Search

In recent years, fully convolutional networks (FCNs) have been applied to various medical image segmentation tasks. However, it is difficult to generate a large amount of high-quality annotation data to train FCNs for medical image segmentation. Thus, it is desired to achieve high segmentation performances even from incomplete training data. We aim to evaluate performance of FCNs to clean noises and interpolate labels from noisy and sparsely given label images. To evaluate the label cleaning and propagation performance of FCNs, we used 2D and 3D FCNs to perform volumetric brain segmentation from magnetic resonance image volumes, based on network training on incomplete training datasets from noisy and sparse annotation. The experimental results using pseudo-incomplete training data showed that both 2D and 3D FCNs could provide improved segmentation results from the incomplete training data, especially by using three orthogonal annotation images for network training. This paper presented a validation for label cleaning and propagation based on FCNs. FCNs might have the potential to achieve improved segmentation performances even from sparse annotation data including possible noises by manual annotation, which can be an important clue to more efficient annotation.

A gait training method that can adjust the target leg load is required to meet the varied demands of trainees. In this study, we propose a running training method using a soft gait-assistive suit—a power-assistive device that adds an external force to the leg of a trainee by actuating a pneumatic artificial muscle (PAM). Herein, the assisting force is applied to the leg of the trainee so that muscle activity is reduced throughout the gait cycle. We detected the terminal stance phase as an appropriate assist timing from the pressure information, and the detected timing was used for the PAM control. We conducted a verification experiment of the PAM pressure variation. In a running mode of 7.0 to 10.0 km/h, all data showed that the PAM at the calf front has better detectability than that at the calf back, which is used in conventional detection methods. The experimental results confirmed that the PAM for gait phase detection should be positioned at the calf front during running. The effectiveness of the proposed running training method was verified through experiments on four trainees using a treadmill. The leg electromyogram of the trainees was measured and changes observed. Accordingly, 14 out of 16 channels of the four trainees showed a significant difference, and the electromyograms decreased by the suit assistance in the range of −1.4 to − 80.3% maximum voluntary contraction. Thus, the experimental results quantitatively confirmed that the proposed running assistance method is effective in reducing muscular activity.

Considering the widespread use of surgical robots in recent years, computer-assisted surgery is becoming significantly popular. Because the automation of surgical tasks without human intervention remains complex owing to individual patient differences, a human cooperative control is proposed in this study. A system in which an operator manipulates one surgical instrument to insert a suture needle was developed, along with another surgical instrument that automatically pulls out the needle from the operated instrument. In the proposed method, YOLOv3 and a standard convolutional neural network (CNN) are used to estimate the penetration and pull state of the needle. An image-based state estimator classifies the state regardless of the stiffness of the object to which the suture needle is inserted. Furthermore, after the pull state is detected, despite a failure of the needle pulling, the position can be corrected and the automated surgical instrument can approach the needle again. Experiments on human cooperative control demonstrated the effectiveness of state estimation using the proposed method. In addition, the failure of grasping observed in a previous study caused by the needle angle error was reduced.

Various applications using pneumatic artificial muscles (PAMs) have been developed in recent years. When the pressure of the PAM is controlled, it is desirable to position a pressure sensor at the control port of a servo valve through a pipeline for improvement in usability, environment resistance, and elimination of mechanical complexity. However, the responsiveness of the pressure control worsens due to the volume and pressure losses through the pipeline. In this paper, we propose a pneumatic circuit model to compensate for such losses through the pipeline and a method to estimate the pressure at the PAM inlet. The proposed method improves the responsiveness of the pressure control by positioning a pressure sensor at the control port of the servo valve. We developed an experimental apparatus that simulates the mechanism of our power assist robot, and it was used for performance evaluation experiments of the conventional and proposed pressure control methods. When a sinusoidal pressure target value with a frequency of 1.5 Hz and load mass of 8 kg was fed as input, the errors between the measured and target values in the conventional and proposed control were approximately 45 and 20 kPa, respectively. The reduced error confirmed that the responsiveness of the pressure control was improved by the proposed method.

