In recent years, the field of robots for medical applications has been expanding rapidly. Robots effectively augment their operators’ skills, enabling them to achieve accuracy and high precision during complex procedures. The use of robots improves the quality of life of patients and the quality of medical research. Therefore, the research and development of robots for medical applications will become more active in aging societies. This special issue focuses on the design and control of robots as well as integrated technologies for robots for medical applications. These include navigation, simulator, image guidance, training, and validation technologies for robots.

The special issue consists of 17 papers with various studies related to medical robots. There are 7 papers on assistant robots, including their passive and active controls, devices, and sensors. There are 10 papers related to minimally invasive surgery and neurosurgery involving robots, including papers on sensors, actuators, navigation, haptic display devices, the mechanical design of devices, and other topics. The editors are confident that this special issue will greatly contribute to further progress in robotics

We sincerely thank the authors for their fine contributions and the reviewers for their generous contributions of time and effort. We would also like to thank the Editorial Board of the Journal of Robotics and Mechatronics for their help with this special issue.

Intraoperative tool change is a time-consuming and labor-intensive task for robot-assisted laparoscopic surgery. Serial multi-DOF manipulators are potential devices for realizing automatic intraoperative tool changes because of the layout flexibility and motion range, and multi-DOF makes it feasible for the manipulator to access and fetch the surgical tools by itself. However, the direction of the trocar may change because of the soft abdomen, and the lack of a fixed RCM makes it difficult for manipulators to reinsert a new surgical tool through the trocar. This study proposes a system prototype using a 7-DOF manipulator to automatically conduct the intraoperative tool-changing task. The newly designed surgical tool docking station facilitates surgical tool coupling/decoupling by rotating the manipulator’s end effector once. The proposed trocar recognition method with position error compensation is reliable for aligning a new surgical tool to the trocar port, even when the direction of the trocar is changed. The experimental results confirms that the manipulator can accomplish an intraoperative tool changing task without additional assistance or correction from the human.

Highly rigid surgical robots are capable of precise positioning; however, there is a risk of injury to the surrounding organs owing to undesired contact. To solve this problem, surgeons can change their stiffness according to the desired motion by contracting and relaxing the muscles. Therefore, surgical robots that can change their stiffness according to their application, similar to a surgeon, are useful in improving safety. However, existing variable stiffness actuators cannot easily achieve a wide variable stiffness range while maintaining a small size and lightweight, which are critical factors for surgical robots. This study presents the design, fabrication, and evaluation of a variable stiffness actuator that is compact and provides a wide range of variable stiffness, with elastic elements arranged in a circumferential direction.

Several micro-robotic medical tools, including catheters, guide wires, and endoscopes, employing shape memory alloy (SMA) actuators have been proposed and developed for use in the human body. This paper describes the basic principle of SMAs and the characteristics of several mechanisms, such as unidirectional bending, multi-directional bending, torsional movement, extension, and stiffness control. Moreover, some medical applications, such as insertion assistance and endoscopic and laser positioning, are described.

Micro-anastomosis is considered to be a difficult task even for skilled surgeons. Our group has developed a surgical robotic system to assist surgeons. Going further, the detection of surgically relevant objects in the microscopic view is indispensable for the automation or semi-automation of the system. This paper proposes a novel surgical thread detector inspired by an automatic crack detection method. The proposed method achieved a Dice score of 76.30% and an intersection over union (IOU) of 66.08% at 34.50 fps.

Laparoscopic surgery is a minimally invasive procedure that is performed by viewing endoscopic camera images. However, the limited field of view of endoscopic cameras makes laparoscopic surgery difficult. To provide more visual information during laparoscopic surgeries, augmented reality (AR) surgical navigation systems have been developed to visualize the positional relationship between the surgical field and organs based on preoperative medical images of a patient. However, since earlier studies used preoperative medical images, the navigation became inaccurate as the surgery progressed because the organs were displaced and deformed during surgery. To solve this problem, we propose a mixed reality (MR) surgery navigation system in which surgical instruments are tracked by a motion capture (Mocap) system; we also evaluated the contact between the instruments and organs and simulated and visualized the deformation of the organ caused by the contact. This paper describes a method for the numerical calculation of the deformation of a soft body. Then, the basic technology of MR and projection mapping is presented for MR surgical navigation. The accuracy of the simulated and visualized deformations is evaluated through basic experiments using a soft rectangular cuboid object.

