Recently, in medical field, different training methods for medical staff have been proposed. However, the lack of knowledge on the real improvements of trainees makes difficult the real effectiveness of those proposed methods. Therefore, we are proposing an Active Training system for the effective medical training. The Active training system is characterized by providing quantitative information of the trainee's performance of the task to the trainee, simulating real-world conditions of the task, and assuring training effectiveness. In order to fulfill each of these characteristics, we have developed Waseda Kyotokagaku Airway No.3 (WKA-3) which makes it possible to obtain quantitative information for the trainee's performance and reproduce the various cases, individual differences for the airway difficulties for the simulating real-world conditions of the task. In this paper, we are proposing a patient scenario generation which can reproduce the characteristics of the patient simulating real-world conditions of the task for WKA-3. In emergent situation or surgical operation, the characteristics of patients are presented in the three parameters such as: patient initial conditions, time variant status change, and reflex action. By adjusting, and combining each of the three parameters, we can reproduce the various patient scenarios. In this paper, we also state how to generate the Patient Scenario Generation using the position control and virtual compliance control. Finally, a set of experiments has been carried out to the doctor subjects in order to verify effectiveness of the proposed Patient Scenario Generation, and discuss the doctors about the result of the experiments.

An active training system is characterized by providing quantitative information of the trainee's performance of the task to the trainee, simulating real-world conditions of the task, and assuring training effectiveness. For this purpose, the authors have been developing an Airway management training system. In particular, we have previously presented an evaluation model system to acquire quantitative information for the trainee's performance, and an actuated model system for real-world conditions of the task. In order to fulfill all of the characteristics of the active training system; in this paper, we would like to propose an integrated system, Waseda Kyotokagaku Airway No.3 (WKA-3) which integrates the function of both system models. For this reason, we have focused on embedding sensors and actuators with the link drive mechanism and the wire drive mechanism to a conventional patient model towards the development of a patient robot. In this paper, we present WKA-3 how to design the mechanism and how to redesign several organs to embed the sensors for the measurement of the trainee's performance. A set of experiment have been carried out to doctor subjects using the WKA-3 to verify the effectiveness of the proposed system by asking doctors to evaluate the proposed system.

The emerging field of medical robotics aims to introduce intelligent tools to assist a physician. The main challenges for developing efficient medical robotic training systems are simulating real-world conditions of the task and assuring training effectiveness. High anatomic fidelity has been achieved in current systems. However, those training systems are designed to reproduce specific conditions of the task (passive training). In this research, we are focusing in developing an airway management training systems designed to reproduce various cases of difficult airway. Such difficulties (i.e. restricted mouth opening, various shapes of oral cavity including tongue swallowing, etc) may provoke traumas on different organs of the patient during an emergency situation. For this purpose, the authors have proposed the development of more advanced training tools. For this purpose, we have focused in embedding sensors and actuators to a conventional patient model towards the development of a patient robot (in previous research, authors have presented the evaluation model which embeds sensors). In this paper, we present an airway training system which embeds actuators into a mannequin. In particular, the mechanism design of the Waseda Kyotokagaku Airway No.2 (WKA-2) is detailed. The WKA-2 is composed of twelve active and one passive degrees of freedom; which are designed to reproduce the various cases of difficult airway. For this purpose, the WKA-2 reproduces the human muscles around the upper airway based on a wire driving mechanisms (a total of sixteen wires are used). In particular, the head of the model is composed of a tongue and mandible with translational and rotational movements around kinematic axis. In addition, we present the details of the kinematic model of WKA-2 which enable the robot to reproduce the airway difficulties. Finally, we presented the design of a tension sensor designed to measure the applied tension on each of the wire driving mechanism of WKA-2. As preliminary experiments, we reproduce several cases of difficult airways using WKA-2.

We have developed a vocal control method, based on forward and inverse models, to allow the anthropomorphic talking robot Waseda Talker No. 7 (WT-7) to produce various kinds of voices. The control parameters of the vocal cords on WT-7 are pressure, vocal cord tension and glottal opening, and the acoustic parameters are sound pressure, sound pitch and spectrum slope. The relationships among these parameters are complicated and difficult to model using conventional methods. Here we present a neural network (NN) control method. The learning process consists of creation of the NN forward model by back propagation methods and optimization of the inverse model using the forward model. In addition, a real-time auditory feed-back mechanism is used to reduce the error between the target and the generated acoustic parameters. Using this method, the control parameters can be adjusted to follow the target voice well.