Bio Link to heading

Hello! My name is Jiaming (Jeremy) Zhang. I am a first-year Ph.D. student in Computer Sciecne at The Johns Hopkins University. My reasearch focuses on the application of robotics in medical scenarios. Prior joining in Hopkins, I received my Bachelor degree in Mechanical Engineering from Huazhong University of Science and Technology, where I graduated with honors. Afterwards, I obtained my Master’s degree in Robotics at Hopkins. I was an intern engineer at Narwal Robotics and HUST-Wuxi Research Institute. During my leisure time, I enjoy playing tennis, cooking and reading.

Research Interest and Experience Link to heading

As a robotics enthusiast with a strong interest in visual perception, motion planning, and medical robots, I have conducted research in various areas such as hand-eye calibration, LiDAR-based SLAM, exoskeleton design and analysis, AI-based medical image analysis, and medical robots. Thess experiences equip me with solid foundation for Mathematics, ROS development, Robot Hardware Design and the state-of-the-art Computer Vision models.

Ongoing Project Link to heading

Currently, I’m working as a research assisant in BIOMECHANICAL AND IMAGE GUIDED SURGICAL SYSTEMS (BIGSS) laboratory under the supervision of Prof. Mehran Armand and Prof. Alejandro Martin-Gomez. Here are the projects that I’m carrying out.

• Deformable Object Simulation
• Robot TMS
• Position-based Visual Servoing
• Learning-based 2D/3D Fluoroscopic Image Registration
• 3D Slicer Integration with Segment Anything Model, aka SAMM (Segment Any Medical Model)

Deformable Object Simulation Link to heading

Realtime shape estimation of continuum objects and manipulators is essential for developing accurate planning and manipulation paradigms. The existing methods that create dense point clouds from camera images, and/or use distinguishable markers on a deformable body have limitations in realtime tracking of large continuum objects/manipulators and the physical occlusion of markers can often compromise accurate shape estimation. We propose a robust method to estimate the shape of linear deformable objects in realtime using scattered and unordered key points. By utilizing a robust probability-based labeling algorithm, our approach identifies the true order of the detected key points and then reconstructs the shape using piecewise spline interpolation. We demonstrate the robustness of the method when key points are partially occluded. We also show that the method can be used to estimate the shape of a continuum manipulator in realtime. The proposed method is implemented in C++ and integrated with ROS. The code is available here.

Robot TMS Link to heading

Position-based Visual Servoing Link to heading

Learning-based 2D/3D Fluoroscopic Image Registration Link to heading

We developed and maintained an open-source python package xregi that is used to compute the projection matrix between 2D x-ray images and 3D CT scans. This process is also called as 2D/3D fluoroscopic image registration problem. The package is now available on Pypi. Install the pacakge through pip install xregi to try out!

SAMM Link to heading

I’m excited to announce that our project SAMM has received many attentions from all over the world! Please reach out if you want to collaborate. Here is a demo video of SAMM.