A magnetorheological (MR) fluid is a composite material comprising ferromagnetic particles, medium oils, and several types of additives. We developed an MR fluid clutch for haptics (H-MRC) and installed it in a haptic interface that simulates teleoperation endoscopic surgery (ES). To enhance its operability, we redesigned the H-MRC to reduce its weight and improve its control system. We reduced the weight of the H-MRC and haptic gripper by 77.0 g and 137.0 g, respectively. To evaluate the influence of the improvement and force feedback functions on remote operation skills, we conducted pick-and-place tests with a remotely controlled system. In the tests, we subjectively evaluated the NASA-TLX and quantitatively evaluated the success rate of the task. The results of the subjective assessment showed significant reductions in mental stress during the teleoperation task. In addition, the results of the quantitative evaluation showed that the force feedback function was effective against the teleoperation skills of the operators.

Single-port water-filled laparo-endoscopic surgery (WaFLES) is a surgical procedure used for treatment in an environment filled with isotonic water in the abdominal cavity under a single-port condition. In this study, we developed two leader controllers for the forceps manipulator and retractor to generate and maintain a surgical workspace for a single-port WaFLES support robot. The development of the specific controller for each device increased the operation time and complicated the motion, such as regripping. We integrated the two functions as a controller to prevent the problem above. We performed grasping and retracting tasks in the virtual surgical workspace to evaluate the proposed controller. Based on the experimental results, we clarified the effect on the operation time by a different mechanism and observed that arranging the switch decreased the operation time. In addition, one of the proposed leader controllers improved operability in terms of operation time during selection and switching from the retractor to the forceps manipulator. However, the arrangement of the switch could adversely affect controller operability when switching from a simple operation (requiring only position control during retractor operation) to a complex operation (requiring both position and posture control during forceps operation). Furthermore, manipulation errors were observed using either of the proposed controllers. Therefore, the sensing procedure of the controller should be improved by addressing these errors in software and hardware.

Introduction of surgical simulators, which enable repeated learning of new surgical techniques, is advancing and they are desired in the field of neurosurgery. This study aims to make a two-fold contribution. First is the development of a haptic interface, which can be used while changing the operative tools necessary for training the cerebral fissure opening technique while using both the hands. Second is to develop a module for the haptic interface, which can display the cutting force when using micro scissors. To realize the operation with both the hands, the haptic interfaces for the right and left hands are designed so that they do not interfere. In addition, surgical tools, such as retractors, micro dissectors, and micro scissors, can be exchanged. In the cutting experiment carried out prior to the development of the haptic interface, it was clarified that the force when the dura mater was cut using micro scissors was 0.5 N. For comparison, the cutting forces required to cut two and three sheets of paper were measured to be 0.4 N and 0.6 N, respectively. The developed micro scissors module was designed using one motor and planetary gear mechanism. The gear mechanism is designed such that the right and left handles rotate in reverse directions around the rotation axis of the micro scissors using only one motor. This mechanism enables the micro scissors to cut the virtual tissues in the middle of the blade. The developed module could display a force of 0.4 N.

Neuroendovascular catheterization using fluoroscopy poses the problem to operators and staffs of cumulative radiation exposure. To solve this problem, we are developing a remote-controlled master-slave robot. Because a wire-like elongated treatment device is inserted into a blood vessel using a catheter, the robot requires a sensor to detect the insertion force of the wire. The proposed sensor is integrated into a robot installed in an X-ray fluoroscopy room that is remotely controlled from another room. The features of this sensor include measurement of the insertion force with sufficient accuracy, simple wire attachment, and an inexpensive disposable sensor head, rendering it very suitable for practical application. In this paper, we report on these features, as well as the results of a practical test of the sensor using a cerebrovascular model.

Neuronavigation is a computer-assisted technique for presenting three-dimensional images of a patient’s brain to facilitate immediate and precise lesion localization by surgeons. Neuronavigation systems use preoperative medical images of patients. In neurosurgery, when the dura mater and arachnoid membrane are incised and the cerebrospinal fluid (CSF) drains out, the brain loses the CSF buoyancy and deforms in the direction of gravity, which is referred to as brain shift. This brain shift yields inaccurate neuronavigation. To reduce this inaccuracy, an intraoperative brain shift should be estimated. This paper proposes a dynamic simulation method for brain-shift estimation combining the moving-particle semi-implicit (MPS) method and the finite element method (FEM). The CSF was modeled using fluid particles, whereas the brain parenchyma was modeled using finite elements (FEs). Node particles were attached to the surface nodes of the brain parenchyma in the FE model. The interaction between the CSF and brain parenchyma was simulated using the repulsive force between the fluid particles and node particles. Validation experiments were performed using a gelatin block. The gelatin block was dipped into silicone oil, which was then gradually removed; the block deformation owing to the buoyancy loss was measured. The experimental deformation data were compared with the results of the MPS-FEM coupled analysis. The mean absolute error (MAE) between the simulated deformation and the average across the four experiments was 0.26 mm, while the mean absolute percentage error (MAPE) was 27.7%. Brain-shift simulations were performed using the MPS-FEM coupled analysis, and the computational cost was evaluated.

