Epidemiological characteristics of respiratory disease mortality in residents of Baoshan，Shanghai 2009‒2020 / 上海预防医学

ObjectiveTo investigate the epidemiological characteristics of respiratory disease mortality in Baoshan residents during the period of 2009‒2020. MethodsRespiratory disease deaths of Baoshan residents from 2009‒2020 were collected. ICD-10 codes were used to classify the causes of death， and R-4.2.1 was applied for statistical analysis. The average annual percent change （AAPC） of standardized mortality rates of different respiratory diseases were analyzed by using Joinpoint 4.9.0.0. ResultsThe average annual mortality rate of respiratory diseases in Baoshan from 2009 to 2020 was 58.86/105， and the standardized mortality rate was 35.62/105， which was the 3rd leading cause of mortality. The mortality rate of respiratory diseases was higher in men than in women （χ2=46.70， P<0.001）. COPD ranked first among respiratory diseases in Baoshan from 2009 to 2020， followed by pneumonia， asthma and pneumoconiosis in that order. The standardized mortality rate for COPD decreased from 38.66/105 in 2009 to 19.88/105 in 2020 （AAPC=-6.6%， 95%CI： -8.2% to -4.9%， P<0.001）. The standardized mortality rate of asthma decreased from 2.86/105 in 2009 to 1.43/105 in 2020 （AAPC=-5.8%， 95%CI： -8.8% to -2.8%， P<0.01）. The standardized mortality rate of pneumoconiosis decreased from 0.64/105 in 2009 to 0.12/105 in 2020 （AAPC=-7.4%， 95%CI： -13.0% to -1.5%， P<0.05）. The standardized mortality rate for pneumonia decreased from 2.63/105 in 2009 to 0.70/105 in 2020 （AAPC=-6.2%， 95%CI： -12.2% to 0.2%， P=0.056）， but not statistically significant. The annual average mortality rates of COPD， pneumonia and asthma were all highest in January. Crude mortality rates for COPD （χ2=2 669.01， P<0.001）， pneumonia （χ2=217.82， P<0.001）， asthma （χ2=100.09， P<0.001）， pneumoconiosis （χ2=26.46， P<0.001） and all categories of respiratory diseases （χ2=2 995.84， P<0.001） increased with age showed an increasing trend. The crude mortality rates for COPD （χ2=101.69， P<0.001）， pneumonia （χ2=7.39， P<0.01） and asthma （χ2=7.41， P<0.01） were higher in the central than in the northern part of Baoshan District， while the crude mortality rate for COPD （χ2=19.97， P<0.001） was higher in the central than in the southern part. ConclusionThe attention should be focused on COPD； increased detection in males and the elderly， especially in winter and spring； and a good balance between environmental and economic when planning the regional development.

Malignant tumors in farmers over 60 years old of the high-risk population in a town in Shanghai / 上海预防医学

Objective To screen for malignant tumors and high-risk factors in rural residents over 60 years old， so as to prevent and control the occurrence and development of tumors in the future. Methods The survey was conducted with reference to part of the questionnaire in the "Urban Cancer Early Diagnosis and Treatment Project and Evaluation of High-risk Populations". Clinical examinations included serum tumor marker detection， CT screening for lung cancer， occult blood （+） plus colonoscopy screening for colorectal cancer， and mammography screening. Individuals who were positive in the abovementioned clinical tests were defined as high-risk subjects. Results A total of 271 high-risk subjects （1.91%） were screened out of 14 161. Among the high-risk subjects， 71 cases of malignant tumors （26.19%） were found， with an incidence rate of 501.38 per 105. The top five tumors （63.38% of all diagnosed） were mainly concentrated in lung， upper digestive tract， blood system， urinary system， and rectum-colon. The proportion of malignant tumors detected by positive indicators was 61.54% of blood； 46.15% of carcinoembryonic antigen and carbohydrate antigen 125； 23.08% of alpha-fetoprotein； 16.66% of lung CT； and 3.09% of prostate PSA. The positive indicators in the high-risk subjects were mainly for the tumors in the prostate， lungs， liver， and CEA/CA125. The subjects with positive test indicators had lower average annual income in the last 5 years than the normal subject group （χ2=3.380， P=0.040）. The subjects with positive test indicators had higher proportion in family history of tumors than the normal group （χ2=2.596， P=0.046）. People in thehigh-risk group had a higher proportion than the normal group in suffering from hypertension， liver disease， gastrointestinal disease， respiratory system disease， and surgical treatment. Patients with high-risk tumors were found to have higher proportion than the normal group in showing pre-tumor clinical symptoms in the last 1 year. Study of the tumor-related risk factors found that the high-risk group had a higher proportion of high-fat/high-cholesterol diet， alcohol drinking， passive smoking， and personality depression. Conclusion High tumor risk factors have been identified in this population. It is necessary to strengthen the corresponding intervention and follow-up treatment of precancerous diseases in the future. We recommend the government to conduct tumor screening among high-risk groups to improve cost-effectiveness.

