Link for Live Seminar

Link for Recorded seminars – 2022/2023 school year

Benjamin D. Killeen “An Autonomous X-ray Image Acquisition and Interpretation System for Assisting Percutaneous Pelvic Fracture Fixation”

Abstract: Percutaneous fracture fixation involves multiple X-ray acquisitions to determine adequate tool trajectories in bony anatomy. In order to reduce time spent adjusting the X-ray imager’s gantry, avoid excess acquisitions, and anticipate inadequate trajectories before penetrating bone, we propose an autonomous system for intra-operative feedback that combines robotic X-ray imaging and machine learning for automated image acquisition and interpretation, respectively. Our approach reconstructs an appropriate trajectory in a two-image sequence, where the optimal second viewpoint is determined based on analysis of the first image. The reconstructed corridor and K-wire pose are compared to determine likelihood of cortical breach, and both are visualized for the clinician in a mixed reality environment that is spatially registered to the patient and delivered by an optical see-through head-mounted display. We assess the upper bounds on system performance through in silico evaluation across 11 CTs with fractures present, in which the corridor and K-wire are adequately reconstructed. In post-hoc analysis of radiographs across 3 cadaveric specimens, our system determines the appropriate trajectory to within 2.8 ± 1.3 mm and 2.7 ± 1.8°. An expert user study with an anthropomorphic phantom demonstrates how our autonomous, integrated system requires fewer images and lower movement to guide and confirm adequate placement compared to current clinical practice.

Bio: A 4th year Ph.D. candidate at Johns Hopkins University, Benjamin D. Killeen is interested in intelligent surgical systems that improve patient outcomes. His recent work involves realistic simulation of interventional X-ray imaging for the purpose of developing AI-integrated surgical systems. Benjamin is a member of the Advanced Robotics and Computationally Augmented Environments (ARCADE) research group, led by Mathias Unberath, as well as the President of the LCSR Graduate Student Association (GSA) and Sports Officer for the MICCAI Student Board. In 2019, he earned a B.A. in Computer Science with Honors from the University of Chicago, with a minor in Physics, and he has completed internships at IBM Research – Almaden, Epic Systems, and Intuitive Surgical. In his spare time, he enjoys bouldering and creative writing.

Divya Ramesh “Studying terrestrial fish locomotion on wet deformable substrates”

Abstract: Many amphibious fishes can make forays onto land. The water-land interface often has wet deformable substrates like mud and sand, whose strength changes as they get dryer or wetter, challenging locomotion. Most previous terrestrial locomotion studies of fishes focused on quantifying kinematics, muscle control, and functional morphology. Yet, without quantifying how the complex mechanics of wet deformable substrates affect ground reaction forces during locomotion, we cannot fully understand how these locomotor features interact with the environment to permit performance. Here, we used controlled mud as a model wet deformable substrate and developed methods to prepare mud into spatially uniform and temporally stable states and tools to characterize its strength. As a first step to understand how mud strength impact locomotion, we studied the Atlantic mudskipper (Periophthalmus barbarus) moving on a thicker and a thinner mud, which differs in strength by a factor of two. The animal performed similar “crutching” walks on mud of both strengths, with only a slight reduction in speed on the thinner mud (from 0.39 ± 0.12 to 0.32 ± 0.14 body length/s, P < 0.05, ANOVA). However, it jumped more frequently on the thinner mud (from 1.2 ± 0.7 to 3.2 ± 1.6 times per minute, P < 0.05, ANOVA), likely due to it sticking more to the belly and fins and hindering walking.

Bio: Divya Ramesh is a fourth year PhD student in Dr. Chen Li’s lab (Terradynamics lab). Her current work focuses in studying and understanding amphibious fish locomotion on wet deformable substrates. Her previous work focused in using contact sensing to study and understand limbless locomotion of snakes and snake-robot on 3-D terrains. She received a BTech in Electronics and Communication Engineering from VIT University (Vellore, India) and MSE in Electrical Engineering from University of Pennsylvania. She has published in IEEE RA-L (presented at ICRA 2020) and presented at ICRA 2022. This work was presented in SICB 2023 where she was a finalist for Best Student Presentation in the Division of Comparative Biomechanics.

Gargi Sadalgekar “Template-level robophysical models for studying sustained terrestrial locomotion of amphibious fish”

Abstract: Studying terrestrial locomotion of amphibious fishes informs how early tetrapods may have invaded land. The water-land interface often has wet, deformable substrates like mud and sand that challenge locomotion. Recent progress has been made on understanding limbed and limbless tetrapod locomotion by studying robots as active physical models of model organisms. Robophysical models complement animals with their high controllability and repeatability for systematic experiments. They also complement theoretical and computational models because they enact physical laws in the real world, which is especially useful for studying locomotion in complex terrain. Here, we created the first robophysical models for studying sustained terrestrial locomotion of amphibious fishes on controlled mud as a model web deformable substrate. Our three robots are on the template level (lowest degree-of-freedom to generate a target locomotor behavior) and represent mudskippers, ropefish, and bichirs that use appendicular, axial, and axial-appendicular strategies, respectively. The mudskipper robot rotated two fins in phase to raise the body and “crutch” forward on mud. The ropefish robot used body lateral undulation to “surface-swim” on mud. The bichir robot combined body undulation and out-of-phase fin rotations to “army-crawl” forward on mud. Each robot generated qualitatively similar locomotion on mud as its model organism. We are currently refining the robots and performing systematic experiments on mud of a wide range of strengths.

Bio: Gargi Sadalgekar is a 2^nd year master’s student in the Terradynamics Lab at Johns Hopkins University and is interested in developing bio-inspired robots to investigate locomotion in extreme environments. Her current work focuses on developing low-order robophysical models of amphibious fish to uncover general principles of locomotion over wet deformable substrates, and this work was presented in SICB 2023 where she was a finalist for Best Student Presentation in the Division of Comparative Biomechanics. Gargi received a BSE in Mechanical and Aerospace Engineering from Princeton University with a minor in Robotics and Information Systems.

Yaqing Wang “Force sensing can help robots reconstruct potential energy landscape and guide locomotor transitions to traverse large obstacles”

Abstract: Legged robots already excel at maintaining stability during upright walking and running to step over small obstacles. However, they must further traverse large obstacles comparable to body size to enable a broader range of applications like search and rescue in rubble and sample collection in rocky Martian hills. Our lab’s recent research demonstrated that legged robots can traverse large obstacles if they can be destabilized to transition across various locomotor modes. When viewed on a potential energy landscape of the system, which results from physical interaction with obstacles, these locomotor transitions are strenuous barrier-crossing transitions between landscape basins. Because potential energy landscape gradients are closely related to terrain reaction forces and torques, we hypothesize that sensing obstacle interaction forces allows landscape reconstruction, which can guide robots to cross barriers at the saddle to make transitions more easily (analogous to crossing a mountain ridge at its saddle). Here, we created a robophysical model with custom 3-axis force sensors and surface contact sensors to measure forces and contacts during interaction with large obstacles. We found that the measured forces indeed well captured potential energy landscape gradients and we could use the locally measured gradients to roughly reconstruct the potential energy landscape. Our future work should understand how to enable robots to make locomotor transitions at the landscape saddle based on local landscape reconstruction.

Bio: Yaqing Wang is a fourth-year PhD student in Dr. Chen Li’s lab (Terradynamics lab). His work focuses on understanding locomotor transitions in bio and bio-inspired terrestrial locomotion. He received a B.S. in Mechanical Engineering at Tsinghua University in China. He recently presented this work at the annual APS March meeting.