumor cardiovascular disease co-occurrence group was higher(all P<0.05),and the proportion of subjects with depression and anxiety was also higher(all P<0.05).Furthermore,compared with the tumor-free and non-cardiovascular disease group,the single cardiovascular disease group had lower compliance rates of poultry,fish,fruit and liquid milk(all P<0.05).Among the four groups,only the compliance rate of vegetable intake exceeded 50%,while the compliance rates of poultry,fish,fruit,liquid milk and tubers were all below 20%.Conclusion·In the elderly population of Shanghai communities,over half of malignant tumor patients are concomitant with cardiovascular diseases.Unhealthy daily habits are prevalent among those with cardiovascular diseases,tumors and tumor-cardiovascular disease co-occurrence.The intake of many foods in the elderly of the community do not reach the levels recommended by Chinese Dietary Guidelines.
		                        		
		                        		
		                        		
		                        	
9.Risk factors and predictive model of cerebral edema after road traffic accidents-related traumatic brain injury
Di-You CHEN ; Peng-Fei WU ; Xi-Yan ZHU ; Wen-Bing ZHAO ; Shi-Feng SHAO ; Jing-Ru XIE ; Dan-Feng YUAN ; Liang ZHANG ; Kui LI ; Shu-Nan WANG ; Hui ZHAO
Chinese Journal of Traumatology 2024;27(3):153-162
		                        		
		                        			
		                        			Purpose::Cerebral edema (CE) is the main secondary injury following traumatic brain injury (TBI) caused by road traffic accidents (RTAs). It is challenging to be predicted timely. In this study, we aimed to develop a prediction model for CE by identifying its risk factors and comparing the timing of edema occurrence in TBI patients with varying levels of injuries.Methods::This case-control study included 218 patients with TBI caused by RTAs. The cohort was divided into CE and non-CE groups, according to CT results within 7 days. Demographic data, imaging data, and clinical data were collected and analyzed. Quantitative variables that follow normal distribution were presented as mean ± standard deviation, those that do not follow normal distribution were presented as median (Q 1, Q 3). Categorical variables were expressed as percentages. The Chi-square test and logistic regression analysis were used to identify risk factors for CE. Logistic curve fitting was performed to predict the time to secondary CE in TBI patients with different levels of injuries. The efficacy of the model was evaluated using the receiver operator characteristic curve. Results::According to the study, almost half (47.3%) of the patients were found to have CE. The risk factors associated with CE were bilateral frontal lobe contusion, unilateral frontal lobe contusion, cerebral contusion, subarachnoid hemorrhage, and abbreviated injury scale (AIS). The odds ratio values for these factors were 7.27 (95% confidence interval ( CI): 2.08 -25.42, p = 0.002), 2.85 (95% CI: 1.11 -7.31, p = 0.030), 2.62 (95% CI: 1.12 -6.13, p = 0.027), 2.44 (95% CI: 1.25 -4.76, p = 0.009), and 1.5 (95% CI: 1.10 -2.04, p = 0.009), respectively. We also observed that patients with mild/moderate TBI (AIS ≤ 3) had a 50% probability of developing CE 19.7 h after injury (χ 2= 13.82, adjusted R2 = 0.51), while patients with severe TBI (AIS > 3) developed CE after 12.5 h (χ 2= 18.48, adjusted R2 = 0.54). Finally, we conducted a receiver operator characteristic curve analysis of CE time, which showed an area under the curve of 0.744 and 0.672 for severe and mild/moderate TBI, respectively. Conclusion::Our study found that the onset of CE in individuals with TBI resulting from RTAs was correlated with the severity of the injury. Specifically, those with more severe injuries experienced an earlier onset of CE. These findings suggest that there is a critical time window for clinical intervention in cases of CE secondary to TBI.
		                        		
		                        		
		                        		
		                        	
10.Expert consensus on the evaluation and rehabilitation management of shoulder syndrome after neek dissection for oral and maxillofacial malignancies
Jiacun LI ; Moyi SUN ; Jiaojie REN ; Wei GUO ; Longjiang LI ; Zhangui TANG ; Guoxin REN ; Zhijun SUN ; Jian MENG ; Wei SHANG ; Shaoyan LIU ; Jie ZHANG ; Jicheng LI ; Yue HE ; Chunjie LI ; Kai YANG ; Zhongcheng GONG ; Qing XI ; Bing HAN ; Huaming MAI ; Yanping CHEN ; Jie ZHANG ; Yadong WU ; Chao LI ; Changming AN ; Chuanzheng SUN ; Hua YUAN ; Fan YANG ; Haiguang YUAN ; Dandong WU ; Shuai FAN ; Fei LI ; Chao XU ; Wei WEI
Journal of Practical Stomatology 2024;40(5):597-607
		                        		
		                        			
		                        			Neck dissection(ND)is one of the main treatment methods for oral and maxillofacial malignancies.Although ND type is in con-stant improvement,but intraoperative peal-pull-push injury of the accessory nerve,muscle,muscle membrane,fascia and ligament induced shoulder syndrome(SS)is still a common postoperative complication,combined with the influence of radiochemotherapy,not only can cause pain,stiffness,numbness,limited dysfunction of shoulder neck and arm,but also may have serious impact on patient's life quality and phys-ical and mental health.At present,there is still a lack of a systematic evaluation and rehabilitation management program for postoperative SS of oral and maxillofacial malignant tumors.Based on the previous clinical practice and the current available evidence,refer to the relevant lit-erature at home and abroad,the experts in the field of maxillofacial tumor surgery and rehabilitation were invited to discuss,modify and reach a consenusus on the etiology,assessment diagnosis,differential diagnosis,rehabilitation strategy and prevention of SS,in order to provide clinical reference.
		                        		
		                        		
		                        		
		                        	
            

Result Analysis
Print
Save
E-mail