دانشگاه علوم پزشکی کرمان Journal of Health and Biomedical Informatics 2423-3870 5 2 2018 9 1 Evaluation of the Effect of Feature Selection and Different kernel Functions on SVM Performance for Breast Cancer Diagnosis 244 251 FA Azam Orooji PhD Student in Medical Informatics, Health Information Management Dept., School of Health Management and Information Sciences, Iran University of Medical Sciences, Tehran, Iran Mostafa Langarizadeh PhD of Medical Informatics, Health Information Management Dept., School of Health Management and Information Sciences, Iran University of Medical Sciences, Tehran, Iran. Introduction: Breast cancer is one of the most common cancers affecting women. In mammography, differentiating a malignant tumor from a benign one is a very tedious task due to their structural similarities. Machine Learning (ML) is a subfield of Artificial Intelligence that can be used as an effective tool to help physicians to make decisions. Support vector machine (SVM) is one of the most common ML techniques that its performance depends on kernel parameters tuning and input features. The aim of this study was to investigate the effect of feature selection and different kernel functions on SVM performance. Methods: This analytic study was performed through comparative method. Genetic algorithm was used for feature selection. SVM models based on different kernel functions, including polynomial, Linear, Radial Basis Function (RBF), Quadratic and Multi-Layer Perceptron (MLP), were first performed with all features and then, with the selected features. The Wisconsin original breast cancer data set was used as a training set to evaluate the performance of the classifiers. All implementations were done in MATLAB environment. Results: According to the obtained results, by applying feature selection, the performance of SVM with MLP kernel function decreased and with quadratic kernel function increased. However, the performances of the linear and RBF kernels were desirable in both conditions. Generally, after the dimension reduction, the best accuracy, specificity, sensitivity and accuracy were dropped by 0.663, 0.833, 1.077 and 0.138 percent respectively. Conclusion: The ML-based methods can help physicians in diagnosis and decision makings for treatment.

دانشگاه علوم پزشکی کرمان Journal of Health and Biomedical Informatics 2423-3870 5 2 2018 9 1 Comparing the Performance of Data Mining Algorithms in Predicting Coronary Artery Diseases 252 264 FA Haleh Ayatollahi Department of Health Information Management, School of Health Management and Information Sciences, Iran University of Medical Sciences, Tehran, Iran. Leila Gholamhosseini PhD Student in Health Information Management, School of Health Management and Information Sciences, Iran University of Medical Sciences, Tehran, Iran. Masoud Salehi Department of Biostatistics, School of Public Health, Iran University of Medical Sciences, Tehran, Iran. Introduction: Cardiovascular diseases are the first leading cause of death worldwide. World health organization has estimated that the morality rate due to heart diseases will mount to 23 million cases by 2030. Hence, the use of data mining algorithms will be useful in predicting coronary artery disease. The objective of the present study was to compare the accuracy of the CAD predictions made by ANN and SVM techniques. Methods: The present study was conducted via descriptive-analytical method. The research sample included all CAD patients hospitalized in three hospitals affiliated to AJA University of Medical Sciences from March 2016 to March 2017. Totally, 1324 records with 26 characteristics affecting the disease were extracted and after normalizing, and cleaning of the data, they were entered in SPSS statistics V23.0 & IBM Excel 2013; then, R3.3.2 data mining software was used to format data. Results: SVM model with lower MAPE (112.03) and higher Hosmer-lemeshow statistic (16.71), sensitivity (92.23) and specificity (74.42) yielded better fitness of data and provides more accurate prediction than ANN model. On the other hand, since the area under the ROC curve in SVM algorithm was more than that in ANN, it could be concluded that this model had higher accuracy. Conclusion: According to the results, SVM algorithm presented higher accuracy and better performance than ANN model and showed higher sensitivity and accuracy. It is suggested that in future studies, the results of the present study be compared with the findings resulted from applying other data mining algorithms.

