Learning internal representations by error propagation
Parallel distributed processing: explorations in the microstructure of cognition, vol. 1
Artificial intelligence: a modern approach
Artificial intelligence: a modern approach
Neural Networks for Pattern Recognition
Neural Networks for Pattern Recognition
Pattern Classification (2nd Edition)
Pattern Classification (2nd Edition)
Hi-index | 0.00 |
Neural networks are a practical tool for solving various problems of approximation, classification, prediction or control. In the paper we use multi-layer perceptrons to determine the character of stress in healthy femur and after endoprosthesoplasty. Inserting metal prosthesis to the bone changes the stress character what can lead to local decalcification and weakening of its strength in certain areas. Dynamic bone load resulting from non-anatomical load can cause fracture in the weak area. Neural network was learned with the data obtained from numerical simulations using the finite element analysis. The input to the network was stress state in twelve points of femur and body mass.