ALTAIR Robotics Lab

Call for Applications in Medical Robotics at the University of Verona, Italy

The ALTAIR Robotics Laboratory (http://metropolis.scienze.univr.it/altair), at the Computer Science
Department (http://www.di.univr.it/) of the University of Verona (http://www.univr.it) is looking for
candidates to fill the following research positions in Medical Robotics:

A. Positions within the project ARS (Autonomous Robotic Surgery http://www.ars-project.eu), funded by
    the European Research Council Advanced Research grant (ERC-ADG) that addresses the science and
    engineering of autonomous robots in surgery. The project aims at developing the various elements of a
    cognitive surgical robot that will demonstrate simple surgical interventions on mannequins using the da
    Vinci Research Kit (dVRK) available in the laboratory.
    1. Two full time PhD students (3-year fully-funded positions). The specific research topics are:
        a. Machine learning techniques to extract robotic surgery models from data sets.
            This research will investigate the structure and the actual evolution of robot-assisted surgery
            through the examination of data collected during different types of real and simulated
            robotic interventions. The research will focus on the development of data-driven models that
            use/embed prior knowledge in the data analysis, identify possible variants to prior
            knowledge and propose the reward functions optimized by the operating surgeons.
        b. Modeling, analysis and control of a cognitive robot as a hierarchy of hybrid systems.
            This research will analyze the interaction between the reasoning and the control parts of a
            cognitive robot, by modeling the complete system as a hierarchy of discrete and continuous
            controllers whose properties, including safety, should be theoretically analyzed and
            experimentally verified.

        Applicants should have a MSc/MEng (or equivalent) in Engineering, Computer Science, Mathematics
        or related disciplines. Applicants must have strong programing skills and background in Machine
        Learning or Control Engineering preferably with some experience in the integration of control and
        machine learning. The successful candidate to the first position is expected to have a strong
        background in machine learning and mathematical modeling. The successful candidate to the second
        position is expected to have a strong background in system theory, hybrid system control, and formal
        property analysis. Both candidates must be able to work in a heterogeneous multi-disciplinary team,
        and be able to test their results on the laboratory robots. The positions will start in October 2019.
   

    2. Two full-time Post-Doc positions (1-year renewable contract). The specific research topics are:
        a. Real time reasoning and situation awareness in robotic surgery.
            This research will investigate the structure and the actual implementation of a real-time
            reasoning system that analyzes the data collected during simulated and real robotic
            interventions. The system should be able to compute the conditions determining the
            evolution of the surgical intervention, as well as identifying possible risk situations related to
            the patient conditions.
        b. Modeling and real time simulation of deformable anatomical environments.
            This research will analyze the best bio-mechanical model to represent a deformable organ,
             identify the most appropriate software formulation to implement the model and integrate the
            model into real-time anatomical simulator under development within the ARS project.
        Applicants should submit the following documents:
                • A short CV. Only CVs with considerable scientific achievements will be considered.
                  Candidates should have an excellent track record of research; they should demonstrate
                  the ability to grow into scientific/technological leadership roles.
                • A statement on how the research topics will be addressed.
                • A letter of presentation.
                • Contacts of two supervisors/mentors who could act as references.
        The positions will start as soon as possible.

Interested candidates should contact Paolo Fiorini (paolo.fiorini@univr.it) as early as possible.
Preliminary, in person interviews will be possible during ISMR19.


