The emerging field of medical robotics is aiming in introducing intelligent tools to perform medical procedures. In particular, robotic researchers have been proposed advanced medical training systems to enhance motor dexterities of trainees. An efficient medical training system should be designed to simulate real-world conditions and to assure their effectiveness as the representation of the motor skills often differs among trainees. Up to now, several training simulators have been developed by medical companies designed to reproduce with high fidelity the human body. However, such devices are not designed to provide any information to trainees. Therefore, we have proposed the development of a patient robot which embeds sensors and actuators into a conventional human model. Due to its complexity, as a first approach; we are presenting the development of a suture training system designed to simulate the real-world task conditions as well as providing quantitative assessments. In particular; the Waseda-Kyotokagaku Suture No.2 Refined II is presented, which includes a new evaluation function to provide more detailed information of the task. A set of experiments were proposed to analyze the performance of trainees. From the experimental results, we could confirm its effectiveness to detect differences of the performance of trainees.

Up to now, there is no widely accepted quantitative evaluation scheme. Nowadays, an objective structured clinical examination has been proposed as a modern type of examination often used in medicine to test skills such as medical procedures, etc. The assessment of skills is realized by practical exams, in which students are evaluated by experienced examiner using a check list. However, the examiner lacks of information which cannot be obtained trough the simple observation of the task. Thanks to the advances in robot technology in embedded systems, etc.; more advanced evaluation tools can be conceived. For this reason, at Waseda University, we have proposed the development of a patient robot as an advanced evaluation tool to provide more detailed information of the task. As a first approach of our long-term research target, we have proposed the development of a suture/ligature training system which provides quantitative information of the movement of a dummy skin as well as information of the quality of task. In this paper; we describe the functionalities of the newest version, the Waseda-KyotoKagaku Skin No.2 Refined (WKS-2R), which has been designed to provide quantitative information of the task. In addition, we are proposing a new evaluation function which includes performance indexes and weighting coefficients. As a first approach, the weighting coefficients were determined by using the discriminant analysis. A set of experiments were proposed to confirm the effectiveness of the proposed evaluation function. From the preliminary results, the evaluation function was useful in detecting differences among different levels of expertise as well as detecting improvements during the training process by computing the learning curve.