Objective:To explore the diagnostic value of different machine learning models and optimal machine learning model combined with clinical data for diagnosing Hashimoto′s thyroiditis (HT).Methods:The thyroid gland images of 643 patients with 643 thyroid nodules who underwent preoperative ultrasound examination and had pathological results in the Second Affiliated Hospital of Xi′an Jiaotong University from December 2018 to March 2024 were retrospectively collected, and the images were divided into training set and test set according to a ratio of 7 to 3. Twenty ultrasound imaging omics models were constructed using pairwise combination of 5 feature screening components and 4 classifiers. The area under the curve (AUC) of each model in the test set was compared. Meanwhile, 3 basic network models were respectively used to construct deep learning models for diagnosing HT, and the diagnostic efficacies of the deep learning models and the ultrasound imaging omics models for HT were compared. The model with the greatest efficacy was selected as the optimal machine learning model. Further, the optimal machine learning model was combined with clinical data to construct a combined model. The ROC curves were plotted to compare the diagnostic efficacy of the optimal machine learning model and the combined model for HT.Results:In the comparison of the efficacies of ultrasound imaging omics models and deep learning models in diagnosing HT, the efficacy of stable feature screening-logistic regression (LR) model was the greatest, and the accuracy, sensitivity and specificity of using the LR model in diagnosing HT in the test set were 78%, 75%, 74%, respectively, with an AUC of 0.82(95% CI=0.76-0.88). After combining the LR model with clinical data, the accuracy, sensitivity, and specificity of the combined model in the test set were 87%, 74%, and 95%, respectively, with an AUC of 0.91(95% CI=0.87-0.95), which was strongly consistent with pathology (Kappa value=0.708, P<0.001). Conclusions:The optimal machine learning model (LR model) constructed in this study demonstrates a strong ability to diagnose HT and can accurately detect patients with atypical ultrasound manifestations of HT. The combination with clinical data can improve its diagnostic efficacy with higher accuracy and specificity.

Objective:To explore the diagnostic value of different machine learning models and optimal machine learning model combined with clinical data for diagnosing Hashimoto′s thyroiditis (HT).Methods:The thyroid gland images of 643 patients with 643 thyroid nodules who underwent preoperative ultrasound examination and had pathological results in the Second Affiliated Hospital of Xi′an Jiaotong University from December 2018 to March 2024 were retrospectively collected, and the images were divided into training set and test set according to a ratio of 7 to 3. Twenty ultrasound imaging omics models were constructed using pairwise combination of 5 feature screening components and 4 classifiers. The area under the curve (AUC) of each model in the test set was compared. Meanwhile, 3 basic network models were respectively used to construct deep learning models for diagnosing HT, and the diagnostic efficacies of the deep learning models and the ultrasound imaging omics models for HT were compared. The model with the greatest efficacy was selected as the optimal machine learning model. Further, the optimal machine learning model was combined with clinical data to construct a combined model. The ROC curves were plotted to compare the diagnostic efficacy of the optimal machine learning model and the combined model for HT.Results:In the comparison of the efficacies of ultrasound imaging omics models and deep learning models in diagnosing HT, the efficacy of stable feature screening-logistic regression (LR) model was the greatest, and the accuracy, sensitivity and specificity of using the LR model in diagnosing HT in the test set were 78%, 75%, 74%, respectively, with an AUC of 0.82(95% CI=0.76-0.88). After combining the LR model with clinical data, the accuracy, sensitivity, and specificity of the combined model in the test set were 87%, 74%, and 95%, respectively, with an AUC of 0.91(95% CI=0.87-0.95), which was strongly consistent with pathology (Kappa value=0.708, P<0.001). Conclusions:The optimal machine learning model (LR model) constructed in this study demonstrates a strong ability to diagnose HT and can accurately detect patients with atypical ultrasound manifestations of HT. The combination with clinical data can improve its diagnostic efficacy with higher accuracy and specificity.

Objective:To summarize the methods and results of human motion intention recognition based on the surface electromyography. Methods:Literatures were retrieved and reviewed from the databases of PubMed, Web of Science, CNKI, Wanfang and VIP until December, 2020. The experimental researches about human motion intention recognition based on surface electromyography were summarized. Results:The methods of motion intention recognition were divided into three models: musculoskeletal model, traditional machine learning model and deep learning model. Conclusion:It is difficult to fully estimate human motion intention using surface electromyography in a single way. More researches are needed to develop more accurate and real-time human motion intention recognition methods.

