Pub Date : 2021-11-17DOI: 10.1109/ismr48346.2021.9661513
Mishek Musa, Saikat Sengupta, Yue Chen
In this work, the design, analysis, and characterization of a 6 DoF parallel robot for MRI guided applications is presented. The primary motivation for developing this robot is to create a general purpose robotic platform capable of producing accurate 6 DoF motion inside the MRI bore to perform needle-based interventional procedures (i.e., radio-frequency ablation, biopsy) or generate accurate motion for other MRI-based experiments (i.e., motion compensation imaging sequence development, HIFU probe manipulation). The robot is driven by 6 pneumatic cylinder actuators and controlled via a robust sliding mode controller. Pneumatic actuator tracking experiments indicate that the system is able to achieve an average error of 0.69 ± 0.14 mm and 0.67 ± 0.40 mm for step signal tracking and sinusoidal signal tracking respectively. To demonstrate the feasibility of the parallel robot for needle insertion interventions, a tissue-mimic phantom experiment was performed in the benchtop environment, which indicated an average position error of 1.20 0.43 mm and an average orientation error of 1.09 0.57°, respectively.
{"title":"Design of a 6 DoF Parallel Robot for MRI-guided Interventions","authors":"Mishek Musa, Saikat Sengupta, Yue Chen","doi":"10.1109/ismr48346.2021.9661513","DOIUrl":"https://doi.org/10.1109/ismr48346.2021.9661513","url":null,"abstract":"In this work, the design, analysis, and characterization of a 6 DoF parallel robot for MRI guided applications is presented. The primary motivation for developing this robot is to create a general purpose robotic platform capable of producing accurate 6 DoF motion inside the MRI bore to perform needle-based interventional procedures (i.e., radio-frequency ablation, biopsy) or generate accurate motion for other MRI-based experiments (i.e., motion compensation imaging sequence development, HIFU probe manipulation). The robot is driven by 6 pneumatic cylinder actuators and controlled via a robust sliding mode controller. Pneumatic actuator tracking experiments indicate that the system is able to achieve an average error of 0.69 ± 0.14 mm and 0.67 ± 0.40 mm for step signal tracking and sinusoidal signal tracking respectively. To demonstrate the feasibility of the parallel robot for needle insertion interventions, a tissue-mimic phantom experiment was performed in the benchtop environment, which indicated an average position error of 1.20 0.43 mm and an average orientation error of 1.09 0.57°, respectively.","PeriodicalId":405817,"journal":{"name":"2021 International Symposium on Medical Robotics (ISMR)","volume":"142 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116564261","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-11-17DOI: 10.1109/ismr48346.2021.9661502
Alex Tinggaard Årsvold, Andreas Sørensen Zeltner, Zhuoqi Cheng, K. Schwaner, Pernille Tine Jensen, T. Savarimuthu
Lymphadenectomy is frequently performed as a surgical treatment for cancer. Lymph nodes grow inside fat and have similar color as fat, making them difficult to detect. In Robotic Assisted Minimally Invasive Surgery (RAMIS), it can be even more challenging due to the lack of haptic feedback. This study proposes a novel method to measure the electrical property of a target tissue site and determine whether a lymph node is present underneath through an Artificial Neural Network classifier. The proposed system and method are built, analyzed, and evaluated based on simulation and ex vivo tissue phantom experiments. The experimental results show a very high accuracy (93.49%) in detecting a lymph node that is embedded deep inside fat. Given the promising results and the portability of the proposed system, we believe it has great potential to improve the quality of related surgical procedures.
