Cross Sectional Survey Based Intelligent Framework for the Diagnosis of Frozen Shoulder

Abstract

Mostly the efforts made in medical diagnosis are considered an art. However in today’s computing machinery, it’s really difficult how to cope with uncertainty. After the advancement in data mining techniques, the researchers are encouraged to develop software to assist doctors in making decision without the specialists. Data mining techniques also help in extracting useful data from unstructured reports. It also has high potential for extracting the hidden information from the datasets. In Pakistan the work done in orthopedics related to predicting diseases is very less. Frozen shoulder is a state where a shoulder becomes stiff. Accurate diagnosis of frozen shoulder is helpful in providing economical and effective treatment for patients. This research provides the classification of unstructured data using data mining techniques. It also uses predictive models like Naïve Bayesian and Random forest to improve the diagnosis of frozen shoulder. Intelligent system is based on personal data and symptoms history to classify frozen shoulder into two groups “Intact” and “No-Intact” which is gathered from doctors and patients. A complete software requirement based datasets is formed after gathering all knowledge and by analyzing the surveys. We analyzed our results by finding performance measures like sensitivity, recall, precision etc. The proposed system also detects all important variables which are necessary for disease analysis by using P-value method. The validity of variables is tested using datasets and results show the worth of proposed system. This research will also help the researcher in future in the area of orthopedics.