The use of fuzzy linear regression models for tumor size in colorectal cancer in hospital of Malaysia

Regression analysis has become popular among several fields of research and standard tools in analysing data. This structure was represented by four commonly statistical models such as multiple linear regression, fuzzy linear regression (Tanaka, 1982), fuzzy linear regression (Ni, 2005) and extended fuzzy linear regression by benchmarking models under fuzziness (Chung, 2012). Colorectal cancer (CRC) was applied on CRC cases in Malaysia. The CRC patients’ quality of life in order to detect the CRC at an early stage is still very poor, the programmes are mainly ad-hoc and not implemented as a national wide programme. This study aims to determine the best model to measure the tumor size at hospitals using mean square error and root mean square error. Secondary data was used where 180 patients having colorectal cancer and receiving treatment in hospitals was recorded by nurses and doctors. Based on the results, fuzzy linear regression (Ni, 2005) is the best model to predict the tumor size developed by patients after receiving treatment in hospital.