Term Project Progress Report

Medical Decision Support System
18 March 2016

Submitted To : Submitted By:

Dr. Vatcharaporn Esichaikul Raja Vyshnavi (st118183)

Tata Rohit (st118238)

Sachham Man Buddhacharya (st117891)

Prithvi Raj (st118177)

Tejasree (st118187)

INDEX S.NO TOPIC NAME PAGE NUMBER 1 INTRODUCTION 2 1.2 IMPLEMENTATION 6 3.2 CONVERSION OF UCI DATA TO EXCEL 4 3 METHODOLOGY 6 3.1 OBJECTIVE 2 1.2 SCOPE 2 2 DATA 2 2.2.1 DATA SET 2 2.1 SCREENSHOTS 6 4 REFERENCES 12 5 FUTUREWORK 12 1.1 MODEL 6 3.Introduction 1 .

The diagnosis of heart diseases needs clinical and pathological data. One of the common heart disease is Cardiovascular disease. Table 1. there remains some grey area in their diagnosis decision. The dataset consist of 303 instances of which 164 belonging to healthy person and 139 instances belonging to heart diseases. We have specifically used the heart dataset of Cleveland Clinic Foundation. The data consist of 76 raw attributes.1 Data set The data is retrieved from heart disease data from UCI machine learning repository. Doctors do not possess expertise in all matters. Some of the instances contained missing values which were filled by random numbers. Our system helps the doctor to make rigid decision and complete the patient's diagnosis with higher level of precis 1.Data 2.2 Scope ● The scope of this DSS extends to the prediction of occurrence of a cardiovascular disease using Decision Tree model. we have tried to predict the cardiovascular diseases using decision trees. 2. of which only 13 have been used. In our research.1 Objective ● To develop a DSS for predicting Cardiovascular disease using Prediction model. Heart diseases are one of the main cause of death worldwide.Clinical features and their description Name Type Description Age Continuous Age in years Sex Discrete 1 = male 0 = female Cp Discrete Chest pain type: 1 = typical angina 2 = atypical angina 3 = non-anginal pain 4 =asymptomatic Trestbps Continuous Resting blood pressure (in mm Hg) Chol Continuous Serum cholesterol in mg/dl 2 . ● To reduce the workload of the end user as the system can predict the occurrence of a cardiovascular disease from clinical and pathological data 1.

individual data is extracted using space as delimiter in text import wizard.2 Conversion of UCI data to excel The Data retrieved from UCI machine learning library is in the form of plain text file with comma as delimiter. The formatted data is saved in CSV format to be imported in RapidMiner software. Thal Discrete 3 = normal 6= fixed defect 7= reversible defect Diagnosis Discrete Diagnosis classes: 0 = healthy 1= patient who is subject to possible heart disease 2. Fbs Discrete Fasting blood sugar > 120 mg/dl: 1 = true 0 = false Restecg Discrete Resting electrocardiographic results: 0 = normal 1 = having ST-T wave abnormality 2 =showing probable or define left ventricular hypertrophy by Estes’criteria Thalach Continuous Maximum heart rate achieved Exang Discrete Exercise induced angina: 1 = yes 0 = no Old peak Continuous Depression induced by exercise relative to rest Slope Discrete The slope of the peak exercise segment : 1 = up sloping 2 = flat 3= downsloping Ca Discrete Number of major vessels colored by fluoroscopy that ranged between 0 and 3. 3 . The downloaded file is imported to Microsoft Excel.

One Instance with 76 attributes Figure. Collection of selected 13 attributes and the diagnosis result(0-4) 4 . Figure.

Excel data 3.2 Implementation 3.2.Methodology 3.1 Screenshots 5 . Figure. 3.1 Model In order to construct a DSS for Cardiovascular disease we use Decision Tree type model.

Figure : Rapidminer Home page Figure : Adding Data to the software. 6 .

Figure : Locating the data on the system Figure : Imported Data 7 .

Figure : Labeling a column Figure : Choosing the type of role 8 .

Figure : Choosing label as the role Figure : Storing the data in the repository 9 .

+++++++++++++++++++++++++++++++++++++++++++++++++++ 10 .

References Pandey. 6. K. 23-30). Inc. December)... & Stocker. 11 . R. Australian Computer Society. (2014). L. & Jaiswal. K. Using decision tree for diagnosing heart disease patients. A heart disease prediction model using decision tree. T. M. M. Tamil Nadu. IUP Journal of Computer Sciences.. Heart Disease Diagnosis Using Predictive Data mining. 43. Venkatalakshmi. India. Shouman. Turner.7(3). In 2014 IEEE International Conference on Innovations in Engineering and Technology (ICIET’14).. Figure : Decision Tree 5. P. Pandey. A. (2013). & Shivsankar.. In Proceedings of the Ninth Australasian Data Mining Conference- Volume 121 (pp. Futurework ● Based on the further training using the data set we expect to predict decision accurately. B.. (2011.

12 .