{"title":"Lymph Node Detection Using Robot Assisted Electrical Impedance Scanning and an Artificial Neural Network","authors":"Alex Tinggaard Årsvold, Andreas Sørensen Zeltner, Zhuoqi Cheng, K. Schwaner, Pernille Tine Jensen, T. Savarimuthu","doi":"10.1109/ismr48346.2021.9661502","DOIUrl":"https://doi.org/10.1109/ismr48346.2021.9661502","url":null,"abstract":"Lymphadenectomy is frequently performed as a surgical treatment for cancer. Lymph nodes grow inside fat and have similar color as fat, making them difficult to detect. In Robotic Assisted Minimally Invasive Surgery (RAMIS), it can be even more challenging due to the lack of haptic feedback. This study proposes a novel method to measure the electrical property of a target tissue site and determine whether a lymph node is present underneath through an Artificial Neural Network classifier. The proposed system and method are built, analyzed, and evaluated based on simulation and ex vivo tissue phantom experiments. The experimental results show a very high accuracy (93.49%) in detecting a lymph node that is embedded deep inside fat. Given the promising results and the portability of the proposed system, we believe it has great potential to improve the quality of related surgical procedures.","PeriodicalId":405817,"journal":{"name":"2021 International Symposium on Medical Robotics (ISMR)","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114107094","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-11-17DOI: 10.1109/ismr48346.2021.9661511
Zhen Li, M. Mancini, G. Monizzi, D. Andreini, G. Ferrigno, J. Dankelman, E. Momi
Cardiologists highlight the need for an intra-operative 3D visualization to assist interventions. The intra-operative 2D X-ray/Digital Subtraction Angiography (DSA) images in the standard clinical workflow limit cardiologists’ views significantly. Compared with image-to-image registration, model-to-image registration is an essential approach taking advantage of the reuse of pre-operative 3D models reconstructed from Computed Tomography Angiography (CTA) images. Traditional optimized-based registration methods suffer severely from high computational complexity. Moreover, the consequence of lacking ground truth for learning-based registration approaches should not be neglected. To overcome these challenges, we introduce a model-to-image registration framework via deep learning for image-guided endovascular catheterization. This work performs autonomous vessel segmentation from intra-operative fluoroscopy images via a deep residual U-net and a model-to-image matching via a convolutional neural network. For this study, image data were collected from 10 patients who performed Transcatheter Aortic Valve Implantation (TAVI) procedures. It was found that vessel segmentation of test data results in median values of Dice Similarity Coefficient, Precision, and Recall of (0.75, 0.58, 0.67) for femoral artery, and (0.71, 0.56, 0.74) for aortic root. The segmentation network behaves better than manual annotation, and it recognizes part of vessels that were not labeled manually. Image matching between the transformed moving image and the fixed image results in a median value of Recall of 0.90. The proposed approach achieves a good accuracy of vessel segmentation and a good recall value of model-to-image matching.
{"title":"Model-to-Image Registration via Deep Learning towards Image-Guided Endovascular Interventions","authors":"Zhen Li, M. Mancini, G. Monizzi, D. Andreini, G. Ferrigno, J. Dankelman, E. Momi","doi":"10.1109/ismr48346.2021.9661511","DOIUrl":"https://doi.org/10.1109/ismr48346.2021.9661511","url":null,"abstract":"Cardiologists highlight the need for an intra-operative 3D visualization to assist interventions. The intra-operative 2D X-ray/Digital Subtraction Angiography (DSA) images in the standard clinical workflow limit cardiologists’ views significantly. Compared with image-to-image registration, model-to-image registration is an essential approach taking advantage of the reuse of pre-operative 3D models reconstructed from Computed Tomography Angiography (CTA) images. Traditional optimized-based registration methods suffer severely from high computational complexity. Moreover, the consequence of lacking ground truth for learning-based registration approaches should not be neglected. To overcome these challenges, we introduce a model-to-image registration framework via deep learning for image-guided endovascular catheterization. This work performs autonomous vessel segmentation from intra-operative fluoroscopy images via a deep residual U-net and a model-to-image matching via a convolutional neural network. For this study, image data were collected from 10 patients who performed Transcatheter Aortic Valve Implantation (TAVI) procedures. It was found that vessel segmentation of test data results in median values of Dice Similarity Coefficient, Precision, and Recall of (0.75, 0.58, 0.67) for femoral artery, and (0.71, 0.56, 0.74) for aortic root. The segmentation network behaves better than manual annotation, and it recognizes part of vessels that were not labeled manually. Image matching between the transformed moving image and the fixed image results in a median value of Recall of 0.90. The proposed approach achieves a good accuracy of vessel segmentation and a good recall value of model-to-image matching.","PeriodicalId":405817,"journal":{"name":"2021 International Symposium on Medical Robotics (ISMR)","volume":"12 Suppl 8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126053316","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-11-17DOI: 10.1109/ismr48346.2021.9661492
Thulani Tsabedze, Erik Hartman, Cianan Brennan, Jun Zhang
Robotic rehabilitation is advantageous as it allows for dynamic exercise routines that are more accurate and efficient than human-led routines. However, existing robotic wrist rehabilitation devices are often bulky and are tethered in rehabilitation centers. It is difficult to realize compact wearable wrist orthosis that is capable of inducing three degrees of freedom (DOF) of the wrist. This paper presents the first wearable wrist orthosis with three DOF driven by twisted string actuators (TSAs) using both stiff fishing lines and compliant super-coiled polymer (SCP) strings. The design considerations of the robotic wrist orthosis are provided in detail. Experiments are performed to characterize the compliance and force production of TSAs. The device is capable of inducing pronation or supination, flexion or extension, and abduction or adduction, with a range of 117.9°, 115.5°, and 73.4° respectively. In addition, it is demonstrated that the device can fully induce wrist movement of a human subject without the need of activating the human subject’s muscles.
