Search

Background: Achieving healthy ageing has become the only way for China to alleviate the pressure of ageing, especially in rural areas. However, the factors affecting the health of rural older adults are numerous and complex. It is important to identify the critical factors that affecting the health of older adults in rural areas and provide decision-making support for targeted health interventions.Methods: To overcome some limitations of existing works, an extended probabilistic linguistic fuzzy cognitive map model is proposed in this paper as a useful tool for modeling the cause-effect relationship between factors. The proposed model integrates the advantages of probabilistic linguistic term sets and fuzzy cognitive maps. In the end, to rank and identify the critical factors affecting the health, a novel similarity measure based on Euclidean distance and Z-mapping function is proposed.Results: The proposed model can effectively deal with the uncertainty of experts and reflect different opinions of groups well. In terms of representing uncertainty and ambiguity, the proposed method outperforms other models in modeling complex systems. In the real-world case analysis, we find that education is the most important factor affecting the health of rural older adults, followed by previous occupational experiences, psychology, and physical exercise, among other things. Intergenerational relationship has become another important factor affecting the health of rural older adults in China as the development of Chinese society.Conclusions: From a macro perspective, social economic status, living environment, lifestyle, and health management, are the variables that have the greatest impact on the health of rural older adults. As a result, providing more precise health interventions with the characteristics of factors influencing health is a crucial guarantee for preserving and improving the health of rural older adults in China. [ABSTRACT FROM AUTHOR]

Background: With the character of high incidence, high prevalence and high mortality, stroke has brought a heavy burden to families and society in China. In 2009, the Ministry of Health of China launched the China national stroke screening and intervention program, which screens stroke and its risk factors and conducts high-risk population interventions for people aged above 40 years old all over China. In this program, stroke risk factors include hypertension, diabetes, dyslipidemia, smoking, lack of exercise, apparently overweight and family history of stroke. People with more than two risk factors or history of stroke or transient ischemic attack (TIA) are considered as high-risk. However, it is impossible for this criterion to classify stroke risk levels for people with unknown values in fields of risk factors. The missing of stroke risk levels results in reduced efficiency of stroke interventions and inaccuracies in statistical results at the national level. In this paper, we use 2017 national stroke screening data to develop stroke risk classification models based on machine learning algorithms to improve the classification efficiency.Method: Firstly, we construct training set and test sets and process the imbalance training set based on oversampling and undersampling method. Then, we develop logistic regression model, Naïve Bayesian model, Bayesian network model, decision tree model, neural network model, random forest model, bagged decision tree model, voting model and boosting model with decision trees to classify stroke risk levels.Result: The recall of the boosting model with decision trees is the highest (99.94%), and the precision of the model based on the random forest is highest (97.33%). Using the random forest model (recall: 98.44%), the recall will be increased by about 2.8% compared with the method currently used, and several thousands more people with high risk of stroke can be identified each year.Conclusion: Models developed in this paper can improve the current screening method in the way that it can avoid the impact of unknown values, and avoid unnecessary rescreening and intervention expenditures. The national stroke screening program can choose classification models according to the practice need. [ABSTRACT FROM AUTHOR]

Background: Standardized coding of plays an important role in radiology reports' secondary use such as data analytics, data-driven decision support, and personalized medicine. RadLex, a standard radiological lexicon, can reduce subjective variability and improve clarity in radiology reports. RadLex coding of radiology reports is widely used in many countries, but translation and localization of RadLex in China are far from being established. Although automatic RadLex coding is a common way for non-standard radiology reports, the high-accuracy cross-language RadLex coding is hardly achieved due to the limitation of up-to-date auto-translation and text similarity algorithms and still requires further research.Methods: We present an effective approach that combines a hybrid translation and a Multilayer Perceptron weighting text similarity ensemble algorithm for automatic RadLex coding of Chinese structured radiology reports. Firstly, a hybrid way to integrate Google neural machine translation and dictionary translation helps to optimize the translation of Chinese radiology phrases to English. The dictionary is made up of 21,863 Chinese-English radiological term pairs extracted from several free medical dictionaries. Secondly, four typical text similarity algorithms are introduced, which are Levenshtein distance, Jaccard similarity coefficient, Word2vec Continuous bag-of-words model, and WordNet Wup similarity algorithms. Lastly, the Multilayer Perceptron model has been used to synthesize the contextual, lexical, character and syntactical information of four text similarity algorithms to promote precision, in which four similarity scores of two terms are taken as input and the output presents whether the two terms are synonyms.Results: The results show the effectiveness of the approach with an F1-score of 90.15%, a precision of 91.78% and a recall of 88.59%. The hybrid translation algorithm has no negative effect on the final coding, F1-score has increased by 21.44% and 8.12% compared with the GNMT algorithm and dictionary translation. Compared with the single similarity, the result of the MLP weighting similarity algorithm is satisfactory that has a 4.48% increase compared with the best single similarity algorithm, WordNet Wup.Conclusions: The paper proposed an innovative automatic cross-language RadLex coding approach to solve the standardization of Chinese structured radiology reports, that can be taken as a reference to automatic cross-language coding. [ABSTRACT FROM AUTHOR]