To prevent postoperative complications in corrective surgery for foot deformities such as hallux valgus and pes planus, it is critical to quantitatively predict the postoperative standing-position plantar pressure distribution during the operation. The authors have previously proposed an intraoperative plantar pressure measurement system (IPPM) that allows for the measurement of a supine patient’s plantar pressure distribution that is equivalent to that in the standing position. This system consists of an IPPM device comprising of a force plate and pressure distribution sensor, an optical three-dimensional position measurement device, a navigation monitor, and a PC. The plantar pressure distribution in the standing position is reproduced by navigating the operator, as he or she presses the IPPM device against the patient’s sole so that the weight-bearing axis (floor reaction force vector) and femoral head center are as close to each other as possible. However, in our previous study, the reproducibility of the standing position plantar pressure distribution was insufficient. Therefore, in the present study, we add a navigational function that can be used to bring the centers of pressure in the standing position and under measurement, as well as to correct the IPPM’s self-weight in the measured force. The improved device was used in an experiment with nine healthy subjects, and the similarity of the plantar pressure distribution in the standing and supine positions was evaluated using normalized cross-correlation, yielding an average of 0.90. Furthermore, in an evaluation experiment with ten orthopedic surgeons, it was observed that using the system reproduced the plantar pressure distribution significantly better than when the system was not used. These results indicate that the present system can predict the plantar pressure distribution in the standing position. We believe that this system can contribute to reducing complications after foot surgery.

In recent years, a rapid increase in bedridden elderly, due to cerebral strokes and other diseases, has become a social problem due to soaring medical costs and heavy burden of long term care support. Therefore, development of walk rehabilitation devices for the elderly, suffered from cerebral strokes, is strongly expected. In the research so far, the authors have developed a rollator with a chest support pad that can rotate freely with one degree of freedom, and the effectiveness of its walk assistance was evaluated. In this study, a novel rollator using a motor-driven chest support pad was developed. A compact system to combine data collect system and motor control was configured. Walk measurements for six subjects were conducted using the developed rollator. The relationship between the movements of the lower limbs and rotation of the motor driven chest support pad was investigated. Time series and power spectrum results were analyzed, and the effectiveness of the motor-driven chest support pad on movement of the lower limb during walking was discussed.

Greeting is important for socially assistive robot to smoothly initiate conversations with older adults. Because of their decreased cognitive function, older adults may occasionally be unaware of the presence of a robot. The purpose of this study was to investigate and evaluate the characteristic motion and utterance time for greeting older adults in comparison with those for greeting non-older adults. The motion and utterance for greeting a seated target imitating an older adult and greeting a non-older adult were measured. The utterance times of the greeting and motion parameters such as the maximum joint angles were calculated from the measured data. The parameters were compared using statistical methods. According to the results, the hip bending angle in older adults was 36.6° greater than in the non-older adults. The utterance lag for greeting the older adults was 0.7 s longer than that for greeting the non-older adults at this time. The impressions of the robot that greeted the participants based on the extracted motion parameters were compared to verify these parameter differences. Although the greeting styles did not differ significantly, it was verified that the robot’s greeting was more impressive than that of a computer-graphics robot.

This paper proposes a three-layer elastic cloth fabric mechanism for an assistive suit with adjustable structure (based on a two-layer non-adjustable structure) to achieve different assistive force profiles. This increases the assistive force on the lower-back muscle group and alleviates the undesired pre-tension that acts on a user when the rubber belt located on the back is pulled to provide a higher assistive force. With the lower pre-tension, users would not encounter body fatigue as rapidly as in the past. The adjustable feature enables the structure to provide a force that increases gradually to a high level over a short distance without pre-tension. An experiment involving the measurement of muscle activities is conducted to evaluate the variation in assistive force in the lower back by comparing the three-layer suit to the two-layer non-adjustable suit. The experimental results show that the new three-layer structure successfully assists without pre-tension in the lower-back muscle group similar to the two-layer structure with pre-tension. A simple questionnaire is also administered to collect feedback from participants on the differences between the three-layer suit and two-layer suit in terms of wearing perception. Over half of the participants reported that the perception of pre-tension in the three-layer suit is lower than that in the two-layer suit.