{"title":"A Compliant Robotic Wrist Orthosis Driven by Twisted String Actuators","authors":"Thulani Tsabedze, Erik Hartman, Cianan Brennan, Jun Zhang","doi":"10.1109/ismr48346.2021.9661492","DOIUrl":"https://doi.org/10.1109/ismr48346.2021.9661492","url":null,"abstract":"Robotic rehabilitation is advantageous as it allows for dynamic exercise routines that are more accurate and efficient than human-led routines. However, existing robotic wrist rehabilitation devices are often bulky and are tethered in rehabilitation centers. It is difficult to realize compact wearable wrist orthosis that is capable of inducing three degrees of freedom (DOF) of the wrist. This paper presents the first wearable wrist orthosis with three DOF driven by twisted string actuators (TSAs) using both stiff fishing lines and compliant super-coiled polymer (SCP) strings. The design considerations of the robotic wrist orthosis are provided in detail. Experiments are performed to characterize the compliance and force production of TSAs. The device is capable of inducing pronation or supination, flexion or extension, and abduction or adduction, with a range of 117.9°, 115.5°, and 73.4° respectively. In addition, it is demonstrated that the device can fully induce wrist movement of a human subject without the need of activating the human subject’s muscles.","PeriodicalId":405817,"journal":{"name":"2021 International Symposium on Medical Robotics (ISMR)","volume":"67 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127118983","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-11-17DOI: 10.1109/ismr48346.2021.9661545
Naghmeh Zamani, Ashkan Pourkand, Heather Culbertson, David I. Grow
This paper presents the design of a 6-DOF hybrid impedance/admittance haptic device that targets simulation of bone drilling and related tasks. We present a prototype optimized for applications like bone drilling, cutting bone, spinal awl probe use, and other surgical techniques where a combination of high force and low impedance is required in different directions. The required performance cannot be satisfied by existing, off-the-shelf haptic devices. This design may allow critical improvements in simulator fidelity for surgery training. The device consists of two low-mass (carbon fiber) plates with a rod passing through them and constrained to move in only 2 DOF, allowing axial torque to be displayed to the user’s hand. These two parallel plates are controlled by four cables pulled by motors. We derive the forward kinematic equations and present the predicted distribution of location error, cable velocity, cable tension, and output force. These results and hardware tests indicate that this design may provide a revolutionary approach for the haptic display of many surgical procedures by means of an architecture that allows arbitrary workspace scaling.
