Identification of the inclusion biomechanical properties in soft tissues by artificial neural network

In this paper an artificial neural network model for identifying inclusion properties based on measured biomechanical data is demonstrated. The force-displacement curves for a set of breast phantoms under different loading and exploration conditions were accumulated and analyzed. Curve features were extracted. It is successfully to formulate the relationship between the inclusion biomechanical properties and curve features. The inverse biomechanical model is then solved using an artificial neural network (ANN) model. The results show that the ANN model combined with lateral exploration strategy has the capability accurately to predict the inclusion properties, such as inclusion stiffness, when the indentation depth is close to the underlying depth of the inclusion.