Intra-Operative Distinction of Breast Tissues Based on Neural Network Integration of Impedance Information

Bioelectrical impedance measurement technology used in intra-operative breast tissues distinction can shorten the waiting time and lighten patients´ pain in partial mastectomy. This paper focuses on researching the differences of electrical properties between breast cancer and normal breast tissues (mammary gland and adipose). Therefore, electrode structure is studied in this paper to design a suited hardware system, in order to acquire impedance information. Through simulation analyses and experiments, the arc-shaped electrode is proved to have powerful character distinguishing ability. Frequency-resistance curves are painted and their slopes are extracted as characteristic parameters to express the overall trend. BP neural network is selected and the integration of several neural networks is researched. The results indicate that the integral neural network has a good performance in distinction.