Chien-Yeh Hsu

Building a portable data and information interoperability infrastructure-framework for a standard Taiwan Electronic Medical Record Template

Traditional electronic health record (EHR) data are produced from various hospital information systems. They could not have existed independently without an information system until the incarnation of XML technology. The interoperability of a healthcare system can be divided into two dimensions: functional interoperability and semantic interoperability. Currently, no single EHR standard exists that provides complete EHR interoperability. In order to establish a national EHR standard, we developed a set of local EHR templates. The Taiwan Electronic Medical Record Template (TMT) is a standard that aims to achieve semantic interoperability in EHR exchanges nationally. The TMT architecture is basically composed of forms, components, sections, and elements. Data stored in the elements which can be referenced by the code set, data type, and narrative block. The TMT was established with the following requirements in mind: (1) transformable to international standards; (2) having a minimal impact on the existing healthcare system; (3) easy to implement and deploy, and (4) compliant with Taiwans current laws and regulations. The TMT provides a basis for building a portable, interoperable information infrastructure for EHR exchange in Taiwan.

The relationship between varicoceles and obesity in a young adult population.

To determine whether a relationship between obesity and varicocele occurrence exists, the prevalence and severity of varicoceles related to obesity were investigated in a general population of young males. A total of 1050 young males attending the Navy Recruit Training Center were evaluated from their physical screening examinations. All subjects underwent history taking and physical examinations to evaluate for the presence and severity of varicocele. The anthropometric indexes including body mass index (BMI), waist circumference (WC) and waist-to-hip ratio (WHR) were recorded. All subjects were categorized by quartiles according to each anthropometric index. Means were compared with the Students t-test. Severity was compared by analysis of variance testing and frequency was analysed using the chi-square method. Statistical significance was considered at p <0.05. A total of 490 (46.67%) subjects had varicoceles. The means of BMI, WC and WHR of those without varicoceles was 23.99 +/- 3.82 kg/m(2), 83.20 +/- 9.97 cm and 0.85 +/- 0.05, respectively. These judged values were greater than those with varicoceles (22.02 +/- 3.18 kg/m(2), 79.19 +/- 9.01 cm and 0.83 +/- 0.05) (p < 0.001). In the univariate regression analysis, BMI, WC and WHR all had a significantly negative correlation with severity of varicocele (all p < 0.001). Analysis comparing varicocele frequency based on each grade per anthropometric index group was performed. The logistic regression revealed that the prevalence of grade II and III varicoceles showed a statistically inverse association with all three anthropometric indexes. The prevalence and severity of varicoceles inversely correlated with obesity. The present data support the explanation that obesity may result in a decreased nutcracker effect, which accounts for prevention of the renal vein compression by the adipose tissue.

Constructing a nutrition diagnosis expert system

This paper presents a research of constructing a web-based expert system for nutrition diagnosis by utilizing the expert system techniques in artificial intelligence. The research implements Nutritional Care Process and Model (NCPM) defined by American Dietetic Association (ADA) in 2008 and integrate the nutrition diagnosis knowledge from dietetics professionals to establish the basics of building the rule-based expert system with its knowledge base. The system is built using Microsoft Visual Studio 2008 on .NET Framework 3.5SP1 utilizing the built in rule engine which comes with Windows Workflow Foundation. With the help of this system, it is easier for dietetics professionals to adapt to the newly introduced concept of nutrition diagnosis. At the heart of the web based expert system is a knowledge base, it has a rule engine which contains the nutrition diagnosis rules converted from signs and symptoms for nutrition diagnosis from dietetics professionals and are expressed in XML format which are then stored in a SQL database. A knowledge engineer will be able to use a rule editor to add new rules or to update existing rules within the rule database. Dietetics professionals would be able to enter patients basic data, anthropometric data, physical exam findings, biochemical data, and food/nutrition history into the program. After dietetics professionals complete nutrition assessment, the program will make inference to the rule base and make nutrition diagnosis. Dietetics professionals could then make the final diagnosis decision for the patient based on the diagnosis report generated by the web based nutrition diagnosis expert system. For this study, I have selected 100 chronic kidney disease patients under hemodialysis from a university hospital, recorded their albumin, cholesterol, creatinine before dialysis, height, and dry weight and then use these data to perform nutrition diagnosis with both the expert system and a practicing dietitian. After comparing the result, I found that the expert system is faster and more accurate than human dietitian.

