An MRI’s superior soft tissue contrast and ability to perform parametric scanning make it a powerful tool for use during medical procedures; from surgery to rehabilitation. However, the MRI’s strong static magnetic field, fast switching gradients and constrained space make accomplishing procedures within it difficult. Recent advances in the field of robotics have enabled the creation of devices capable of assisting medical practitioners in this environment. In this work, technologies to enable the use and control of robotic assistive devices for MRI interventions are presented. This includes a modular controller that is designed, built and used to control two surgical systems with minimal effect on image quality. Progressive improvements to an MRI conditional actuator including the construction of a first of a kind plastic piezoelectric resonant motor stator that improves the motor’s compatibility with the MRI is presented. Finally, control algorithms are evaluated for real-time functional MRI based control of a rehabilitation robot which includes the use of a robot for controlling brain activity of a subject in an online experiment.

No description available.

The percutaneous biopsy is a critical intervention for diagnosis and staging in cancer therapy. Robotic systems can improve the efficiency and outcome of such procedures while alleviating stress for physicians and patients. However, the high complexity of operation and the limited possibilities for robotic integration in the operating room (OR) decrease user acceptance and the number of deployed robots. Collaborative systems and standardized device communication may provide approaches to overcome named problems. Derived from the IEEE 11073 SDC standard terminology of medical device systems, we designed and validated a medical robotic device system (MERODES) to access and control a collaborative setup of two KUKA robots for ultrasound-guided needle insertions. The system is based on a novel standard for service-oriented device connectivity and utilizes collaborative principles to enhance user experience. Implementing separated workflow applications allows for a flexible system setup and configuration. The system was validated in three separate test scenarios to measure accuracies for 1) co-registration, 2) needle target planning in a water bath and 3) in an abdominal phantom. The co-registration accuracy averaged 0.94 ± 0.42 mm. The positioning errors ranged from 0.86 ± 0.42 to 1.19 ± 0.70 mm in the water bath setup and from 1.69 ± 0.92 to 1.96 ± 0.86 mm in the phantom. The presented results serve as a proof-of-concept and add to the current state of the art to alleviate system deployment and fast configuration for percutaneous robotic interventions.

Percutaneous needle placement, a minimally invasive procedure performed dozens of millions in the U.S. each year, relies on dedicated skill and long-term training due to difficult control of needle trajectory inside tissue and mental registration of images to locations inside the patient. Inaccurate needle placement may miss cancer tumors during diagnosis or eradicate healthy tissue during therapy. MRI provides ideal procedure guidance with the merit of excellent soft tissue contrast and volumetric imaging for high spatial resolution visualization of targets and surgical tool. However, manual insertion in the bore of an MRI scanner has awkward ergonomics due to difficult access to the patient, making both training and intervention even harder. To overcome the challenges related to MRI electromagnetic compatibility and mechanical constraints of the confined close-bore, a modular networked robotic system utilizing piezoelectric actuation for fully actuated prostate biopsy and brachytherapy is developed and evaluated with accuracy study. To enhance manipulation dexterity, two kinds of steerable continuum needle robots are developed. The asymmetric tip needle robot performs needle rotation and translation control to minimize tissue deformation, and increase steering dexterity to compensate placement error under continuous MRI guidance. The MRI-guided concentric tube robot is deployed to access delicate surgical sites that are traditionally inaccessible by straight and rigid surgical tools without relying on tissue reaction force. The master-slave teleoperation system with hybrid actuation is the first of its kind for prostate intervention with force feedback. The teleoperation controller provides the feel and functionality of manual needle insertion. Fabry- Perot interferometer based fiber optic force sensor is developed for the slave manipulator to measure needle insertion force and render proprioception feedback during teleoperation.

La robotique chirurgicale vise à rendre les gestes du chirurgien plus précis et moins invasifs. La complexité d’une salle d’opération conduit à rechercher des dispositifs robotiques aussi compacts que possible et pouvant être facilement stérilisés. Une conception robotique basée sur l’emploi de mécanismes compliants à structures monolithiques et d’actionneurs piézoélectriques est particulièrement intéressante sur ce point. Des travaux précédents conduits au laboratoire ont permis de proposer un dispositif robotique pour le pontage coronarien qui facilite la réalisation des gestes minimalement invasifs sur cœur battant. Ce dispositif répond au besoin médical mais manque aujourd’hui de la compacité souhaitée pour une intégration optimale. À partir du cas d’application où nous cherchons à réduire la taille du dispositif de compensation, nous nous intéressons, dans cette thèse, aux problématiques de conception de mécanismes compliants à fortes contraintes d’intégration. Nous étudions d’abord la possibilité d’intégrer le dispositif de compensation directement dans la tige du stabilisateur cardiaque passif. Puis, nous étudions la possibilité de réduire la taille du dispositif de compensation en amont, en explorant les possibilités de réaliser des mécanismes dans un plan. Nous avons notamment proposé une méthode originale de conception de mécanismes compliants plans à partir de l‘analyse des singularités de mécanismes à architectures parallèles en configuration plane. Afin d’optimiser les différents mécanismes très contraints par les volumes imposés, une méthode originale d’optimisation à base d’un algorithme de colonie de fourmis est employée. / Surgical robotics helps to increase the surgeon’s accuracy and limits the invasiveness of the surgery. The complexity of an operation room implies to design surgical devices that are as compact as possible and that can be easily sterilized. One interesting design approach is to combine compliant mechanisms, which have a monolithic structure, and piezoelectric actuators. Based on this approach, a robotic device for minimally invasive coronary artery bypass grafting has been proposed previously in our laboratory. This device successfully helps to increase the stabilization of the heart surface during the surgery but its compactness needs to be increased for an optimal integration in the operation room. Based on the need to reduce the compensation mechanism of this device, the problem of the design of compliant mechanisms with strong integration constrains is studied in this PhD thesis. First, the possibility to integrate the compensation mechanism directly in the shaft is considered. Then, the possibility to reduce the compensation mechanism at the end of the shaft by considering an assembly of planar manufactured structures is considered. Among the contributions, we propose an original design method based on the analysis of singularities of parallel manipulators in planar configuration. We also propose an original optimization method based on ant colony optimization in order to optimize the compliant architectures proposed in this work.