{"title":"Plate-and-Cable (PAC) Haptic Device for Orthopaedic Training","authors":"Naghmeh Zamani, Ashkan Pourkand, Heather Culbertson, David I. Grow","doi":"10.1109/ismr48346.2021.9661545","DOIUrl":"https://doi.org/10.1109/ismr48346.2021.9661545","url":null,"abstract":"This paper presents the design of a 6-DOF hybrid impedance/admittance haptic device that targets simulation of bone drilling and related tasks. We present a prototype optimized for applications like bone drilling, cutting bone, spinal awl probe use, and other surgical techniques where a combination of high force and low impedance is required in different directions. The required performance cannot be satisfied by existing, off-the-shelf haptic devices. This design may allow critical improvements in simulator fidelity for surgery training. The device consists of two low-mass (carbon fiber) plates with a rod passing through them and constrained to move in only 2 DOF, allowing axial torque to be displayed to the user’s hand. These two parallel plates are controlled by four cables pulled by motors. We derive the forward kinematic equations and present the predicted distribution of location error, cable velocity, cable tension, and output force. These results and hardware tests indicate that this design may provide a revolutionary approach for the haptic display of many surgical procedures by means of an architecture that allows arbitrary workspace scaling.","PeriodicalId":405817,"journal":{"name":"2021 International Symposium on Medical Robotics (ISMR)","volume":"96 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127234223","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-11-17DOI: 10.1109/ismr48346.2021.9661520
Jack Kaplan, Yana Sosnovskaya, M. Arnold, B. Hannaford
Minimally Invasive Surgery lacks tactile feedback that surgeons find useful for finding and diagnosing tissue abnormalities. The goal of this paper is to calibrate sensors of a motorized Smart Grasper surgical instrument to provide accurate force and position measurements. These values serve two functions with the novel calibration hardware. The first is to control the motor of the Grasper to prevent tissue damage. The second is to act as the base upon which future work in multi-modal sensor fusion tissue characterization can be built. Our results show that the Grasper jaw distance is a function of both applied force and motor angle while the force the jaws apply to the tissue can be measured using the internal load cell. All code and data sets used to generate this paper can be found on GitHub at https://github.com/Yana-Sosnovskaya/Smart_Grasper_public
{"title":"Sensor Fusion for Force and Position Calibration of a Motorized Surgical Smart Grasper","authors":"Jack Kaplan, Yana Sosnovskaya, M. Arnold, B. Hannaford","doi":"10.1109/ismr48346.2021.9661520","DOIUrl":"https://doi.org/10.1109/ismr48346.2021.9661520","url":null,"abstract":"Minimally Invasive Surgery lacks tactile feedback that surgeons find useful for finding and diagnosing tissue abnormalities. The goal of this paper is to calibrate sensors of a motorized Smart Grasper surgical instrument to provide accurate force and position measurements. These values serve two functions with the novel calibration hardware. The first is to control the motor of the Grasper to prevent tissue damage. The second is to act as the base upon which future work in multi-modal sensor fusion tissue characterization can be built. Our results show that the Grasper jaw distance is a function of both applied force and motor angle while the force the jaws apply to the tissue can be measured using the internal load cell. All code and data sets used to generate this paper can be found on GitHub at https://github.com/Yana-Sosnovskaya/Smart_Grasper_public","PeriodicalId":405817,"journal":{"name":"2021 International Symposium on Medical Robotics (ISMR)","volume":"107 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133813582","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-11-17DOI: 10.1109/ismr48346.2021.9661539
K. Schwaner, Iñigo Iturrate, J. K. Andersen, C. R. Dam, Pernille Tine Jensen, T. Savarimuthu
MOPS is a platform for surgical robotics research. It consists of hardware and software components for building surgical robot systems based on existing robot arms. The reference system presented in this paper has two robot arms, each with an adapter for mounting and actuating different types of surgical instruments. Additionally, the system has a high-resolution stereo camera and a simple operator console with haptic devices and foot pedals for teleoperating the manipulators. Software components are based on Robot Operating System (ROS) to keep them modular and reusable for different types of hardware. We evaluate the precision of our system for vision-based tasks and demonstrate its potential for surgical task automation. The mean end-to-end error, including that of the vision system, for a trajectory following task was 1.2mm / 0.3° for one manipulator and 2.7mm / 0.7° for the other. MOPS software and hardware components are made available under an open-source license to encourage collaboration and facilitate the advancement of surgical robotics research.