    B. Two full time PhD students (3-year fully-funded positions) within the project ATLAS (AuTonomous
         intraLuminAl Surgery, https://atlas-itn.eu), funded by the Marie Skłodowska-Curie Actions Innovative
        Training Networks (MSCA-ITN) that will develop smart flexible robots that autonomously propel and
        navigate through complex deformable tubular structures such as fragile lumens or vessels. The specific
        research topics are:
        a. ESR9: Surgical episode segmentation from multi-modal data.
            An autonomous robot must reliably recognize the current surgical phase to decide what action to
            perform, especially in the case of shared autonomy, i.e. when the human surgeon carries out part
            of the procedure and the robot intelligently/semi-automatically assists the manual gestures. Deep
            learning methods (e.g. combined CNN for video and RNN for lower dimensionality data) could
            be used to extract the intervention phases and trajectories. This research aims at breaking new
            ground by using multi-modal data: it will build on phase detection in endoscopic video data, and
            it will include data from intra-operative sensors such as EM trackers, ultrasound images, preoperative
            data. Feature detectors and descriptors will be developed to perform optimal
            discrimination of different areas of interest and to reduce the data dimensionality for improved
            computational performance. Since methods developed in this research use multi-modal data,
            they may be also applied to intraluminal procedure based on video feedback (i.e. colonoscopy
            and ureteroscopy) and also to cardiovascular catheterization
            Main institution and supervisor: University of Verona, Diego Dall’Alba
            Secondary institution and supervisor: University of Strasbourg, Nicolas Padoy
        b. ESR15: Optimal learning method for autonomous control and navigation.
            Learning optimal control strategies for autonomous anatomical navigation and on-line decision
            making is a challenging problem. Currently 2 main strategies could be adopted: learning from
            data acquired during the execution of surgical procedure by expert surgeon or learning by
            experimentation. This research should compare motion control strategies for intraluminal
            navigation learned from intervention data with those learned in simulated environment. Then, it
            will be possible to identify the optimal control strategies for different clinical scenarios given
            specific robotic configurations. The trajectories identified in the research ESR9 will be used to
            define the initial trajectory of the autonomous endoscope. The performance of the different
            approaches will be evaluated in a realistic setting (physical phantoms) and in a simulated
            environment using a set of objective evaluation metrics. An integrated testing environment,
            including advanced visualization, will be developed to improve the evaluation of the different
            methods (extending/integrating the results of ESR10). The proposed navigation strategies will be
            tested in the colonoscopy and ureteroscopy clinical scenarios, but possible extensions to
            cardiovascular catheterization will be considered.
            Main institution and supervisor: University of Verona, Paolo Fiorini
            Secondary institution and supervisor: Universidad Politécnica de Catalunya, Alicia Casals

Interested candidates should contact Paolo Fiorini (paolo.fiorini@univr.it) as early as possible.
and apply to the positions using the web form at https://atlas-itn.eu/we-are-hiring/


    C. Two Expressions of Interest for the Marie Skłodowska-Curie Actions Individual Fellowships (MSCAIF).
        Researchers for these positions will be working, if funded, on the ARS project and on a smart
        exoskeleton for the upper body, to support people affected by muscular weakness.
        a. To establish the fundaments of intelligent surgical instruments
            These instruments will be endowed with sensing device and strategies that will be capable to
            autonomously perform the measurements required to achieve situation awareness and acquire
            unprecedented level of detail about its surroundings. Aim of this research is the investigation of
            the proper design of soft and hard instruments with embedded sensors and intelligence. New
            paradigms for sensor integration will be taken into consideration during the design for
            guaranteeing a proper fabrication compatible with the working conditions. The design should
            incorporate adequate intra-operative sensing technologies (FBG, OCT, EM, US) and proper
            actuation schemes. The optimal sensor/actuation combination will be implemented in dedicated
            surgical robotic systems by specific integration procedure with the final aim to enhance the selfawareness
            of the whole robotic system and optimizing application safety. This new generation of
            self-aware instruments will be tested during in ex-vivo and in-vitro experiments. Through their
            proprioceptive capabilities these instruments can exceedingly put anatomical information into
            context, improve the operator’s awareness of the surgical site, and increasing the intuitiveness
            and the efficiency of the intervention.
        b. Upper Limb Exoskeleton Control for ADL Assistance
            This research aims at developing advanced sEMG decoding and exoskeleton control algorithms
            considering one or more of the following specific topics:
                  i. Advanced sEMG decoding algorithms with a specific focus on robustness to sensor
                     misplacements and self-calibration.
                 ii. Advanced assistive algorithms working at the interface between sEMG decoding and
                     exoskeleton control.
                iii. Advanced force and impedance control of series elastic systems based on low cost
                    implementations and with specific focus on identification and estimation techniques to
                    compensate for system uncertainties.
                iv. Mechatronic design of mobility support systems (e.g. exoskeletons) to help patients with
                    reduced physical mobility exploiting low-cost materials (e.g. plastic materials) and
                    components (low-end sensors and actuators). See the concept of Series Elastic Link
                    which has been recently introduced by our group.
            The selected candidate will be invited to submit a MSCA IF proposal together with Prof. Fiorini,
            supervisor of the proposal. The involvement of the selected candidate in the proposal writing process
            will provide ample opportunity to tailor the proposal to his / her research interests. A successful
            application will result in a one-year, or a two-year appointment.
            Research Field: Robotics, Automation and Computer Science Engineering
            Career Stage: Experienced researcher or 4-10 yrs (Post-Doc)
            Research Profiles: Recognized Researcher (R2)

Interested candidates should contact Paolo Fiorini (paolo.fiorini@univr.it) as early as possible.
Preliminary, in person interviews will be possible during ISMR19.