دانشگاه علوم پزشکی کرمان Journal of Health and Biomedical Informatics 2423-3870 5 2 2018 9 1 Factors Related to the Acceptance of Radio Frequency Identification Technology in Health Information Management Departments of Hospitals Affiliated to Iran University of Medical Sciences 265 273 FA Abbas Sheikhtaheri Ph.D. in Health Information Management, Assistant Professor, Health Information Management Dept., School of Health Management and Information Sciences, Iran University of Medical Sciences, Tehran, Iran Soghra Rostami Garavand MSc in Health Information Technology, Health Information Management Dept., School of Health Management and Information Sciences, Iran University of Medical Sciences, Tehran, Iran Hossein Ahmadi Ph.D. in information systems, Assistant Professor, Health Information Management Dept., School of Health Management and Information Sciences, Iran University of Medical Sciences, Tehran, Iran Introduction: Radio Frequency Identification (RFID) technology has been implemented in a wide variety of industries including healthcare field. Due to the introducing RFID technology in the healthcare industry, it is necessary to determine the factors influencing acceptance of this technology. Therefore, in the present study, the factors associated with the acceptance of RFID technology in Health Information Management hospitals affiliated to Iran University of Medical Sciences were studied. Methods: The current applied research was descriptive-analytic. A questionnaire was distributed among all health information management staff of the hospitals affiliated to Iran University of Medical Sciences (n=138) and finally, 135 questionnaires were completed. The results of the study were analyzed through SPSS16 and using descriptive statistics, inferential statistics, and regression. Results: The mean scores of factors were as follow: social effects (4.24 out of 5), management support (4.34 out of 5), expected performance (4.20 out of 5), expected effort (4.09 out of 5), facilitating conditions (3.47 out of 5), security (3.83 out of 5) and intention (4.3 out of 5). The coefficients of influence of social effects (0.125), management support (0.210), expected performance (0.244), expected effort (0.148) and facilitating conditions (0.135) showed positive significant relationships with intention to use RFID. However, security was not an influential factor on use of RFID. Conclusion: Based on the findings, providing general education, facilitating the use of this technology by vendors and developing adequate infrastructure may increase the willingness to accept RFID.

دانشگاه علوم پزشکی کرمان Journal of Health and Biomedical Informatics 2423-3870 5 2 2018 9 1 Diagnosis of Coronary Heart Disease using Mixture of Experts Method 274 285 FA Majid Hassanzadeh Ph.D in Higher Education Management, Assistant Professor, Nursing Dept., Torbat Heydarieh University of Medical Sciences, Torbat Heydarieh, Iran Iman Zabbah PhD Student in Computer, School of Electrical and Computer, Tehran North Branch, Islamic Azad University, Tehran, Iran Kamran Layeghi Ph.D. in Computer, Assistant Professor, Department of Computer, North Tehran Branch, Islamic Azad University, Tehran, Iran. Introduction: Coronary Artery Disease (CAD) is one of the most common heart diseases and the main cause of mortality in men and women. This study aimed to predict the disease status using Neural Network compound (mixture of experts). Methods: The present study was a diagnostic study conducted on 200 patients referred to a heart specialty center in Torbat-e-Heydarieh. Patients' files contained their demographic information including13 risk factors. A model for predicting CAD based on multilayer perceptron neural network and mixture of experts was produced. Results: First, we used a neural network of multilayer perceptron with Propagation algorithm by different architectures. The best architecture could predict closed coronary artery with the accuracy of 71.7%. Then, by increasing the number of neural networks and training process, results were combined. Mixture of experts by liner method (majority voting) and nonlinear method (gating network) was applied and the accuracy rates of 75.8 percent and 78.3 percent were respectively obtained. Conclusion: Angiography is an invasive diagnostic procedure with risk factors such as stroke and heart attack. Therefore, non-invasive methods should be used for the diagnosis of CAD. In this study, with increasing the number of learners and their nonlinear mixture, the accuracy of diagnosis was increased.

