A neuro-fuzzy model of soft tissue in haptic simulator for training diagnosis of breast cancer

Soft tissue force modeling with the approach of creating a force-feedback simulator for training medical skills has been the focus of many attempts up to now. The most important parameter considered in soft tissue modeling, is its being real-time along with its precision and high sensitivity. In this article, using ANFIS (Adaptive Neuro Fuzzy Inference System), a neuro-fuzzy model is presented for soft tissue force modeling. In order to evaluate the efficiency of the model, it has been used for simulated training of breast cancerous tumors diagnosis using a haptic interface. Data needed for training neuro-fuzzy network related to the model, is provided from breast tissue modeling in ANSYS 12.0 software. In validating session, numerical data have been confirmed with the experimental data with an average error of less than 3%. Testing session indicates root mean square error of the model to be about 0.02 (N), which shows a high degree of precision for the model. The presented model's real-time feature is about 100 times of the maximum needed amount for force modeling simulations.