The field of surgery is continually evolving as there is always room for improvement in the post-operative health of the patient as well as the comfort of the Operating Room (OR) team. While the success of surgery is contingent upon the skills of the surgeon and the OR team, the use of specialized robots has shown to improve surgery-related outcomes in some cases. These outcomes are currently measured using a wide variety of metrics that include patient pain and recovery, surgeon’s comfort, duration of the operation and the cost of the procedure. There is a need for additional research to better understand the optimal criteria for benchmarking surgical performance. Presently, surgeons are trained to perform robot-assisted surgeries using interactive simulators. However, in the absence of well-defined performance standards, these simulators focus primarily on the simulation of the operative scene and not the complexities associated with multiple inputs to a real-world surgical procedure. Because interactive simulators are typically designed for specific robots that perform a small number of tasks controlled by a single user, they are inflexible in terms of their portability to different robots and the inclusion of multiple operators (e.g., nurses, medical assistants). Additionally, while most simulators provide high-quality visuals, simplification techniques are often employed to avoid stability issues for physics computation, contact dynamics and multi-manual interaction. This study addresses the limitations of existing simulators by outlining various specifications required to develop techniques that mimic real-world interactions and collaboration. Moreover, this study focuses on the inclusion of distributed control, shared task allocation and assistive feedback -- through machine learning, secondary and tertiary operators -- alongside the primary human operator.

Surgeon expertise can be difficult to capture through direct robot programming. Deep imitation learning (DIL) is a popular method for teaching robots to autonomously execute tasks through learning from demonstrations. DIL approaches have been previously applied to surgical automation. However, previous approaches do not consider the full range of robot dexterous motion required in general surgical task, by leaving out tooltip rotation changes or modeling one robotic arm only. Hence, they are not directly applicable for tasks that require rotation and dual-arm collaboration such as debridement. We propose to address this limitation by formulating a DIL approach for the execution of dual-arm surgical tasks including changes in tooltip orientation, position and gripper actions.<br><br>In this thesis, a framework for multi-step surgical task automation is designed and implemented by leveraging deep imitation learning. The framework optimizes Recurrent Neural Networks (RNNs) for the execution of the whole surgical tasks while considering tooltip translations, rotations as well as gripper actions. The network architecture proposed implicitly optimizes for the interaction between two robotic arms as opposed to modeling each arm independently. The networks were trained directly from the human demonstrations and do not require to create task specific hand-crafted models or to manually segment the demonstrations.<br><br>The proposed framework was implemented and evaluated in simulation for two relevant surgical tasks, the peg transfer task and the surgical debridement. The tasks were tested under random initial conditions to challenge the robustness of the networks to generalize to variable settings. The performance of the framework was assessed using task and subtask success as well as a set of quantitative metrics. Experimental evaluation showed favorable results for automating surgical tasks under variable conditions for the surgical debridement, which obtained a task success rate comparable to the human task success. For the peg transfer task, the framework displayed moderate overall task success. Quantitative metrics indicate that the robot generated trajectories possess similar or better motion economy that the human demonstrations.

The field of surgery is continually evolving as there is always room for improvement in the post-operative health of the patient as well as the comfort of the Operating Room (OR) team. While the success of surgery is contingent upon the skills of the surgeon and the OR team, the use of specialized robots has shown to improve surgery-related outcomes in some cases. These outcomes are currently measured using a wide variety of metrics that include patient pain and recovery, surgeon’s comfort, duration of the operation and the cost of the procedure. There is a need for additional research to better understand the optimal criteria for benchmarking surgical performance. Presently, surgeons are trained to perform robot-assisted surgeries using interactive simulators. However, in the absence of well-defined performance standards, these simulators focus primarily on the simulation of the operative scene and not the complexities associated with multiple inputs to a real-world surgical procedure. Because interactive simulators are typically designed for specific robots that perform a small number of tasks controlled by a single user, they are inflexible in terms of their portability to different robots and the inclusion of multiple operators (e.g., nurses, medical assistants). Additionally, while most simulators provide high-quality visuals, simplification techniques are often employed to avoid stability issues for physics computation, contact dynamics and multi-manual interaction. This study addresses the limitations of existing simulators by outlining various specifications required to develop techniques that mimic real-world interactions and collaboration. Moreover, this study focuses on the inclusion of distributed control, shared task allocation and assistive feedback -- through machine learning, secondary and tertiary operators -- alongside the primary human operator.

