مجله انفورماتیک سلامت و زیست پزشکی
سال سوم شماره 4 (پیاپی 10، زمستان 1395)

The importance of users’ role in success or failure of hospital information system and effective factors in successful implementation of these systems is obvious. This study aimed to identify acceptance and use of hospital information systems among medical record department employees at Isfahan teaching hospitals.
This was an applied and analytic study. Research population was staff of medical record departments of 11 teaching hospitals in Isfahan/ Iran. One available user from each medical record unit (admission, statistics, coding and filing) was selected and since in some of the hospitals, the filling unit is not active, totally, 39 users were included in the study. Data were collected through a questionnaire. The validity of the questionnaire was confirmed by 10 academic members and its reliability was determined by Cronbach's alpha (89%). Data were analyzed using Pearson correlation coefficient, one-way analysis of variance, corresponding non-parametric and Kruskal Wallis Test and through SPSS21.
The results showed a strong and positive correlation of behavioral intention with performance expectation (r=0.488), Social effects (r=0.607) and effort expectation (r=0.304). In contrast, there was weak correlation between behavioral intention and facilitating conditions (r=0.197).
The results showed that from the point of views of medical record department’s staff of Isfahan teaching hospitals, use of hospital information system improves job performance and it seems that for improvement of the present systems, users’ expectations and needs are important factors and cause successful implementation of information systems in health sector.

Coronary Artery disease is the most common type of heart disease and one of the leading causes of death in industrialized countries. The aim of this study was to design an expert system with high accuracy for Coronary artery disease diagnosis.
In this applied study, 14 features of 303 patients underwent coronary angiography were used. Dempster-Shafer theory of evidence combination was used to combine the results of three classifying methods including Decision Tree, K-Nearest Neighbor and Neural Network, in order to design a more accurate coronary artery disease diagnostic system. The data mining tool (Weka version 3.7) and C# in Net Framework environment were used for the implementation of model. The 10-fold cross-validation was used for the efficiency assessment.
According to the results, mean accuracy, sensitivity, and specificity of the proposed system were 90.1%, 89.09% and 91.3% respectively. These values were higher in comparison with each of the participated classifiers in the combination. Moreover, in comparison to the similar studies, this method showed higher accuracy for the diagnosis of coronary artery disease.
The results of this research indicates that in the studied population, the proposed method has better accuracy in the diagnosis of coronary heart disease. This method, as an expert system, can help clinicians in making decisions, reducing clinical errors, improving the time to get a diagnostic through reducing waiting time and reducing unnecessary medical tests.

Electronic records can be used as one of the key technologies in health care field. Electronic record, with its several capabilities, is an efficient tool for documentation, information exchange and cooperation of health care organizations. The aim of this study was to design an electronic record based on the traditional medicine approach for infertile patients.
The present study was an applied- development research. First, information requirements questionnaire was filled out by 20 traditional medicine specialists and the obtained data were analyzed through SPSS22 software. Then, a system based on data was designed and given to 10 traditional medicine specialists for evaluation of its practical use and patients’ satisfaction rate.
A majority of data elements in the questionnaire were found necessary by the respondents and in relation to practical use, it gained mean score of 8.6 (out of 9). The designed system, in addition to providing data storage, retrieval and reporting is capable of changing the used data, creating new forms and displaying items with different sizes and types according to the user’s comments and without any need for programming.
According to the obtained results, the electronic medical record designed based on traditional medicine approach was an effective step in managing health data of infertile patients. Designing electronic record based on traditional medicine approach for infertile patient’s increases efficient performance, function, storage and retrieval of health data.

Decision making Support Systems, along with Semantic Web technology, develop a new approach to help physicians in the diagnosis of a variety of illnesses. On the other hand, medical ontology is a knowledge model of clinical domain including all the concepts related to the diagnosis, treatment, clinical procedures and patient data.
In this developmental-applied study, a system for the diagnosis of diseases in various fields of medicine based on the chest pain as a common symptom has been developed. Various symptoms, diagnostic tests and also the intervening nature of disease, meaning that one disease could be a diagnostic symptom of another disease, have been considered in this system and by applying it, real data of patients hospitalized in Shafa hospital in Kerman/ Iran have been studied through Protege software and program language based on Jena rule.
The proposed method was compared to conventional and multilevel methods in regard to the disease diagnosis and the number of required rules. The proposed method, as opposed to the conventional method, is capable of detecting disease at the most inner level. Furthermore the proposed multi-level and entity-related method is capable of detecting disease by means of just seven rules even in the worst cases.
Multilevel entity-based approach accompanied with semantic technology is an effective approach in medical diagnosis systems.

