Abstract: The disclosure is to constitute, while reducing a cost for collecting training data used in machine learning that makes a control module acquire an ability to control a robot device, the control module operatable in an actual environment by the machine learning. A learning device according to one aspect of the present invention executes machine learning of an extractor by using a first learning data set constituted by a combination of simulation data and first environmental information and a second learning data set constituted by a combination of actual data and second environmental information. Further, a learning device according to one aspect of the present invention executes machine learning of a controller by using a third learning data set constituted by a combination of third environmental information, state information, and a control command.

Abstract: MRI-guided robotics offers possibility for physicians to perform interventions remotely on confined anatomy. While the pathological and physiological changes could be visualized by high-contrast volumetric MRI scan during the procedure, robots promise improved navigation with added dexterity and precision. In cardiac catheterization, however, maneuvering a long catheter (1-2 meters) to the desired location and performing the therapy are still challenging. To meet this challenge, this invention presents an MRI-conditional catheter robotic system that integrates intra-op MRI, MR-based tracking units and enhanced visual guidance with catheter manipulation. This system differs fundamentally from existing master/slave catheter manipulation systems, of which the robotic manipulation is still challenging due to the very limited image guidance. This system provides a means of integrating intra-operative MR imaging and tracking to improve the performance of tele-operated robotic catheterization.

Abstract: Provided is an image generation device capable of generating, on the basis of inputted images, a learning image for training an action of a robot which carries out a prescribed operation on a workpiece. The image generation device comprises: a first image acquisition unit which acquires a first image capturing a real operation space including the robot and not including the workpiece; a second image acquisition unit which acquires a second image in which a virtual operation space including a virtual robot corresponding to the robot and a virtual workpiece corresponding to the workpiece is rendered; and a learning apparatus so that, in response to an input of the first image and the second image, the apparatus outputs a third image obtained by transforming the second image such that at least the virtual robot included in the second image is made to approximate the robot included in the first image.

Abstract: The disclosure is to constitute, while reducing a cost for collecting training data used in machine learning that makes a control module acquire an ability to control a robot device, the control module operatable in an actual environment by the machine learning. A learning device according to one aspect of the present invention executes machine learning of an extractor by using a first learning data set constituted by a combination of simulation data and first environmental information and a second learning data set constituted by a combination of actual data and second environmental information. Further, a learning device according to one aspect of the present invention executes machine learning of a controller by using a third learning data set constituted by a combination of third environmental information, state information, and a control command.

Abstract: Provided are a simulation device, a simulation method, and a computer-readable storage medium, which reduce a computation load required for simulation of movement of an operation subject. This simulation device is provided with: a first setting unit that sets framing conditions for a model representing the operation subject; a second setting unit that sets conditions for external force applied to the operation subject; a first simulation unit that simulates the movement of the operation subject under the framing conditions and the conditions for external force; a generation unit that generates learning data; and a learning unit that, by supervised learning using the learning data, generates a learning model that takes the framing conditions, the conditions for external force, and the initial conditions of the plurality of representative points as input and outputs data representing the movement of the plurality of representative points.

Abstract: MRI-guided robotics offers possibility for physicians to perform interventions remotely on confined anatomy. While the pathological and physiological changes could be visualized by high-contrast volumetric MRI scan during the procedure, robots promise improved navigation with added dexterity and precision. In cardiac catheterization, however, maneuvering a long catheter (1-2 meters) to the desired location and performing the therapy are still challenging. To meet this challenge, this invention presents an MRI-conditional catheter robotic system that integrates intra-op MRI, MR-based tracking units and enhanced visual guidance with catheter manipulation. This system differs fundamentally from existing master/slave catheter manipulation systems, of which the robotic manipulation is still challenging due to the very limited image guidance. This system provides a means of integrating intra-operative MR imaging and tracking to improve the performance of tele-operated robotic catheterization.