Abstract: A method of training neural network for obtaining medical sagittal image includes: using a first neural network on a 3-D medical image to generate a prediction sagittal mask; generating a prediction result according to the 3-D medical image and the prediction sagittal mask; generating a ground truth result according to the 3-D medical image and a ground truth sagittal mask; using a second neural network on the prediction result and the ground truth result; generating a loss function data according to an output of the second neural network; and adjusting parameters of the first neural network or the second neural network according to the loss function data.

Abstract: A wearable image display device for surgery includes a display unit, a wireless receiver and a processing core. The wireless receiver wirelessly receives a medical image or a medical instrument information in real-time. The processing core is coupled to the wireless receiver and the display unit for displaying the medical image or the medical instrument information on the display unit.

Abstract: An optical tracking system for a medical equipment includes optical markers, optical sensors and a computing device. The optical markers are disposed on the medical equipment. The optical sensors optically sense the optical markers to respectively generate sensing signals. The computing device is coupled to the optical sensors for receiving the sensing signals, and comprises a surgical situation 3-D model. The computing device is configured to adjust a relative position between a virtual medical equipment object and a virtual surgical target object in the surgical situation 3-D model according to the sensing signals.

Abstract: A method of training neural network for obtaining medical sagittal image includes: using a first neural network on a 3-D medical image to generate a prediction sagittal mask; generating a prediction result according to the 3-D medical image and the prediction sagittal mask; generating a ground truth result according to the 3-D medical image and a ground truth sagittal mask; using a second neural network on the prediction result and the ground truth result; generating a loss function data according to an output of the second neural network; and adjusting parameters of the first neural network or the second neural network according to the loss function data.

Abstract: A hand rehabilitation device includes a base having an arm support. A first actuator is mounted to the base and includes a first output end. A second actuator is connected to the first output end of the first actuator and includes a second output end. A handle is connected to the second output end. One of the first and second output ends rotates about an axis. The other of the first and second output ends reciprocates in a direction.

Abstract: A hand rehabilitation device includes a base having an arm support. A first actuator is mounted to the base and includes a first output end. A second actuator is connected to the first output end of the first actuator and includes a second output end. A handle is connected to the second output end. One of the first and second output ends rotates about an axis. The other of the first and second output ends reciprocates in a direction.

Abstract: An electronic deodorization trash receptacle includes a shell, a lid, a removable sterilization device and a baffle. The baffle is provided between the shell and lid. The removable sterilization device may be sterilization lamps, with the ultraviolet radiation to provide sterilization and deodorization effects. The baffle can protect users from harm of ultraviolet radiation.