Predicting Hospital-Acquired Infections by Scoring System with Simple Parameters

Background Hospital-acquired infections (HAI) are associated with increased attributable morbidity, mortality, prolonged hospitalization, and economic costs. A simple, reliable prediction model for HAI has great clinical relevance. The objective of this study is to develop a scoring system to predict HAI that was derived from Logistic Regression (LR) and validated by Artificial Neural Networks (ANN) simultaneously. Methodology/Principal Findings A total of 476 patients from all the 806 HAI inpatients were included for the study between 2004 and 2005. A sample of 1,376 non-HAI inpatients was randomly drawn from all the admitted patients in the same period of time as the control group. External validation of 2,500 patients was abstracted from another academic teaching center. Sixteen variables were extracted from the Electronic Health Records (EHR) and fed into ANN and LR models. With stepwise selection, the following seven variables were identified by LR models as statistically significant: Foley catheterization, central venous catheterization, arterial line, nasogastric tube, hemodialysis, stress ulcer prophylaxes and systemic glucocorticosteroids. Both ANN and LR models displayed excellent discrimination (area under the receiver operating characteristic curve [AUC]: 0.964 versus 0.969, p = 0.507) to identify infection in internal validation. During external validation, high AUC was obtained from both models (AUC: 0.850 versus 0.870, p = 0.447). The scoring system also performed extremely well in the internal (AUC: 0.965) and external (AUC: 0.871) validations. Conclusions We developed a scoring system to predict HAI with simple parameters validated with ANN and LR models. Armed with this scoring system, infectious disease specialists can more efficiently identify patients at high risk for HAI during hospitalization. Further, using parameters either by observation of medical devices used or data obtained from EHR also provided good prediction outcome that can be utilized in different clinical settings.

Developing guideline-based decision support systems using protégé and jess

The Institute of Medicine has identified both computerized physician order entry and electronic prescription as keys to reducing medication errors and improving safety. Many computerized clinical decision support systems can enhance practitioner performance. However, the development of such systems involves a long cycle time that makes it difficult to apply them on a wider scale. This paper presents a suite of guideline modeling and execution tools, built on Protégé, Jess and Java technologies, which are easy to use, and also capable of automatically synthesizing clinical decision support systems for clinical practice guidelines of moderate complexity.

Developing Electronic Health Records in Taiwan

The design of the Taiwan Electronic Medical Record Template (TMT) provides the basis for developing a document-based information standard and an information interoperability infrastructure for Taiwans healthcare system. An additional benefit of TMT is its strong data security structure.

Application of Multiscale Amplitude Modulation Features and Fuzzy C-Means to Brain–Computer Interface

This study proposed a recognized system for electroencephalogram (EEG) data classification. In addition to the wavelet-based amplitude modulation (AM) features, the fuzzy c-means (FCM) clustering is used for the discriminant of left finger lifting and resting. The features are extracted from discrete wavelet transform (DWT) data with the AM method. The FCM is then applied to recognize extracted features. Compared with band power features, k-means clustering, and linear discriminant analysis (LDA) classifier, the results indicate that the proposed method is satisfactory in applications of brain–computer interface (BCI).