In this study, we investigated trainees’ adaptation by conducting static balance training in a tandem standing posture. The horizontal tensile force loads in the front, back, left, and right directions were applied using pneumatic artificial muscles. We analyzed the adaptation that occurred during training by changing the direction of the horizontal tensile load on the lumbar region according to the tendency of the trainee. We conducted the experiments using the following protocol. Ten trainees participated in the experiment. In Phase 1, we applied loads in four directions the same number of times in random order to investigate the weak direction in the balance of each trainee. In Phase 2, we measured five trainees in each group: Group 1 was trained in the same way as Phase 1, and Group 2 was intensively trained in two directions in which the balance found in Phase 1 was difficult to maintain. In Phase 3, we performed the same experiment as in Phase 1. We analyzed the adaptation of the trainees using the margin of stability (MoS), a balance evaluation index. We compared the experimental results of Phases 1 and 3. In Group 1, the tendency for improvement in balance was unclear. On the other hand, the balance index in Group 2 improved in four out of five trainees in both the front-back and left-right directions. These results suggest that the training method concentrating on the weak direction could provide a clear directionality to the training effect.

In response to the shortage, uneven distribution, and high cost of rehabilitation resources in the context of the COVID-19 pandemic, we developed a low-cost, easy-to-use remote rehabilitation system that allows patients to perform rehabilitation training and receive real-time guidance from doctors at home. The proposed system uses Azure Kinect to capture motions with an error of just 3% compared to professional motion capture systems. In addition, the system provides an automatic evaluation function of rehabilitation training, including evaluation of motion angles and trajectories. After acquiring the user’s 3D motions, the system synchronizes the 3D motions to the virtual human body model in Unity with an average error of less than 1%, which gives the user a more intuitive and interactive experience. After a series of evaluation experiments, we verified the usability, convenience, and high accuracy of the system, finally concluding that the system can be used in practical rehabilitation applications.

The world’s aging population is increasing. The number of elderly individuals having walking impairments is also increasing. Adequate exercise is becoming necessary for them. Therefore, several walking assistive devices have been developed or are under development. However, elderly individuals may have low motivation for exercising, or they may experience physical damage by excessive fatigue. This study proposed a method to enable elderly individuals to exercise with a positive emotion and prevent damage such as muscle fatigue. We proposed a 3D human condition model to control the walking assistive device. It includes the arousal, pleasure, and fatigue dimensions. With regard to the arousal and pleasure dimensions, we used heartbeat and electromyography (EEG) signals to train a deep neural network (DNN) model to identify human emotions. For fatigue detection, we proposed a method based on near-infrared spectroscopy (NIRS) to detect muscle fatigue. All the sensors are portable. This implies that it can be used for outdoor activities. Then, we proposed a walking strategy based on a 3D human condition model to control the walking assistive device. Finally, we tested the effectiveness of the automatic control system. The wearing of the walking assistive device and implementation of the walking strategy can delay the fatigue time by approximately 24% and increase the walking distance by approximately 16%. In addition, we succeeded in visualizing the distribution of emotion during each walking method variation. It was verified that the walking strategy can improve the mental condition of a user to a certain extent. These results showed the effectiveness of the proposed system. It could help elderlies maintain higher levels of motivation and prevent muscle damage by walking exercise, using the walking assistive device.

It remains challenging for robots to accurately perform sound source localization and speech recognition in a real environment with reverberation, noise, and the voices of multiple speakers. Accordingly, we propose “U-TasNet-Beam,” a speech extraction method for extracting only the target speaker’s voice from all ambient sounds in a real environment. U-TasNet-Beam is a neural beamformer comprising three elements: a neural network for removing reverberation and noise, a second neural network for separating the voices of multiple speakers, and a minimum variance distortionless response (MVDR) beamformer. Experiments with simulated data and recorded data show that the proposed U-TasNet-Beam can improve the accuracy of sound source localization and speech recognition in robots compared to the conventional methods in a noisy, reverberant, and multi-speaker environment. In addition, we propose the spatial correlation matrix loss (SCM loss) as a loss function for the neural network learning the spatial information of the sound. By using the SCM loss, we can improve the speech extraction performance of the neural beamformer.