This paper reports the precision of shape-from-focus (SFF) imaging according to the texture frequencies and window sizes of a focus measure. SFF is one of various depth measurement techniques for optical imaging, such as microscopy and endoscopy. SFF measures the depth of an object according to focus measure, which is generally computed with a fixed window. The window size affects the performance of SFF and should be adjusted for the texture of an object. In this study, we investigated the precision difference of SFF in texture frequencies and by window size. Two experiments were performed: precision validation in texture frequencies with a fixed window size, and precision validation in various window sizes related to pixel-cycle lengths. The first experimental results showed that a smaller window size could not provide a correct focus measure, and the second results showed that a window size that is approximately equal to a pixel-cycle length of the texture could provide better precision. These findings could potentially contribute to determining the appropriate window size of focus measure operation in shape-from-focus reconstruction.

Brain structure segmentation on magnetic resonance (MR) images is important for various clinical applications. It has been automatically performed by using fully convolutional networks. However, it suffers from the class imbalance problem. To address this problem, we investigated how loss weighting strategies work for brain structure segmentation tasks with different class imbalance situations on MR images. In this study, we adopted segmentation tasks of the cerebrum, cerebellum, brainstem, and blood vessels from MR cisternography and angiography images as the target segmentation tasks. We used a U-net architecture with cross-entropy and Dice loss functions as a baseline and evaluated the effect of the following loss weighting strategies: inverse frequency weighting, median inverse frequency weighting, focal weighting, distance map-based weighting, and distance penalty term-based weighting. In the experiments, the Dice loss function with focal weighting showed the best performance and had a high average Dice score of 92.8% in the binary-class segmentation tasks, while the cross-entropy loss functions with distance map-based weighting achieved the Dice score of up to 93.1% in the multi-class segmentation tasks. The results suggested that the distance map-based and the focal weightings could boost the performance of cross-entropy and Dice loss functions in class imbalanced segmentation tasks, respectively.

In this paper, a forceps manipulator using a pneumatic soft actuator for a flexion joint of forceps tip is developed. Many forceps manipulators are driven by wires, however there are problems such as friction between wires and parts and interference between wires. Therefore, this problem is solved by using the soft actuator, which performs bending of two degrees of freedom joint driven by air pressure instead of wires. The soft actuator is made of a silicone rubber and reinforced by two metal springs. It is easy to fabricate the soft actuator by using a metal spring. The outer spring restrains the radial expansion of the soft actuator, and the inner spring is placed in the center of the soft actuator to reinforce the bending stiffness. The actuator has four chambers, and it is possible to bend the soft actuator by inputting air into each chamber. An air gripper is mounted on the tip of the soft actuator. The bending angle of the forceps is estimated by the pressure in the chambers. The total length of the developed forceps is 340 mm, the weight is 23 g, and the maximum bending angle is 53 deg. By using the soft actuator for the bending joint itself, a simple and lightweight forceps manipulator with no driving unit at the rear end is realized.

In this paper, we propose a three-fingered end-effector driven by pneumatic soft actuators. The proposed end-effector has three thin sticks as the finger, and grips a food with three points to achieve force-closure. Since the sticks have a small diameter, the proposed end-effector will be able to achieve the safely picking from a food basketwithout breaking the foods. In addition, by using the soft actuator, it is possible to control the gripping force softly. We developed a curved extension type soft actuator for the rotation motion of the proposed end-effector. In addition, we propose a disturbance observer that estimates an external force applied to the stick tip from the food to realize the human-like soft touch. To evaluate the performance of the proposed end-effector, we carried out the three type experiments: an angle control test, an external force estimation, and a food gripping test. As a result, the measured angles sufficiently tracked the given target trajectories, and the proposed external force estimation was effective with its accuracy about 0.1 N. Finally, we confirmed that the proposed end-effector succeeded to grip four different type foods stably.