Abstract Minimally invasive surgery (MIS) techniques have seen rapid growth as methods for improved operational procedures. The main technology of MIS is based on image guided surgery (IGS) devices, namely surgical navigators, surgical robotics and image scanners. With their widespread use in various fields of surgery, methods and tools that may be used routinely in the hospital setting for “real world” assessment of the accuracy of these devices are lacking. In this thesis the concept of accuracy testing was developed to meet the needs of quality assurance of navigators and robots in a hospital environment. Thus, accuracy was defined as the difference between actual and measured distances from an origin, also including determination of directional accuracy within a specific volume. Two precision engineered accuracy assessment phantoms with assessment protocols were developed as advanced materials and methods for the community. The phantoms were designed to include a common region of surgical interest (ROSI) that was determined to roughly mimic the size of the human head. These tools and methods were utilized in accuracy assessment of two commercial navigators, both enabling the two most widely used tracking modalities, namely the optical tracking system (OTS) and the electromagnetic tracking system (EMTS). Also a study of the accuracy and repeatability of a prototype surgical interactive robot (SIRO) was done. Finally, the phantoms were utilized in spatial accuracy assessment of a commercial surgical 3D CT scanner, the O-Arm. The experimental results indicate that the proposed definitions, tools and methods fulfill the requirements of quality assurance of IGS devices in the hospital setting. The OTS and EMTS tracking modalities were nearly identical in overall accuracy but had unique error trends. Also, the accuracy of the prototype robot SIRO was in the range recommended in the IGS community. Finally, the image quality of the O-Arm could be analyzed using the developed phantoms. Based on the accuracy assessment results, suggestions were made when setting up each IGS device for surgical procedures and for new applications in minimally invasive surgery. / Tiivistelmä Mini-invasiivisen eli täsmäkirurgian tekniikoita ja teknologioita on alettu hyödyntää viime aikoina yhä enemmän. Tavoitteena on ollut parantaa kirurgisten operaatioiden tarkkuutta ja turvallisuutta. Täsmäkirurgiassa käytetyt teknologiat pohjautuvat kuvaohjattuihin kirurgisiin paikannuslaitteisiin. Kuvaohjattuihin laitteisiin kuuluvat navigaattorit, kirurgiset robotit ja kuvantalaitteet. Näiden laitteistojen kehittyminen on mahdollistanut tekniikoiden hyödyntämisen monialaisessa kirurgiassa. Paikannuslaitteistojen ja robottien yleistyminen on kuitenkin nostanut sairaaloissa esiin yleisen ongelman paikannustarkkuuden määrittämisessä käytännön olosuhteissa. Tässä väitöskirjassa esitetään kirurgisten yksiköiden käyttöön menetelmä sekä kaksi uutta fantomia ja protokollaa käytössä olevien paikannuslaitteistojen tarkkuuden määrittämiseen. Fantomit suunniteltiin sisältämään ennalta määritetty kirurginen kohdealue, mikä rajattiin käsittämään ihmisen kallon tilavuus. Fantomeita ja protokollaa hyödynnettiin kahden kaupallisen paikannuslaitteen tarkkuuden määrityksessä. Navigaattorit käyttivät optiseen ja elektromagneettiseen paikannukseen perustuvaa tekniikkaa. Lisäksi työssä kehitetyillä menetelmillä tutkittiin prototyyppivaiheessa olevan kirurgisen robotin paikannus- ja toistotarkkuutta sekä tietokonetomografialaitteen O-kaaren kuvan tarkkuuden määritystä. Kokeellisten tulosten perusteella työssä kehitetyt fantomit ja protokollat ovat luotettavia ja tarkkoja menetelmiä kirurgisten paikannuslaitteistojen tarkkuuden määrittämiseen sairaalaoloissa. Kirurgisten navigaattoreiden tarkkuuden määritystulokset osoittivat optisen ja elektromagneettisen paikannustekniikan olevan lähes yhtä tarkkoja. Prototyyppirobotin tarkkuus oli tulosten perusteella kirjallisuudessa esitettyjen suosituksien mukainen. Lisäksi O-kaaren kuvanlaatua voitiin tutkia kehitetyillä fantomeilla. Tarkkuudenmääritystulosten perusteella työssä ehdotetaan menetelmiä laitteistojen optimaalisesta käytöstä leikkaussalissa sekä laajennetaan niiden käyttömahdollisuuksia. Tuloksia voidaan hyödyntää myös paikannuslaitteistojen kehittämistyössä.

No description available.