Teaching pendants with multiple buttons are commonly employed to control working robots; however, such devices are not easy to operate. As an alternative, gesture-based manipulation methods using the operator’s upper limb movements have been studied as a way to operate the robots intuitively. Previous studies involving these methods have generally failed to consider the changing viewpoint of the operator relative to the robot, which may adversely affect operability. This study proposes a novel evaluation method and applies it in a series of experiments to compare the influence of viewpoint change on the operability of the gesture- and button-based operation methods. Experimental results indicate that the operability of the gesture-based method is superior to that of the button-based method for all viewpoint angles, due mainly to shorter non-operating times. An investigation of trial-and-error operation indicates that viewpoint change makes positional operation with the button-based method more difficult but has a relatively minor influence on postural operation.

The rubber band gripper is an automatic contraction ring that is used to wrap and fix objects of various shapes by vacuuming the inside of the ring. It can be used to handle objects as a robot gripper, as well as fix bodies to pedestals, and objects to bodies. This study geometrically analyzes object shapes that can be wrapped using the ring, without causing gaps in the gripping mechanism. The analysis and experimental results show that the object shapes that can be wrapped without gaps are determined by the maximum shrinkage rate after the ring contacts the object as well as the circumference of the object. The range of object shapes that can be wrapped without gaps narrows as the object position moves away from the center of the ring. However, this influence is small, except when the object is small or has a large aspect ratio. In particular, when the object shape is cylindrical, it can be wrapped without gaps regardless of its position within the ring.

To enhance the step climbing performance of omni-directional wheelchair, we developed a front-wheel-steering-drive dual-wheel caster drive mechanism with rocker links and a differential mechanism. The dual-wheel caster drive mechanism has the advantage of simple structure and is suitable for wheelchair applications. Based on the motion characteristics of this mechanism, we have found that the step climbing performance can be improved by adopting a front-wheel-steering-drive mechanism. In addition, the rocker links and the differential mechanism are employed as a suspension mechanism to improve wheel ground contact. In this study, a 3D dynamics simulator was constructed to compare the performance of wheelchairs employing several mechanisms, including the proposed mechanism, in step climbing. Based on the motion characteristics of the dual-wheel caster drive mechanism, simulations were carried out under two conditions: steady state and transient state. Simulation results confirm that the proposed mechanism has high step climbing performance under both conditions. Furthermore, it is confirmed that the rocker links and differential mechanism work to improve step climbing performance when the step has an angle, which means that the two front wheels do not contact the step at the same time.

Security is considered heavy labor work owing to night shifts and long working hours. In recent years, the number of security guards has been increasing, and the labor shortage for security guards has become a problem in Japan. To overcome these problems, robots have been used for security purposes. However, most are unable to interact with users or guards at night. In this study, a drone called aerial ubiquitous display (AUD) is proposed to improve the problems of existing security methods using robots and to provide night security. The AUD enables human-drone interaction and night security. Moreover, when an AUD interacts with humans, the drone must come close to the human and project information on the ground. Therefore, this study investigated the optimal parameters for a projector-equipped drone to approach a human at night. In addition, by comparing these results with those of the day approach, we verified whether there would be a change in perception between daytime and nighttime. Furthermore, an experiment was conducted to investigate the types of projections that are most likely to capture a user’s attention.

An association is observed between the standing sway posture and falls in patients with stroke; hence, it is important to study their standing balance. Although there are studies on the standing balance in stroke patients, differences in control have not been adequately investigated. This study aims to propose a method to characterize the postural sway in standing stroke patients using a mathematical model. A musculoskeletal model and neural controller model were used to simulate ten stroke patients (five patients with cerebral hemorrhages and five patients with cerebral infarctions) and eight young healthy participants, and their data were monitored during quiet standing. The model parameters were adjusted by focusing on the maximum-minimum difference in sway, which was considered important in a previous study, and sway speed, which is frequently used in the analysis. The adjusted model parameters were subjected to dimension reduction using non-negative matrix factorization. Consequently, the sway characteristics of stroke patients were expressed as the magnitude of gain parameters related to the extension of the entire body. The results of this study demonstrated the possibility of representing the characteristics of postural sway as model parameters in stroke patients using a mathematical model. This characterization could lead to the design of individualized rehabilitation systems in the future.