Author(s): Chaithra N and Madhu B
Background: The huge amounts of data generated by healthcare transactions are complex and voluminous. They need to be processed and analysed by different traditional methods. Data Mining provides the methodology and technology to transform these amounts of data into useful information for decision making. Cardiovascular diseases are one of the highest-flying diseases of the modern world. The treatment of the said disease is quite high and not affordable by most of the patients particularly in India. To solve this problem, Data Mining is the best available technique for classification and prediction.
Aim: Research work was aimed to analyse the various data mining techniques introduced in recent years to design a predictive model for cardiovascular diseases from the data obtained by transthoracic echocardiography.
Methods: A total of 336 records with 24 attributes were highly relevant in predicting heart disease from echocardiography dataset were analysed by applying techniques prospectively. This study investigates three different classification models: J48 Decision Tree, Naive Bayes and Neural Network on cardiovascular disease prediction and the same has been justified with the results of different experiments conducted and the performance of the models was evaluated using the standard metrics of Accuracy, Precision, Recall and F-measure.
Discussion and Conclusion: The results of all the three algorithms performed best in true negative rate which makes it a handy tool to train medical students and junior cardiologists to diagnose patients with heart disease.