Discovering EEG Signals Response to Musical Signal Stimuli by Time-frequency analysis and Independent Component Analysis

In recent years, a lot of research has focus on the physiological effect of music. The electroencephalographic (EEG) is often used to verify the influence of music on human brain activity. In this study, we used frequency distribution analysis and the independent component analysis (ICA) to analyze to discover the EEG responses of subjects with different musical signal stimuli. It is expected that some features on EEG can be demonstrated to reflect the different musical signal stimuli. The EEG of six healthy volunteers listening different music was recorded. We used International 10-20 System to get 19 channels of EEG signal. Musical signal stimuli are metal music, sonata music and the favorite music selected by subjects. Spectra analyses based on Fourier transform were applied to obtain the alpha, beta, gamma and thetas band power of EEG signal under different music stimuli. We used the power at each band of each channel as the features of EEG. The correlation of the features between different situations and subjects was used to show which channel display the difference of EEG signals. Besides, ICA was applied to assist us in the process of isolating noise components and to provide cues to explain the functions of different brain areas in point of neurology. The result showed that some independent components obtained from ICA can demonstrate more significant difference for different music. The features composed of spectral power of each band are very similar in listening metal music, but showed less similarity in listening sonata music. Hence, the response of EEG to sonata is more meaningful and metal music may induce same effect for different subjects. T3 and Pz are the channels with relatively lower correlation under different music stimuli. Therefore, the locations of T3 and Pz of brain may play an important role in feeling music

The Association between Dietary Patterns and Semen Quality in a General Asian Population of 7282 Males

Objective To explore the associations between different dietary patterns and semen quality in a general Asian male population. Methods Cross-sectional study. Healthy Taiwanese men aged 18 years or older who participated in a standard medical screening program from 2008-2013 run by a private firm were included in this study. Semen parameters including sperm concentration (SC), total sperm motility (TSM), progressive motility (PRM) and normal sperm morphology (NSM) were recorded. A dietary questionnaire was used to categorize the participants into 5 groups: “Healthy diet”, “Western diet”, “High-carbohydrate diet”, “High sweet snacks & sugar-sweetened drinks” and “High-sodium diet”. Results A total of 7282 men completed the questionnaire regarding dietary pattern, and examination of anthropometric indexes was performed and laboratory data were obtained. A high intake of a “Western diet” resulted in statistically linear declines of SC and NSM (P < 0.001 and P < 0.001). Similarly, a greater intake of “High sweet snacks & sugar-sweetened drinks” was associated with a lower SC (P = 0.001). Increased intake of a “High-carbohydrate diet” was related to higher prevalences of abnormal TSM and PRM (P = 0.012 and P = 0.025). Similarly, a greater intake of a “High-sodium diet” was correlated with an elevated prevalence of abnormal NSM (P = 0.035). Conclusions This study showed that a greater intake of a “Western diet” is associated with poorer SC and NSM, a “High sweet snacks and sugar-sweetened drinks” intake is correlated with a lower SC, and high-carbohydrate food is related to elevated prevalences of abnormal TSM and PRM.

Exploration of the association between obesity and semen quality in a 7630 male population.

This study aimed to explore the association between body mass index (BMI), other anthropometric indexes and semen quality in a general male population in Taiwan. In this cross-sectional cohort study, the study cohort consisted of 7941 healthy male individuals aged 18 years or older who participated in a standard medical screening program run by a private firm from January 2008 to May 2013. Semen parameters including sperm concentration (SC), total sperm motility (TSM), progressive motility (PRM), and normal sperm morphology (NSM) were recorded. Anthropometric indexes including BMI, waist circumference (WC), hip circumference (HC), waist-to-hip ratio (WHR), waist-to-height ratio (WHtR) and body fat percentage were measured. A total of 7630 men were enrolled for the final analysis, of whom 68.5% had a normal weight distribution and 31.4% were overweight or obese. Total sperm motility, progressive motility, normal sperm morphology and sperm concentration showed a statistically linear decline with increasing age (p < 0.001, p < 0.001, p < 0.001 and p = 0.004). Sperm concentration showed a significantly negatively linear association with BMI (p = 0.005), and normal sperm morphology showed an inverse association with BMI and waist-to-height ratio (p < 0.001 and p = 0.004). The prevalence of abnormal total sperm motility, progressive motility, normal sperm morphology and sperm concentration increased with increasing age (p = 0.011, p < 0.001, p < 0.001 and p = 0.002). Lower normal sperm morphology and sperm concentration were associated with increasing body adiposity (p<0.05). No relationship between obesity and sperm motility was identified.