Predictive modeling of 30-day readmission risk of diabetes patients by logistic regression, artificial neural network, and EasyEnsemble

Objective: To determine the most influential data features and to develop machine learning approaches that best predict hospital readmissions among patients with diabetes. Methods: In this retrospective cohort study, we surveyed patient statistics and performed feature analysis to identify the most influential data features associated with readmissions. Classification of all-cause, 30-day readmission outcomes were modeled using logistic regression, artificial neural network, and EasyEnsemble. F1 statistic, sensitivity, and positive predictive value were used to evaluate the model performance. Results: We identified 14 most influential data features (4 numeric features and 10 categorical features) and evaluated 3 machine learning models with numerous sampling methods (oversampling, undersampling, and hybrid techniques). The deep learning model offered no improvement over traditional models (logistic regression and EasyEnsemble) for predicting readmission, whereas the other two algorithms led to much smaller differences between the training and testing datasets. Conclusions: Machine learning approaches to record electronic health data offer a promising method for improving readmission prediction in patients with diabetes. But more work is needed to construct datasets with more clinical variables beyond the standard risk factors and to fine-tune and optimize machine learning models.

Prediction of near-term breast cancer risk based on virtual optical density image / 中国医学影像技术

Objective To investigate the value of improving the prediction accuracy of near-term risk for developing breast cancer by transforming the original mammography image and fusing the different types of image features using the algorithm of machine learning.Methods The craniocaudal (CC) full-field digital mammography (FFDM) of 185 women were downloaded from the clinical database at the university of Pittsburgh medical center.Firstly,the original gray images were segmented and transformed into virtual optical density images.Then the asymmetry features were separately extracted from original gray images and virtual optical density images.Two decision tree classifiers of the first stage were trained based on the features extracted from two types of image.And the scores output from the two classifiers were used as input to train the second stage of one decision tree classifier.Leave-one-case-out method was used to validate the prediction performance of near-term risk of breast cancer.Results Using two-stage decision tree fusion method to predict breast cancer,the area under the ROC curve (AUC) was 0.9612±0.0132.And the sensitivity,specificity and prediction accuracy were 96.63％(86/89),91.67％(88/96) and 94.05％(174/185).Conclusion The features extracted from virtual optical density image have higher discriminatory power of predicting breast cancer.Fusing the two kinds of image features twice by two-stage decision tree method can help to improve the prediction accuracy of near-term risk of breast cancer.

Study on Remote Sensing Monitoring Method of Aquatic Medicine Materials Euryale ferox Salisb Based on Multi-source Satellite Remote Sensing Image / 世界科学技术-中医药现代化

Multi-source satellite remote sensing technology can be used to monitor the distribution and growth status of aquatic plant species on a large scale.In this paper,C,aoyou Lake was selected as the research area.Aquatic medicine material Euryaleferox Salisb was used as the research object.The spectral characteristics of plants in Euryaleferox Salisb growing area were analyzed by ASD portable spectrometer and the remote sensing image of Pléiades and GF-1.The spectral range of species was obtained.And the decision tree algorithm model was constructed,which were used to extract the information of Euryale ferox Salisb from remote sensing images.Through verification,the results showed that the accuracy of comprehensive classification was 83％.It was concluded that multi-source satellite remote sensing image and GIS spatial analysis technology can accurately reflect the area and distribution of aquatic medicine material Euryale ferox Salisb.

Research Status of Objective Characteristics of TCM Digitized Tongue / 世界科学技术-中医药现代化

Computer science and technology has been used to promote the development of objectification of traditional Chinese medicine (TCM), which is also required for international development of TCM. In this paper, on the basis of current situation of Chinese medicine tongue objective research, the analysis was made on involved computer-related technology, relevant standards, and the future trend was discussed.

Research of Tongue Image Analysis Instrument Based on Image Analysis / 医疗卫生装备

Objective To develop a kind of tongue image analysis instrument which is used to deal with the deviation problem on the process of tongue image analysis currently. Methods The system included two parts:hardware and software. The hardware was made up of special tongue image acquisition device and computer. With the application of digital image analysis technology, the instrument implements functions such as tongue image collection, image preprocess, image segmentation and feature analysis. Results After a certain amount of sample study process, the accuracy of tongue shape recognition was about 90%. Conclusion This instrument can achieve the aim of tongue information objectification and are widely used for clinical medicine and teaching.