دانشگاه علوم پزشکی کرمان Journal of Health and Biomedical Informatics 2423-3870 5 2 2018 9 1 The Effect of Education via Mobile Phones on Procrastination of Iranian Users: Designing a Treatment Aid Application 286 292 FA Mina Nazari Kamal Ph.D. Student in Psychology, General Psychology Dept., Faculty of Literature and Human Science, Islamic Azad University of Birjand, Birjand, Iran Raheleh samouei Ph.D. Student in Health Management in Disaster, Social Determinants of Health Research Center, Health Services Management Dept., Isfahan University of Medical Sciences, Isfahan, Iran. Sheida Sarafzade MSc in Biostatics, Biostatistics and Epidemiology Dept., School of Public Health, Isfahan University of Medical Sciences, Isfahan, Iran. . Ph.D. in Clinical Psychology, Clinical Psychology Dept., Roozbeh Hospital, Tehran University of Medical Sciences, Tehran, Iran Nafiseh Ghaebi MSc in Psychology, General Psychology Dept., Faculty of Psychology and Educational Sciences, Islamic Azad University, Garmsar Branch, Garmsar, Iran Fatemeh Moradi MSc in Information Technology Engineering, Faculty of Computer Engineering and Information Technology, Amirkabir University of Technology, Tehran, Iran Mahtab Moradzadeh BSc in Information Technology Engineering, Faculty of Computer Engineering and Information Technology, Amir Kabir University of Technology, Tehran, Iran Introduction: Procrastination, in addition to wasting time, reducing personal performance and causing feeling of guilt, can cause failure of receiving treatment or counseling. Therefore, using accessible and easy-to-use technologies such as mobile phones can facilitate the treatment of people with this problem. This study aimed to investigate the effects of education via mobile phones on procrastination of Iranian users. Methods: The study was carried out using semi-experimental single group method on 32 Iranian users who had installed the counter procrastination application on their android mobile phones. The participants filled Basco’s procrastination questionnaire before and after the education. After two months of using the application, the results of pre-test and post-test were gathered and analyzed using student t-test. Results: Findings showed that education via mobile phones has been effective in reducing the procrastination in participants. Conclusion: This application, due to ease-of-use, high accessibility and saving cost, time and energy, can be useful in consulting, treatment and education centers.

دانشگاه علوم پزشکی کرمان Journal of Health and Biomedical Informatics 2423-3870 5 2 2018 9 1 A Detection of Type2 Diabetes using C4.5 Decision Tree 293 303 FA Hamed Sabbagh Gol M.Sc in Computer Engineering, Faculty of Computer, Department of Computer Engineering, Payame Noor University (PNU), Iran Introduction: One of the most common diseases in the world is diabetes and the global prevalence of diabetes increases by about six percent annually. The use of data mining techniques to create predictive models is very helpful in identifying people at risk and reducing the complications of the disease. In this study, through using decision tree C4.5, methods of prevention and treatment of diabetes were investigated. Methods: In this applied and descriptive study, we used the standard UCI data and the pima-Indians-diabetes data set. This database contains 768 records with 8 fields. The analysis was done using Weka software using the CRISP3 methodology. In modeling decision tree, C4.5 was created using input variables and determining target variables. Also, the sensitivity, specificity, accuracy, as well as positive and negative predictive values were used to evaluate the model. Results: According to the model, high blood sugar levels, high gravidity, high age, high diastolic blood pressure, familial history and high BMI have respectively the highest effects on type 2 diabetes mellitus. The ranking rate was 73.8% and the accuracy of the C4.5 algorithm was 79%. Conclusion: Compared to the results of studies in the field of data mining for diabetes, the accuracy of the proposed algorithm is acceptable. The most effective factors on diabetes were identified. Also, rules were developed that can be used as a model to predict the risk of diabetes in people.  