{"title":"MOPS: A Modular and Open Platform for Surgical Robotics Research","authors":"K. Schwaner, Iñigo Iturrate, J. K. Andersen, C. R. Dam, Pernille Tine Jensen, T. Savarimuthu","doi":"10.1109/ismr48346.2021.9661539","DOIUrl":"https://doi.org/10.1109/ismr48346.2021.9661539","url":null,"abstract":"MOPS is a platform for surgical robotics research. It consists of hardware and software components for building surgical robot systems based on existing robot arms. The reference system presented in this paper has two robot arms, each with an adapter for mounting and actuating different types of surgical instruments. Additionally, the system has a high-resolution stereo camera and a simple operator console with haptic devices and foot pedals for teleoperating the manipulators. Software components are based on Robot Operating System (ROS) to keep them modular and reusable for different types of hardware. We evaluate the precision of our system for vision-based tasks and demonstrate its potential for surgical task automation. The mean end-to-end error, including that of the vision system, for a trajectory following task was 1.2mm / 0.3° for one manipulator and 2.7mm / 0.7° for the other. MOPS software and hardware components are made available under an open-source license to encourage collaboration and facilitate the advancement of surgical robotics research.","PeriodicalId":405817,"journal":{"name":"2021 International Symposium on Medical Robotics (ISMR)","volume":"44 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114072827","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-11-17DOI: 10.1109/ismr48346.2021.9661571
A. B. Halima, J. Bert, J.-F. Clément, D. Visvikis
Low-dose rate prostate brachytherapy is one of the most widely used radiotherapy techniques for early stage cancer treatment due to its high benefits and low side effects. This method consists in manually inserting small radioactive seeds within the prostate through the perineum under ultrasound image guidance. This implantation is often inaccurate leading to mis-distributed dosimetry. Different robotic devices were proposed to improve the accuracy of seeds implantation. However, most of them are not well suitable for clinical conditions, especially considering bulk and size. Thus, in this paper we developed a 6 degrees of freedom compact and lightweight co-manipulated robot for prostate brachytherapy easy to install in the operating room thanks to its parallel design. The proposed robotic architecture impacts the accuracy of seeds placement along the gravity axis. Therefore, a gear spring mechanism is added into each leg of the robotic system in order to minimize the gravitational torques. The spring stiffness required to compensate gravity forces was calculated by determining the kinematic model and establishing the approximate perfect balancing condition. The final robot fits in a cube of 300×300×300 mm3 and permits to the needle to reach any point within the prostate in a singularity-free workspace superior to 55×55×150 mm3 with an average torque reduction rate of 82%.
{"title":"Development of a 6 Degrees of Freedom Prostate Brachytherapy Robot with Integrated Gravity Compensation System","authors":"A. B. Halima, J. Bert, J.-F. Clément, D. Visvikis","doi":"10.1109/ismr48346.2021.9661571","DOIUrl":"https://doi.org/10.1109/ismr48346.2021.9661571","url":null,"abstract":"Low-dose rate prostate brachytherapy is one of the most widely used radiotherapy techniques for early stage cancer treatment due to its high benefits and low side effects. This method consists in manually inserting small radioactive seeds within the prostate through the perineum under ultrasound image guidance. This implantation is often inaccurate leading to mis-distributed dosimetry. Different robotic devices were proposed to improve the accuracy of seeds implantation. However, most of them are not well suitable for clinical conditions, especially considering bulk and size. Thus, in this paper we developed a 6 degrees of freedom compact and lightweight co-manipulated robot for prostate brachytherapy easy to install in the operating room thanks to its parallel design. The proposed robotic architecture impacts the accuracy of seeds placement along the gravity axis. Therefore, a gear spring mechanism is added into each leg of the robotic system in order to minimize the gravitational torques. The spring stiffness required to compensate gravity forces was calculated by determining the kinematic model and establishing the approximate perfect balancing condition. The final robot fits in a cube of 300×300×300 mm3 and permits to the needle to reach any point within the prostate in a singularity-free workspace superior to 55×55×150 mm3 with an average torque reduction rate of 82%.","PeriodicalId":405817,"journal":{"name":"2021 International Symposium on Medical Robotics (ISMR)","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117041127","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-11-17DOI: 10.1109/ismr48346.2021.9661504
Dayeon Kim, Su-Bin Joo, Joonho Seo, P. Kazanzides
Prosthetic hands have the potential to restore functionality to humans who have lost their hands, but it remains challenging to design a prosthetic hand that mimics a biological hand and then to effectively integrate that mechanical hand with the human brain for both sensing and control. We focus on the human/prosthesis integration and propose an Augmented Reality Manipulation Interface (ARMI) to facilitate that bi-directional integration. ARMI enables the user to specify intent (control) to the prosthesis while providing guidance (feedback) to the user through perception of the environment via artificial intelligence. Specifically, ARMI identifies objects in the environment and automatically determines the grasping configuration and timing; once the user selects an object, ARMI provides guidance for the user to correctly position the prosthesis with respect to the object and then initiate an autonomous grasp. We perform preliminary experiments with the Microsoft HoloLens head-mounted display (HMD) and a robotic hand to demonstrate the concept. Results suggest that ARMI would currently provide the greatest benefit for novice users who have not yet mastered the prosthetic hand, whereas further system improvements are necessary to provide a benefit for more experienced users.
