Track: Data Analytics

Abstract

Data mining provides automatic pattern recognition and uncovers patterns in data that are difficult to detect with traditional statistical methods. For developing countries with limited resources, data mining can be an answer towards improving the overall health of the population at an affordable cost.

In this paper, we initially analyze how data mining can transform various aspects of health care in general such as in evaluating treatment effectiveness, healthcare management by identifying high risk patients, pharmaceutical industry etc. We then focus on two specific diseases viz. Tuberculosis (an infectious disease) and Cardiovascular disease (lifestyle disease) and analyze how data mining can equip public systems better to tackle them.

In the case of Tuberculosis, given a set of variables such as Gender, Loss of Appetite, Erythrocyte, Age group, Weight, Sweating at Night, Fever Cavity, Exhaustion etc. one can estimate the probability of the presence of the disease in a patient. Various techniques such as ANFIS, Multilayer Perception, Rough-Neural-Networks etc. are available for this purpose. Additionally we also analyze the data from India’s National Family and Health Survey and provide our findings and policy recommendations for tackling Tuberculosis.

Similarly, data mining can aid in early detection of Cardiovascular diseases. The information collected from patients is already digitized in most hospitals. Tools such as cluster analysis, decision trees, neural networks etc. can help in identifying patterns among sufferers of cardiovascular diseases. 

Data mining provides automatic pattern recognition and uncovers patterns in data that are difficult to detect with traditional statistical methods. For developing countries with limited resources, data mining can be an answer towards improving the overall health of the population at an affordable cost.

In this paper, we initially analyze how data mining can transform various aspects of health care in general such as in evaluating treatment effectiveness, healthcare management by identifying high risk patients, pharmaceutical industry etc. We then focus on two specific diseases viz. Tuberculosis (an infectious disease) and Cardiovascular disease (lifestyle disease) and analyze how data mining can equip better public systems to tackle them.

In the case of Tuberculosis, given a set of variables such as Gender, Loss of Appetite, Erythrocyte, Age group, Weight, Sweating at Night, Fever Cavity, Exhaustion etc. one can estimate the probability of the presence of the disease in a patient. Various techniques such as ANFIS, Multilayer Perception, Rough-Neural-Networks etc. are available for this purpose. Additionally we also analyze the data from India’s National Family and Health Survey and provide our findings and policy recommendations for tackling Tuberculosis.

Similarly, data mining can aid in early detection of Cardiovascular diseases. The information collected from patients is already digitized in most hospitals. Tools such as cluster analysis, decision trees, neural networks etc. can help in identifying patterns among sufferers of cardiovascular diseases.