دانشگاه علوم پزشکی کرمان Journal of Health and Biomedical Informatics 2423-3870 5 2 2018 9 1 Detection and Extraction of Potential Promoter/Enhancer Interactions in Genome of Cancer Patients using an Evolutionary Multi-Objective Algorithm 304 313 FA Mohammadjavad Hosseinpoor Ph.D. student in computer engineering, Computer Engineering Dept., Yasooj Branch, Islamic Azad University, Yasooj, Iran Hamid Parvin Ph.D. in Computer Engineering - Artificial Intelligence, Assistant Professor, Computer Engineering Dept., Nourabad Mamasani Branch, Islamic Azad University, Nourabad Mamasani, Iran Samad Nejatian Ph.D. in Electrical Engineering, Assistant Professor, Electrical Engineering Dept., Yasooj Branch, Islamic Azad University, Yasooj, Iran vahideh Rezaei Ph.D. in Mathematics, Assistant Professor, Mathematics Dept., Yasooj Branch, Islamic Azad University, Yasooj, Iran Introduction: Cancer, as one of the most common diseases, has influenced the health of many people. The main aim of this study was to present a multi-objective evolutionary algorithm. The algorithm is capable of detecting and extracting potentially promoter/enhancer areas in the chromosomes of the affected people using the information concerning inter-genomic interactions. The correct extraction of these areas can help early diagnosis of cancer. Methods: In this applied and descriptive research, Hi-C data set including information on intergenomic interactions in the GM12878 cell was used. Multi-objective evolutionary algorithm was used in order to discover and extract potential promoter /enhancer interactions. The mentioned algorithm was implemented using MATLAB software. Furthermore, the efficiency of this algorithm was evaluated using two criteria. The first criterion is a proportional function that calculates the magnitude of inter-genomic interactions relative to the length of the genome regions; and the second criterion is the number of discovered potential promoters/enhancers. Results: The results and comparisons showed higher efficiency and optimality of the suggested method in discovering promoter/Enhancer interactions with variable length in comparison to HiCPro method. Therefore, the suggested method is able to discover the potential promoter/ enhancer interactions that cannot be discovered by HiC-Pro method. Conclusion: The suggested algorithm is able to optimally discover and extract potential promoter/ enhancer with variable length. This is a great help in medical science for early diagnosis of cancer.

دانشگاه علوم پزشکی کرمان Journal of Health and Biomedical Informatics 2423-3870 5 2 2018 9 1 Review of the Approaches and Techniques of Text Mining in Clinical Documentation 314 324 FA Farzaneh Feizmanesh MSc in Medical Informatics, Medical Informatics Dept., Faculty of Medical Sciences, Tarbiat Modares University, Tehran, Iran Ali Asghar Safaei Assistant Professor, Computer engineering-software, Medical Informatics Dept., Faculty of Medical Sciences, Tarbiat Modares University Tehran, Tehran, Iran Introduction: With the extensive use of electronic medical records systems, a large amount of medical text data is produced daily in the hospitals and other medical environments that organizing this text information is important and necessary. Also, a need to automatically retrieve useful knowledge from this data to help clinicians is felt. In order to extract the hidden values in the medical text documents, text mining can be used in the field of health. Methods: In this review study, SID, Magiran, Pubmed, ScienceDirect, IEEE, and Google Scholar databases were searched with the keywords including "Text Mining” AND "Medicine” ،"Clinical Text Mining” AND "Predict”،, "knowledge discovery in medical text” and "Text Mining for Medical and Healthcare” in the English databases and keywords such as "text mining” AND " Discovering Knowledge in Medicine” in the Persian databases. Then, all articles that somehow refer to Medical knowledge discovery and text mining applications in the field of health were selected. Results: Text mining is one of the important and powerful techniques for extracting information from health information systems. Text mining in clinical data provides potential for new discoveries and it also improves efficiency and communication in hospital systems for doctors and hospital administrators. Conclusion: Nowadays, text mining in clinical documentation, is one of the developed technologies for the discovery of medical knowledge that its use in medical databases is essential to achieving immediate access to important health resources, and its application can improve patient care and reduce medical costs.