Background: Text Matching (TM) is a fundamental task of natural language processing widely used in many application systems such as information retrieval, automatic question answering, machine translation, dialogue system, reading comprehension, etc. In recent years, a large number of deep learning neural networks have been applied to TM, and have refreshed benchmarks of TM repeatedly. Among the deep learning neural networks, convolutional neural network (CNN) is one of the most popular networks, which suffers from difficulties in dealing with small samples and keeping relative structures of features. In this paper, we propose a novel deep learning architecture based on capsule network for TM, called CapsTM, where capsule network is a new type of neural network architecture proposed to address some of the short comings of CNN and shows great potential in many tasks.Methods: CapsTM is a five-layer neural network, including an input layer, a representation layer, an aggregation layer, a capsule layer and a prediction layer. In CapsTM, two pieces of text are first individually converted into sequences of embeddings and are further transformed by a highway network in the input layer. Then, Bidirectional Long Short-Term Memory (BiLSTM) is used to represent each piece of text and attention-based interaction matrix is used to represent interactive information of the two pieces of text in the representation layer. Subsequently, the two kinds of representations are fused together by BiLSTM in the aggregation layer, and are further represented with capsules (vectors) in the capsule layer. Finally, the prediction layer is a connected network used for classification. CapsTM is an extension of ESIM by adding a capsule layer before the prediction layer.Results: We construct a corpus of Chinese medical question matching, which contains 36,360 question pairs. This corpus is randomly split into three parts: a training set of 32,360 question pairs, a development set of 2000 question pairs and a test set of 2000 question pairs. On this corpus, we conduct a series of experiments to evaluate the proposed CapsTM and compare it with other state-of-the-art methods. CapsTM achieves the highest F-score of 0.8666.Conclusion: The experimental results demonstrate that CapsTM is effective for Chinese medical question matching and outperforms other state-of-the-art methods for comparison. [ABSTRACT FROM AUTHOR]

Background: Many models are published which predict outcomes in hospitalized COVID-19 patients. The generalizability of many is unknown. We evaluated the performance of selected models from the literature and our own models to predict outcomes in patients at our institution.Methods: We searched the literature for models predicting outcomes in inpatients with COVID-19. We produced models of mortality or criticality (mortality or ICU admission) in a development cohort. We tested external models which provided sufficient information and our models using a test cohort of our most recent patients. The performance of models was compared using the area under the receiver operator curve (AUC).Results: Our literature review yielded 41 papers. Of those, 8 were found to have sufficient documentation and concordance with features available in our cohort to implement in our test cohort. All models were from Chinese patients. One model predicted criticality and seven mortality. Tested against the test cohort, internal models had an AUC of 0.84 (0.74-0.94) for mortality and 0.83 (0.76-0.90) for criticality. The best external model had an AUC of 0.89 (0.82-0.96) using three variables, another an AUC of 0.84 (0.78-0.91) using ten variables. AUC's ranged from 0.68 to 0.89. On average, models tested were unable to produce predictions in 27% of patients due to missing lab data.Conclusion: Despite differences in pandemic timeline, race, and socio-cultural healthcare context some models derived in China performed well. For healthcare organizations considering implementation of an external model, concordance between the features used in the model and features available in their own patients may be important. Analysis of both local and external models should be done to help decide on what prediction method is used to provide clinical decision support to clinicians treating COVID-19 patients as well as what lab tests should be included in order sets. [ABSTRACT FROM AUTHOR]

Background: It is significant to model clinical activities for process mining, which assists in improving medical service quality. However, current process mining studies in healthcare pay more attention to the control flow of events, while the data properties and the time perspective are generally ignored. Moreover, classifying event attributes from the view of computers usually are difficult for medical experts. There are also problems of model sharing and reusing after it is generated.Methods: In this paper, we presented a constraint-based method using multi-perspective declarative process mining, supporting healthcare personnel to model clinical processes by themselves. Inspired by openEHR, we classified event attributes into seven types, and each relationship between these types is represented in a Constrained Relationship Matrix. Finally, a conformance checking algorithm is designed.Results: The method was verified in a retrospective observational case study, which consists of Electronic Medical Record (EMR) of 358 patients from a large general hospital in China. We take the ischemic stroke treatment process as an example to check compliance with clinical guidelines. Conformance checking results are analyzed and confirmed by medical experts.Conclusions: This representation approach was applicable with the characteristic of easily understandable and expandable for modeling clinical activities, supporting to share the models created across different medical facilities. [ABSTRACT FROM AUTHOR]

