Using Bayesian networks to built a diagnosis and prognosis model for breast cancer

In recent years, the breast cancer has become one the most common cancer among women. The challenge faced currently is how to implement the early detection and accurate diagnosis for this disease. In this paper, we first introduced the definition of Bayesian network and discussed its advantages in the fields of medicine diagnosis and prognosis. Then, the original breast cancer data records used for case study are collected from the first affiliated hospital of medical college of Xi´an Jiaotong University, China, which are also discretized to build the standard modelling dataset. At last, the physical BN model, living BN model, test BN model, diagnosis BN model and prognosis BN model are learned from the dataset respectively for each treatment process of breast cancer. These BN models can help doctors to estimate the state of cancer by inputting corresponding patients´ condition parameters.