Author: Kazi Amit Hasan
In this project, I tried to analyze the risk factors of Musculoskeletal Pain and tried to find out some interesting patterns among the symptoms. Association Rule mining and Statistical Logistic Regression were used to get the insights and to find out the risks.
The data from this study are a sample of 464 patients, each one assigned to one of the three types of pain by a group of experienced physiotherapists. The assessment was carried out according to a list of 36 binary clinical indicators (“Present”/“Absent”). These indicators are selected on an expert consensus basis regarding symptoms and features of the patients.
- Are there any interesting patterns in the presence/absence clinical criteria for lower back pain?
- Do the patients form groups with similar presence/absence clinical criteria? Do these groups have a connection with the clinical pain types?
- Can we use the presence/absence clinical criteria to accurately predict the clinical pain types?