Association Rules Mining Based Clinical Observations

TL;DR

This paper introduces a novel association rule mining approach to analyze healthcare data, revealing disease co-occurrences and correlations to improve clinical understanding and patient care.

Contribution

It develops a system prototype that applies association rule mining to healthcare data for predicting disease correlations, enhancing clinical analysis capabilities.

Findings

Revealed significant disease co-occurrence patterns

Predicted correlations between primary and secondary diseases

Demonstrated system effectiveness in clinical data analysis

Abstract

Healthcare institutes enrich the repository of patients' disease related information in an increasing manner which could have been more useful by carrying out relational analysis. Data mining algorithms are proven to be quite useful in exploring useful correlations from larger data repositories. In this paper we have implemented Association Rules mining based a novel idea for finding co-occurrences of diseases carried by a patient using the healthcare repository. We have developed a system-prototype for Clinical State Correlation Prediction (CSCP) which extracts data from patients' healthcare database, transforms the OLTP data into a Data Warehouse by generating association rules. The CSCP system helps reveal relations among the diseases. The CSCP system predicts the correlation(s) among primary disease (the disease for which the patient visits the doctor) and secondary disease/s (which is/are other associated disease/s carried by the same patient having the primary disease).