{"title":"A Bi-directional User Interface for a Prosthetic Hand Using a Head-Mounted Display","authors":"Dayeon Kim, Su-Bin Joo, Joonho Seo, P. Kazanzides","doi":"10.1109/ismr48346.2021.9661504","DOIUrl":"https://doi.org/10.1109/ismr48346.2021.9661504","url":null,"abstract":"Prosthetic hands have the potential to restore functionality to humans who have lost their hands, but it remains challenging to design a prosthetic hand that mimics a biological hand and then to effectively integrate that mechanical hand with the human brain for both sensing and control. We focus on the human/prosthesis integration and propose an Augmented Reality Manipulation Interface (ARMI) to facilitate that bi-directional integration. ARMI enables the user to specify intent (control) to the prosthesis while providing guidance (feedback) to the user through perception of the environment via artificial intelligence. Specifically, ARMI identifies objects in the environment and automatically determines the grasping configuration and timing; once the user selects an object, ARMI provides guidance for the user to correctly position the prosthesis with respect to the object and then initiate an autonomous grasp. We perform preliminary experiments with the Microsoft HoloLens head-mounted display (HMD) and a robotic hand to demonstrate the concept. Results suggest that ARMI would currently provide the greatest benefit for novice users who have not yet mastered the prosthetic hand, whereas further system improvements are necessary to provide a benefit for more experienced users.","PeriodicalId":405817,"journal":{"name":"2021 International Symposium on Medical Robotics (ISMR)","volume":"197 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133152292","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-11-17DOI: 10.1109/ismr48346.2021.9661527
A. Soleymani, A. A. S. Asl, Mojtaba Yeganejou, Scott Dick, M. Tavakoli, Xingyu Li
Quality and safety are critical elements in the performance of surgeries. Therefore, surgical trainees need to obtain the required degrees of expertise before operating on patients. Conventionally, a trainee’s performance is evaluated by qualitative methods that are time-consuming and prone to bias. Using autonomous and quantitative surgical skill assessment improves the consistency, repeatability, and reliability of the evaluation. To this end, this paper proposes a video-based deep learning framework for surgical skill assessment. By incorporating prior knowledge on surgeon’s activity in the system design, we decompose the complex task of spatio-temporal representation learning from video recordings into two independent, relatively-simple learning processes, which greatly reduces the model size. We evaluate the proposed framework using the publicly available JIGSAWS robotic surgery dataset and demonstrate its capability to learn the underlying features of surgical maneuvers and the dynamic interplay between sequences of actions effectively. The skill level classification accuracy of 97.27% on the public dataset demonstrates the superiority of the proposed model over prior video-based skill assessment methods. The code of this paper will be available on Github at link: ${color{blue}{text{sourceCode}}}$.
{"title":"Surgical Skill Evaluation From Robot-Assisted Surgery Recordings","authors":"A. Soleymani, A. A. S. Asl, Mojtaba Yeganejou, Scott Dick, M. Tavakoli, Xingyu Li","doi":"10.1109/ismr48346.2021.9661527","DOIUrl":"https://doi.org/10.1109/ismr48346.2021.9661527","url":null,"abstract":"Quality and safety are critical elements in the performance of surgeries. Therefore, surgical trainees need to obtain the required degrees of expertise before operating on patients. Conventionally, a trainee’s performance is evaluated by qualitative methods that are time-consuming and prone to bias. Using autonomous and quantitative surgical skill assessment improves the consistency, repeatability, and reliability of the evaluation. To this end, this paper proposes a video-based deep learning framework for surgical skill assessment. By incorporating prior knowledge on surgeon’s activity in the system design, we decompose the complex task of spatio-temporal representation learning from video recordings into two independent, relatively-simple learning processes, which greatly reduces the model size. We evaluate the proposed framework using the publicly available JIGSAWS robotic surgery dataset and demonstrate its capability to learn the underlying features of surgical maneuvers and the dynamic interplay between sequences of actions effectively. The skill level classification accuracy of 97.27% on the public dataset demonstrates the superiority of the proposed model over prior video-based skill assessment methods. The code of this paper will be available on Github at link: ${color{blue}{text{sourceCode}}}$.","PeriodicalId":405817,"journal":{"name":"2021 International Symposium on Medical Robotics (ISMR)","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-11-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114846855","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：