Background: The social Q&A community quickly becomes a popular platform for consumers to find health information because of its convenience and interactivity.Methods: Based on the 10,861 depression questions collected in the Zhihu, the largest Q&A platform in China, we divided the healthy information needs description into nine categories with Latent Dirichlet Allocation (LDA). We also divided the healthy information needs type into Physiological, affective and cognitive needs based on the Wilson model.Results: The results show that the largest categories are depression symptom and social activities while the less concerned health information is prevention and medical insurance. More attention is paid to cognitive needs. We also find there is no strong correlation between attention and needs type.Conclusions: The purpose of this paper is to refine the consumer health information needs types to better understand the consumer health information characteristic in China. [ABSTRACT FROM AUTHOR]

Background: Due to the increasing complexity in socioeconomic environments and the ambiguity in human cognition, decision makers prefer to give linguistic cognitive information with different granularities according to their own preferences. Consequently, to consider the uncertainty and preferences in the evaluation process, a method based on Multi-Granularity Linguistic Information (MGLI) for evaluating teleconsultation service quality is proposed, which provides a new research direction for scientific evaluation and improvement of teleconsultation service quality.Methods: Firstly, this paper explored a service quality evaluation system from the perspective of regional doctors. And then considering the uncertainty and preferences of decision makers, MGLI was used to optimize the index system according to the similarity degree between the linguistic evaluation information and a given linguistic term set. Finally, the empirical research was conducted using Henan Province Telemedicine Center of China (HTCC) as an example to identify the direction for improving the service quality in teleconsultation.Results: This study found that the number of consulting rooms, attitude of operators, consultation duration, charges, and attitude of experts are the key factors affecting the quality of teleconsultation service.Conclusions: Suggestions for improving the quality of teleconsultation service are put forward in terms of optimizing the allocation of consulting rooms, improving regional doctors' experience and standardizing charging standards, which provides a new direction for improving the quality of teleconsultation service. [ABSTRACT FROM AUTHOR]

Background: Traditional Chinese medicine (TCM) is a highly important complement to modern medicine and is widely practiced in China and in many other countries. The work of Chinese medicine is subject to the two factors of the inheritance and development of clinical experience of famous Chinese medicine practitioners and the difficulty in improving the service capacity of basic Chinese medicine practitioners. Heterogeneous information networks (HINs) are a kind of graphical model for integrating and modeling real-world information. Through HINs, we can integrate and model the large-scale heterogeneous TCM data into structured graph data and use this as a basis for analysis.Methods: Mining categorizations from TCM data is an important task for precision medicine. In this paper, we propose a novel structured learning model to solve the problem of formula regularity, a pivotal task in prescription optimization. We integrate clustering with ranking in a heterogeneous information network.Results: The results from experiments on the Pharmacopoeia of the People's Republic of China (ChP) demonstrate the effectiveness and accuracy of the proposed model for discovering useful categorizations of formulas.Conclusions: We use heterogeneous information networks to model TCM data and propose a TCM-HIN. Combining the heterogeneous graph with the probability graph, we proposed the TCM-Clus algorithm, which combines clustering with ranking and classifies traditional Chinese medicine prescriptions. The results of the categorizations can help Chinese medicine practitioners to make clinical decision. [ABSTRACT FROM AUTHOR]

