Farzad Firouzi Jahantigh, Behzad Kiani, Saina Etemad,
Volume 23, Issue 145 (7-2016)
Abstract
Background: Having precisely analyzed the data of patients with specific diseases we can obtain the patterns and knowledge of these disease or even specific characteristics of patients. A hypothesis is usually considered in medical studies when some data are gathered prospectively to prove or deny this hypothesis, but in many cases there may be relationships between the data of the patients which have never been attended and no hypothesis has been considered. Thus, in this study available hidden patterns in the data of patients with tuberculosis have been discovered.
Methods: Data of the study included 600 patients with tuberculosis who had referred to Masih Daneshvari hospital of Tehran. Data were gathered by reading patients files and observing clinical tests of patients from hospital data system. APPIRIORI data mining technique and WEKA tool of data mining were utilized to discover the associative relationships of the data.
Results: Hypertension diseases, diabetes insipidus and ischemic heart disease have had the most frequency in patients with tuberculosis. Patients with diabetes insipidus or night sweats had also experienced chronic cough. Patients who have had weight loss and had BK+ test one result, had also experienced chronic cough. Patients who have been coughing up blood and had BK+ sputum tests (3), had also experienced chronic cough. Male patients who had night sweats, had also experienced weight loss. Patients who have had weight loss and fever, had also experienced night sweats.
Conclusion: Discovered rules can be considered as primary hypothesis for the upcoming studies especially those of clinical trials, In addition to this, physicians can use these rules to analyze the clinical condition of patients.
Nastran Dehghan, Hamid Hassanpour, M. R. Abbaszadegan,
Volume 23, Issue 152 (2-2017)
Abstract
Background: Microarray technology is a powerful tool to study and analyze the behavior of thousands of genes simultaneously. Images of microarray have an important role in the detection and treatment of diseases. The aim of this study is to provide an automatic method for the extraction and analysis of microarray images to detect cancerous diseases.
Methods: The proposed system consists of three main phases of image processing, data mining, and detection of disease. The image processing phase is accompanied with some operations such as identifying the location of genes, deleting the background, and extracting the raw data from the images. The second phase includes data normalization and selection of more effective genes. The disease is identified and recognized in the third phase using the extracted data.
Results: In this study it has been used from breast cancer microarray images from Stanford University database. The accuracy of the proposed method to locate genes and diagnosis of breast cancer is up to 98 and 95.45%, respectively.
Conclusion: The obtained results indicate that the proposed method is more accurate than other existing methods in microarray analysis. In addition, the proposed method is easily implemented and less costly compared to the clinical tests.
Elham Parvinnia, Meysam Mohammadi, Ali Mohammad Bananzadeh, Seyedeh Parastoo Khayami,
Volume 25, Issue 9 (12-2018)
Abstract
Background: In recent years the growing trend of colon cancer has revealed that we need some safe and new methods to detect and control this disease. Data mining is one of these methods, one of its most important applications is the discovery of hidden patterns between data in a large database. In this study, we explore and discover unknown patterns in a real colon cancer data set.
Methods: In this study, the information of 400 colorectal cancer patients, with 42 cfeature has been studied.This information was collected through the Colorectal Research Center, Shiraz University of Medical Sciences, between 2008 and 2016. After performing the data set preprocessing, the hidden relationships between the features of this data are discovered through the Fp-Growth algorithm.
Results: Ater using this algorithm and discovering the relationship between some of the features, some rules have been developed. Based on the suggestion of the specialist and the importance of the features, the rules have been studied in seven groups.
Conclusion: The results of the review of the laws indicate that the pathologic stage and the age of the patient had a significant effect on the survival rate of the patients.
Also, the percentage of men and women with rectal cancer is greater than that of the clone, and the sex does not affect the survival of the patient.
Other findings from the review of this data can be the lack of a meaningful relationship between the patient's pathologic stage and the demographic information.
Aliyeh Kazemi, Maliheh Fazeli, Shahpar Haghighat,
Volume 25, Issue 12 (3-2019)
Abstract
Background: Lymphadenitis is a common and debilitating complication of breast cancer patients. This study has predicted and classified lymphedema complications. Moreover, identifying effective factors and discovering patterns for faster diagnosis and prevention of this complication is another goal of this study.
Methods: Data from 1113 patients with breast cancer who were referred to Seyed Khandan Lymphedema clinic during 2009 to 2017 were evaluated. Data analysis was performed by using IBM SPSS Modeler software version 18 and CRISP-DM methodology and in the modeling section, logistic regression and decision tree algorithms were used.
Results: Data from 933 patients including 25 variables were entered into the model after pre-processing. Probability of catching of Lymphedema for each patient predicted by logistic regression algorithm and different decision tree algorithms consist of C5.0, Chaid, C&RT, and Quest with the sensitivity of 79.33%, 74.41%, 71.92%, 72.64% and 77.83%, respectively and finally the factors related to Lymphedema were identified. Ratio of the involved lymph node numbers to the removed lymph node numbers, heaviness, type of surgery, stage of disease, age, body mass index, metastasis, number of chemotherapy courses, comorbidity, number of removed lymph nodes respectively.
Conclusion: The results show that C5.0 decision tree algorithm with the highest sensitivity is the best model for predicting Lymphedema. By applying the rules created for a new sample with specific characteristics, it can be predicted that the patient will probably suffer from Lymphedema or not. Considering that BMI is a changeable factor, weight control regimens are recommended for these patients. In addition, it is necessary to pay attention to the patient's heavy feeling in the early stages.