Objective To understand the disease spectrum of prehospital emergency patients in Beijing, so as to provide basis for rational allocation of prehospital emergency care resources and for improvement of prehospital emergency care. Methods Emergency dispatch database from 120 dispatching command system of Beijing Emergency Medical Center to was used study the disease spectrum of 120 prehospital emergency patients from 2013 to 2017 in Beijing, and were analyzed the differences of disease spectrum among patients of different years, genders and ages, and the time distribution of different diseases. Results From 2013 to 2017 in Beijing, 1 643 375 patients used 120 prehospital emergency care. The top 3 diseases of the disease spectrum were: injury and poisoning (29.63%), diseases of the circulatory system (18.37%), and diseases of the nervous system (15.80%). The sorting of diseases remained unchanged during the 5 years. The disease spectrums were different among different age groups and gender groups. The calls in spring, summer, autumn and winter accounted for 25.23%, 24.97%, 25.05% and 24.75% respectively. Most calls were made from 08 ∶〗00 to 09 ∶〗59 (12.41%), 12 ∶〗00 to 13 ∶〗59 (10.05%), and 14 ∶〗00 to 15 ∶〗59 (9.91%) during a day. The peak months and hours of different diseases were different. Conclusions The demand for prehospital emergency care has been increasing in Beijing in recent years. It is important to strengthen the prevention of injuries among all ages, especially male, and to improve the prevention and response capacity for acute attack of chronic diseases among older people. Prehospital emergency care resources should be rationally allocated according to the peak months and hours of calls.

Adult ; Animals ; Female ; Frameshift Mutation/genetics* ; Genes, sry ; Gonadal Dysgenesis, 46,XY/genetics* ; Humans ; Menarche ; Ovary/pathology* ; Oviducts/pathology* ; Sex-Determining Region Y Protein/genetics*

Objective To explore the relationship between blood lipid levels and cerebrovascular hemodynamic indices (CVHI) in a young and middle?aged population, and to identify reference values for early warning of stroke. Method Between June 2015 and June 2016, a total of 6 252 young and middle?aged patients were assessed for cerebrovascular function in The Health Management Center of Affiliated Hospital of Guizhou Medical University. Patients were divided into abnormal and normal blood lipid groups. Differences in CVHI indicators and scores between the groups were determined, and factors influencing CVHI indices were compared using multivariate logistic regression analysis. Results Mean (Vmean), maximum (Vmax), and minimum blood flow velocity (Vmin) and cumulative scores for CVHI indices in the abnormal blood lipid group were significantly lower than in the normal blood lipid group (P<0.001). However, pulse wave velocity, characteristic impedance, peripheral vascular resistance, dynamic resistance, critical pressure, and differences between diastolic blood pressure and critical pressure were greater than those in the normal blood lipid group (P<0.001). Multivariate logistic regression analysis indicated that a high triglyceride level, overweight or obesity, high systolic blood pressure, high diastolic blood pressure, and age were risk factors for low CVHI scores, with odds ratios (95% confidence intervals) of 1.455 (1.195- 1.771), 2.271 (1.782-2.895), 5.967 (4.557-7.815), 4.251 (3.349-5.396), 2.560 (1.993-3.287), and 1.448 (1.189-1.763). Conclusion Abnormal blood lipid levels can lead to abnormal cerebrovascular function in young and middle?aged population; elevated triglyceride level was an independent risk factor for impaired cerebrovascular function.

Objective:By applying moxibustion to the eight confluent points in different periods of time,to observe the changes in thermal pain threshold latency of acupoints based on Fei Teng Ba Fa.Methods:A total of 468 healthy college student volunteers received moxibustion at the eight confluent points in three different periods of time,i.e.Chen (7:00-9:00),Wu (11:00-13:00) and Xu (19:00-21:00).The thermal pain threshold latency was adopted to measure the changes in pain threshold of the eight confluent points under different conditions (different periods of time,different genders,different acupoints and different states of the acupoints) based on Fei Teng Ba Fa.Results:Finally,thirty subjects dropped out and 438 subjects were included.The comparison of thermal pain threshold latencies of the eight confluent points in the same opening or closing state based on Fei Teng Ba Fa:latencies of the closing points and adjunct points were significantly different in different periods of time (P＜0.05);the latencies of the males were significantly longer than those of the females (P＜0.05);there was no significant difference in the latency between the left and right sides (P＞0.05);in the female group,there was a significant difference in the latency between the lower-limb points and the upper-limb points (P＜0.05).The comparison of thermal point threshold latencies of the eight confluent points in different opening or closing state:in the period of Wu (11:00-13:00),the latencies of the opening points were significantly longer than those of the closing points and adjunct points (P＜0.05);for men,their opening and closing points had significantly longer thermal pain threshold latencies than their adjunct points (P＜0.05);despite the gender,the latencies of the upper limb opening and closing points were significantly longer than the latency of the adjunct points (P＜0.05);in the female group,the latencies of the lower-limb opening points were significantly shorter than those of the lower-limb closing and adjunct points (P＜0.05).Conclusion:Based on Fei Teng Ba Fa,the pain thresholds of the eight confluent points vary in different periods of time,gender,acupoint location and opening/closing state,which can be taken as the evidence of making time-based acupuncture-moxibustion prescriptions.