Background: It has been shown that the entities in everyday clinical text are often expressed in a way that varies from how they are expressed in the nomenclature. Owing to lots of synonyms, abbreviations, medical jargons or even misspellings in the daily used physician notes in clinical information system (CIS), the terminology without enough synonyms may not be adequately suitable for the task of Chinese clinical term recognition.Methods: This paper demonstrates a validated system to retrieve the Chinese term of clinical finding (CTCF) from CIS and map them to the corresponding concepts of international clinical nomenclature, such as SNOMED CT. The system focuses on the SNOMED CT with Chinese synonyms enrichment (SCCSE). The literal similarity and the diagnosis-related similarity metrics were used for concept mapping. Two CTCF recognition methods, the rule- and terminology-based approach (RTBA) and the conditional random field machine learner (CRF), were adopted to identify the concepts in physician notes. The system was validated against the history of present illness annotated by clinical experts. The RTBA and CRF could be combined to predict new CTCFs besides SCCSE persistently.Results: Around 59,000 CTCF candidates were accepted as valid and 39,000 of them occurred at least once in the history of present illness. 3,729 of them were accordant with the description in referenced Chinese clinical nomenclature, which could cross map to other international nomenclature such as SNOMED CT. With the hybrid similarity metrics, another 7,454 valid CTCFs (synonyms) were succeeded in concept mapping. For CTCF recognition in physician notes, a series of experiments were performed to find out the best CRF feature set, which gained an F-score of 0.887. The RTBA achieved a better F-score of 0.919 by the CTCF dictionary created in this research.Conclusions: This research demonstrated that it is feasible to help the SNOMED CT with Chinese synonyms enrichment based on physician notes in CIS. With continuous maintenance of SCCSE, the CTCFs could be precisely retrieved from free text, and the CTCFs arranged in semantic hierarchy of SNOMED CT could greatly improve the meaningful use of electronic health record in China. The methodology is also useful for clinical synonyms enrichment in other languages. [ABSTRACT FROM AUTHOR]

Background: Healthcare providers generate a huge amount of biomedical data stored in either legacy system (paper-based) format or electronic medical records (EMR) around the world, which are collectively referred to as big biomedical data (BBD). To realize the promise of BBD for clinical use and research, it is an essential step to extract key data elements from unstructured medical records into patient-centered electronic health records with computable data elements. Our objective is to introduce a novel solution, known as a double-reading/entry system (DRESS), for extracting clinical data from unstructured medical records (MR) and creating a semi-structured electronic health record database, as well as to demonstrate its reproducibility empirically.Methods: Utilizing the modern cloud-based technologies, we have developed a comprehensive system that includes multiple subsystems, from capturing MRs in clinics, to securely transferring MRs, storing and managing cloud-based MRs, to facilitating both machine learning and manual reading, and to performing iterative quality control before committing the semi-structured data into the desired database. To evaluate the reproducibility of extracted medical data elements by DRESS, we conduct a blinded reproducibility study, with 100 MRs from patients who have undergone surgical treatment of lung cancer in China. The study uses Kappa statistic to measure concordance of discrete variables, and uses correlation coefficient to measure reproducibility of continuous variables.Results: Using the DRESS, we have demonstrated the feasibility of extracting clinical data from unstructured MRs to create semi-structured and patient-centered electronic health record database. The reproducibility study with 100 patient's MRs has shown an overall high reproducibility of 98 %, and varies across six modules (pathology, Radio/chemo therapy, clinical examination, surgery information, medical image and general patient information).Conclusions: DRESS uses a double-reading, double-entry, and an independent adjudication, to manually curate structured data elements from unstructured clinical data. Further, through distributed computing strategies, DRESS protects data privacy by dividing MR data into de-identified modules. Finally, through internet-based computing cloud, DRESS enables many data specialists to work in a virtual environment to achieve the necessary scale of processing thousands MRs within days. This hybrid system represents probably a workable solution to solve the big medical data challenge. [ABSTRACT FROM AUTHOR]

Background: The probability of heart failure during the perioperative period is 2% on average and it is as high as 17% when accompanied by cardiovascular diseases in China. It has been the most significant cause of postoperative death of patients. However, the patient is managed by the flow of information during the operation, but a lot of clinical information can make it difficult for medical staff to identify the information relevant to patient care. There are major practical and technical barriers to understand perioperative complications.Methods: In this work, we present three machine learning methods to estimate risks of heart failure, which extract intraoperative vital signs monitoring data into different modal representations (statistical learning representation, text learning representation, image learning representation). Firstly, we extracted features of vital signs monitoring data of surgical patients by statistical analysis. Secondly, the vital signs data is converted into text information by Piecewise Approximate Aggregation (PAA) and Symbolic Aggregate Approximation (SAX), then Latent Dirichlet Allocation (LDA) model is used to extract text topics of patients for heart failure prediction. Thirdly, the vital sign monitoring time series data of the surgical patient is converted into a grid image by using the grid representation, and then the convolutional neural network is directly used to identify the grid image for heart failure prediction. We evaluated the proposed methods in the monitoring data of real patients during the perioperative period.Results: In this paper, the results of our experiment demonstrate the Gradient Boosting Decision Tree (GBDT) classifier achieves the best results in the prediction of heart failure by statistical feature representation. The sensitivity, specificity and the area under the curve (AUC) of the best method can reach 83, 85 and 84% respectively.Conclusions: The experimental results demonstrate that representation learning model of vital signs monitoring data of intraoperative patients can effectively capture the physiological characteristics of postoperative heart failure. [ABSTRACT FROM AUTHOR]