Today, one of the most common diseases and causes of death in the world is heart diseases. Data mining techniques are very useful to create predictive models for identifying people at risk and decreasing the disease complications. In this study, using C4.5 decision tree method, the prevention and diagnosis of this disease are discussed.
This was an applied descriptive study. UCI standard data and Cleveland data collection were used. The database contains 297 records. Analysis was performed through Weka software and using CRISP3 methodology. The C4.5 decision tree model, using input variables and determining the target variable, was created.
According to the applied model, it was found that high levels of cholesterol, sex, age, high maximum heart rate, scan thallium higher than 3 and abnormal ECG have the greatest impact on the risk of coronary heart disease. Furthermore, by using the created decision tree, some rules were extracted that can be used as a model to predict the risk of coronary heart disease. The accuracy of the model created by using decision tree was over 80 percent.
According to our calculations, the rate of categorization was 72.6% and the accuracy of C4.5 algorithm was 80.2% that in comparison with the results of studies in the field of data mining of heart diseases, the obtained accuracy for the suggested algorithm is acceptable.

Detection of pulmonary nodules using CT scan images is one of the methods for early detection of cancer. One of the main challenges for the detection of pulmonary nodules is identifying pulmonary nodules and differentiating them from lung components. In this study, a computer-aided detection system is proposed for the detection of these nodules.
In this descriptive analytical study, 97 chest CT-scan images were studied. To detect pulmonary nodules, support vector machine classifier and Genetic algorithm by MATLAB software were used.
In this research on the lung, the areas of images were classified into the two groups of with nodule and without nodule and it was tried to create a fully automated framework to detect lung nodules in the chest CT images. This framework is an essential part of the computer-aided detection system that helps radiologists to detect lung nodules more accurately and rapidly.
According to the results of this study, the proposed system is more efficient than the previous methods for detecting suspicious nodules.

The majority of drugs act through binding with target proteins. Prediction of the interaction between small molecules and proteins is a key element in the process of drug discovery. Advances in Structural Genomics have provided this possibility to access three-dimensional structures of the proteins which are targeted by drugs. Since laboratory processes through which drug-target protein interactions are investigated require high cost and energy, in silico methods can be used as effective strategies for providing useful information in support of experimental methods.
In this study, the interaction between androgen receptor and Bicalutamide, a widely used drug in the treatment of prostate cancer, was investigated via computational analyses. The docking analysis of this receptor with Bicalutamide was done using Autodock4.2.6 and analysis of complexes was done through LigPlot4.5.3 and Chimera1.5.3.
The obtained results showed that amino acid residues Met-895, Trp-741, Arg-752, Ile-899, Leu-707, Gly-708, Gln-711, Met-745, Met-749, Thr-87, Phe-764, and Met-749 (through forming hydrophobic bonds with the drug) and amino acid residues Asn-705 and Leu-704 (through forming hydrogen bonds with the drug) play a significant role in the protein-drug interaction and cause proper positioning of the drug in protein and consequently its efficacy.
These results could be used in the future studies for investigation of drug resistance to Bicalutamide and to develop more efficient medicines.

Due to the improvement of technology during the last decade, using machine learning algorithms for predicting diseases has found great importance. The goal of this research was to investigate the importance of Naïve Bayesian network as the most applied algorithm in predicting diseases and classifying relevant articles related to disease prediction with data mining algorithms.
This was a systematic review study. A comprehensive search was performed from 2007 to 2017 in online databases and search engines including Scopus, Science Direct, web of science and MEDLINE.
From a total of 90 identified abstracts through the research, 27 ones were compatible with inclusion and exclusion criteria. Naïve Bayesian network was compared with other algorithms and in 92% of articles (25 articles out of 27), it had better accuracy in disease prediction. Results of this research showed effectiveness of Naïve Bayesian algorithm in disease prediction.
Naïve Bayesian network is one of the best algorithms for disease prediction in comparison with experts’ decision and other algorithms. This algorithm can be used beside physicians’ decision to improve the accuracy of disease prediction.