Tunable stiffness mechanisms can increase the noninvasiveness and stability of organ manipulation in laparoscopic liver resection. We have developed an organ-grasping device using beam-shaped tunable stiffness mechanism. Increasing the change ratio of stiffness will improve the performance of the device by offering high flexibility when adhering to the liver surface and high rigidity during the manipulation of the liver, however, optimal design of the beam has not been investigated. In this study, we investigate the wavy structure shape of the device that enhances the change in the ratio of stiffness. To increase the stiffness in a high-stiffness state, we used principal stress lines in the device to design the edge curve of the wavy shape material in the beams. We also investigated the arrangement of the wavy shape to decrease the stiffness in a low-stiffness state. Simulation using finite element method showed that the change ratio of stiffness was improved up to 13.0 by the new wavy shape arranged with the uniformly thick bottom of the waves.

In this study, we developed a selective driving joint forceps (SDJF) for laparoscopic surgery. The SDJF has a mechanism that the driving joints can be selected arbitrarily, therefore each joint doesn’t require an individual actuator for operating. The developed SDJF has six joints that can be operated using only four actuators. Each joint has 2-degrees-of-freedom (DOF) of flexion. Therefore, the SDJF has the same working area as the forceps having six driving joints (each joints can bend ±30° around the X and Y axes). The mechanism of the SDJF is realized by fixing each joint with a collar made of shape memory polymer. The proposed mechanism not only reduces the number of actuators required for joint operation, but also has the rigidity of the forceps, which is important in surgery. In addition, a driving section of the forceps is driven by pneumatic cylinders, therefore, the forceps joint has high-back-drivability, lightweight and high-output. We measured the heating and cooling time required to change the driving joint, dynamic response and rigidity of the prototype SDJF.

Background In recent years, there has been significant developments in surgical robots. Image‐based sensing of surgical instruments, without the use of electric sensors, are preferred for easily washable robots. Methods We propose a method to estimate the three‐dimensional posture of the tip of the forceps tip by using an endoscopic image. A convolutional neural network (CNN) receives the image of the tracked markers attached to the forceps as an input and outputs the posture of the forceps. Results The posture estimation results showed that the posture estimated from the image followed the electrical sensor. The estimated results of the external force calculated based on the posture also followed the measured values. Conclusion The method which estimates the forceps posture from the image using CNN is effective. The mean absolute error of the estimated external force is smaller than the human detection limit.

In recent years, fully convolutional networks (FCNs) have been applied to various medical image segmentation tasks. However, it is difficult to generate a large amount of high-quality annotation data to train FCNs for medical image segmentation. Thus, it is desired to achieve high segmentation performances even from incomplete training data. We aim to evaluate performance of FCNs to clean noises and interpolate labels from noisy and sparsely given label images.To evaluate the label cleaning and propagation performance of FCNs, we used 2D and 3D FCNs to perform volumetric brain segmentation from magnetic resonance image volumes, based on network training on incomplete training datasets from noisy and sparse annotation.The experimental results using pseudo-incomplete training data showed that both 2D and 3D FCNs could provide improved segmentation results from the incomplete training data, especially by using three orthogonal annotation images for network training.This paper presented a validation for label cleaning and propagation based on FCNs. FCNs might have the potential to achieve improved segmentation performances even from sparse annotation data including possible noises by manual annotation, which can be an important clue to more efficient annotation.

In this paper, we analyzed the stability of a wave-variables-based bilateral control system driven by pneumatic actuators. A pneumatic system is generally a third-order lag system due to the compressibility of gas. Frequency domain passivity of the system is evaluated using scattering matrix. The passivity is numerically confirmed when an appropriate control gain is given. The pneumatic system is nonlinear and can be unstable if the stability analysis is based on frequency domain passivity condition. The simulation results show that the pneumatic bilateral control system is passive.