Thoracic obstruction is mainly attributed to the scope of coronary heart disease in modern medicine, and traditional Chinese medicine(TCM) shows a significant effect in the treatment of thoracic obstruction. In this research, a network pharmacology method was carried out to systemically study the underlying mechanism of the core herbal compatibility in TCM on the thoracic obstruction. First, we collected the literature about TCM prescriptions for treating thoracic obstruction from CNKI. Then, a prescription database was establish by TCM inheritance support platform system(V2.5) to determine the medication rules and core herbal compatibility in TCM. Finally, to obtain the potential signaling pathways, KEGG pathway analysis was performed by BATMAN-TCM online analysis tool. Results showed that the potential signal pathway of core herbal compatibility in TCM for the clinical treatment of thoracic obstruction was calcium ion and cGMP-PKG signaling pathway. This study provided a new research strategy for the study of the medication rules and mechanism of traditional Chinese medicine in the treatment of thoracic obstruction.

Objective To study the awareness of perimenopausal hormone replacement therapy (HRT) among a part of the medical care personnel in Guiyang . Methods A survey was conducted among 500 medical staff members in 4 hospitals of Guiyang by cluster random sampling using questionnaire about HRT. Results The survey showed that 74.6% (373/500) medical staff thought that the hormone replacement therapy was necessary to perimenopausal women; 96.7% (87/90) of obstetrics and gynecology doctors believed that it was necessary for perimenopausal women to use HRT,which was significantly higher than the doctors of other specialties 68.6% (166/242) and the nurses group 71.4% (120/168) (χ2=28.509, 23.537, P<0.01). Only 5.8%(29/500) of the medical personnel were willing to recommend HRT. In light of the attitude for recommending HRT, the obstetricians and gynecologists group was more significantly higher than the other specialties doctors group (χ2=86.781, P<0.01). Conclusion The knowledge of hormone replacement therapy in part of Guiyang medical personnel is not sufficient;the recommending rate of HRT was low;the side effects of HRT was still a concern. There are differences between obstetrics and gynecology doctors and doctors other specialties and nurses in HRT knowledge.

BACKGROUNDThis study aimed to evaluate the effects of standard rescue procedure (SRP) in improving severe trauma treatments in China.

METHODSThis study was conducted in 12 hospitals located in geographically and industrially different cities in China. A standard procedure on severe trauma rescue was established as a general rule for staff training and patient treatment. A regional network (system) efficiently integrating prehospital rescue, emergency room treatments, and hospital specialist treatments was built under the rule for information sharing and improving severe trauma treatments. Treatment outcomes were compared between before and 1 year after the implementation of the SRP.

RESULTSThe outcomes of a total of 74,615 and 12,051 trauma cases were collected from 12 hospitals before and after the implementation of the SRP. Implementation of the SRP led to efficient cooperation and information sharing of different treatment services. The emergency response time, prehospital transit time, emergency rescue time, consultation call time, and mortality rate of patients were 24.24 ± 4.32 min, 45.69 ± 3.89 min, 6.38 ± 1.05 min, 17.53 ± 0.72 min, and 33.82% ± 3.87% (n = 441), respectively, before the implementation of the standardization and significantly reduced to 10.11 ± 3.21 min, 22.39 ± 4.32 min, 3.26 ± 0.89 min, 3.45 ± 0.45 min, and 20.49% ± 3.11%, separately (n = 495, P < 0.05) after that.

CONCLUSIONSStaff training and SRP can significantly improve the efficiency of severe trauma treatments in China.

Adolescent ; Adult ; Aged ; Child ; Child, Preschool ; China ; Emergency Medical Services ; standards ; Female ; Humans ; Infant ; Infant, Newborn ; Male ; Middle Aged ; Wounds and Injuries ; Young Adult