Improvement of Binary Classification Accuracy by Integration of Haptic and Image Information in Haptic Ultrasonic Probe System

Abstract

Ultrasonic diagnosis is noninvasive method for internal examination. However, it requires a high level of operational skill. It is expected that the new support system for effective operation is developed. In addition, tele-medicine has also been quite attractive for improvement of quality of life. From these social demands, this paper proposes a new haptic ultrasonic probe based on bilateral control. This can communicate both haptic and image information. In order to support the automatic classification of contact objects, Neural Network is implemented. It is confirmed that the integration of haptic and image information can improve the recognition rate from experiment.