Abstract—Heart diseases and strokes are considered as number one killer as they account for around 35 to 40 per cent of the total disease burden in Pakistan. The ratio of heart patients is increasing day by day, which is an alarming condition for the country. This situation needs a detailed analysis which can show the geographical distribution of heart patients and also the city wise attributes (age, weight, income etc) that are aggregating more in the heart disease. A Threshold Based Inference Engine is designed which infers the knowledge base by generating the association rules on each city. These rules infer the clustered data to extract the city wise more risk increasing attributes, and the common disease in that city. Automated Minnesota code is used for the verification of the collected ECGs. The results show that Threshold based Inference Engine successfully and efficiently generates a detailed report of each city including more diseased people and highlights the attributes increasing the risk factor.
Index Terms—Arrhythmias, centroid, ECG, fuzzification, inference engine, membership etc.
S. Safdar and F. Arif are with Military College of Signals (MCS),
National University of Sciences and Technology (NUST) Islamabad,
S. A. Khan is with Electrical and Mechanical College (EME), National University of Sciences and Technology (NUST) Islamabad, Pakistan.
Cite: Saria Safdar, Shoab Ahmad Khan, and Fahim Arif, "Report Generation on ECGs Survey Data Analysis Using Threshold Based Inference Engine," International Journal of Information and Education Technology vol. 2, no. 3, pp. 265-269, 2012.