In this study, we propose a small rotary actuator mounted on the tip of a forceps manipulator for laparoscopic surgery. The proposed actuator comprises a soft actuator (SA) and a cylindrical cam mechanism that converts linear motion into rotary motion. The SA is made from silicone and driven pneumatically. Therefore, it is lightweight, inexpensive, and highly biocompatible. Using an SA, the entire rotary actuator can be made lighter and smaller. In addition, the actuator has the advantage of easy sterilization and disposability due to the absence of electrical parts in it. Furthermore, using air as the power source, one can control the rotation angle with only a simple control that is not affected by the forceps posture. In this study, we developed a prototype of the SA and measured its basic characteristics. Subsequently, we mounted the SA on the proposed rotary actuator and confirmed the basic characteristics of the actuator through cyclic motion and force measurement experiments. As a result of the experiments, a rotational angle of 70∘ or more was measured and the target motion range was achieved. In addition, we confirmed that the rotary actuator exerted a maximum torque of 3.0 N ⋅ mm, which was sufficient for the needling task on a silicone seat.

Respiratory monitoring is a significant issue to reduce patient risks and medical staff labor in postoperative care and epidemic infection, particularly after the COVID-19 pandemic. Oximetry is widely used for respiration monitoring in the clinic, but it sometimes fails to capture a low-functional respiratory condition even though a patient has breathing difficulty. Another approach is breathing-sound monitoring, but this is unstable due to the indirect measurement of lung volume. Kobayashi in our team is developing a sensor measuring temporal changes in lung volume with a displacement sensor attached across the sixth and eighth ribs. For processing these respiratory signals, we propose the combination of complex-valued wavelet transform and the correlation among spectrum sequences. We present the processing results and discuss its feasibility to detect a low-functional condition in respiration. The result for detecting low-functional respiration showed good performance with a sensitivity of 0.88 and specificity of 0.88 to 1 in its receiver operating characteristic (ROC) curve.

The synchronized activity of neuronal populations across multiple distant brain areas may reflect coordinated interactions of large-scale brain networks. Currently, there is no established method to investigate the temporal transitions between these large-scale networks that would allow, for example, to decode finger movements. Here we applied a matrix factorization method employing principal component and temporal independent component analyses to identify brain activity synchronizations. In accordance with previous studies investigating “muscle synergies”, we refer to this activity as “brain activity synergy”. Using electroencephalography (EEG), we first estimated cortical current sources (CSs) and then identified brain activity synergies within the estimated CS signals. A decoding analysis for finger movement in eight directions showed that such CS synergies provided more information for dissociating between movements than EEG sensor signals, EEG synergy, or CS signals, suggesting that temporal activation patterns of the synchronizing CSs may contain information related to motor control. A quantitative analysis of features selected by the decoders further revealed temporal transitions among the primary motor area, dorsal and ventral premotor areas, pre-supplementary motor area, and supplementary motor area, which may reflect transitions in motor planning and execution. These results provide a proof of concept for brain activity synergy estimation using CSs.

A method to improve quality of life of elderly people is to increase muscle strength. The purpose of this study is to develop a machine to improve muscle strength effectively. In this paper, we proposed a muscle strength training machine that exerts variable resistance using pneumatic artificial rubber muscle. In addition, we observed the effect on muscle activity by using the proposed machine with predictable and unpredictable load change patterns. Experimental results showed that there was a difference between the rising time of muscle activity and the maximum amplitude of electromyogram depending on predictability of the load change and showed the possibility of effective muscle strength improvement against unpredictable load using the arm curl machine.

In this study, a robotic forceps with a wrist joint using visual SLAM for joint angle sensing was developed. The forceps has a flexible joint connected to the wrist joint at its rear end and the motion of the rear joint is driven by a parallel linkage. A monocular camera attached on the rear of the parallel linkage is in charge of position sensing, and the joint angles are estimated from the pose of the camera. The pose of the camera is obtained by a visual SLAM. The visual servo system realizes a simple attaching mechanism. The static and dynamic positioning experiments are conducted. We confirmed that the visual servoing system controls the forceps tip within the error of 3 deg in the motion range of 50 deg.