Luohong Wu | Medical Image Analysis

About Me

I am a PhD candidate in the Research in Orthopedic Computer Science (ROCS) group at the University of Zurich and Balgrist University Hospital, supervised by Prof. Philipp Fürnstahl. I started my PhD in October 2022 and expect to graduate in December 2026.

My research focuses on medical ultrasound, 3D anatomical reconstruction, and multimodal registration. I develop methods to extract and reconstruct bone surfaces from ultrasound and align them with preoperative imaging, with applications in computer-assisted orthopedic surgery. I also work on augmented reality for surgical training.

I received my MSc in Computational Science and Engineering from ETH Zurich (2019–2022) and my BSc in Software Engineering from Northwestern Polytechnical University (2015–2019).

Research Interests

Medical image analysis: ultrasound segmentation and anatomical understanding.
3D reconstruction and registration: neural implicit representations and CT–ultrasound alignment.
Computer-assisted intervention: image-guided surgery and mixed-reality training.

Research

Google Scholar ↗

ComReg pipeline: patient-specific training and CT–ultrasound bone surface registration — Pipeline overview · Source ↗

MICCAI2026

Shape-Aware Registration Without Pretraining: CT–Ultrasound Bone Surface Alignment via Instance-Level Shape Completion

Luohong Wu, Elise Cho, Yunke Ao, Lennard Marx, Nicola Cavalcanti, Yiru Yang, Matthias Seibold, Philipp Fürnstahl

Medical Image Computing and Computer Assisted Intervention, 2026 Accepted

Instance-level shape completion for aligning CT and ultrasound bone surfaces without pretraining.

Project Code

Registration demo · Source ↗

MedIA2026

NeuralBoneReg: An instance-specific label-free point cloud-based method for multi-modal bone surface registration

Luohong Wu, Matthias Seibold, Nicola A. Cavalcanti, Yunke Ao, Roman Flepp, Aidana Massalimova, Lilian Calvet, Philipp Fürnstahl

Medical Image Analysis, 2026

Instance-specific registration of bone surfaces across imaging modalities, without training labels.

Paper Code Project

Dataset highlights · Source ↗

arXiv2026

Open-H-Embodiment: A Large-Scale Dataset for Enabling Foundation Models in Medical Robotics

Open-H-Embodiment Consortium: Nigel Nelson, Juo-Tung Chen, Jesse Haworth, et al. (including Luohong Wu)

arXiv preprint, April 2026 · In revision at Science Robotics

arXiv

UltraBoneUDF neural unsigned distance function method overview — Method overview · Source ↗

CMIG2025 online

UltraBoneUDF: Self-supervised bone surface reconstruction from ultrasound based on neural unsigned distance functions

Luohong Wu, Matthias Seibold, Nicola A. Cavalcanti, Giuseppe Loggia, Lisa Reissner, Bastian Sigrist, Jonas Hein, Lilian Calvet, Arnd Viehöfer, Philipp Fürnstahl

Computerized Medical Imaging and Graphics · online 2025; vol. 127, 2026

Reconstructing continuous bone surfaces from ultrasound observations using neural unsigned distance functions.

Paper Code

Simulation demo · Source ↗

NeurIPS2025

SonoGym: High performance simulation for challenging surgical tasks with robotic ultrasound

Yunke Ao, Masoud Moghani, Mayank Mittal, Manish Prajapat, Luohong Wu, Frederic Giraud, Fabio Carrillo, Andreas Krause, Philipp Fürnstahl

Advances in Neural Information Processing Systems (NeurIPS), 2025

Paper

UltraBones100k automated bone labeling workflow — Automatic bone labeling · Source ↗

CBM2025

UltraBones100k: A reliable automated labeling method and large-scale dataset for ultrasound-based bone surface extraction

Luohong Wu, Nicola A. Cavalcanti, Matthias Seibold, Giuseppe Loggia, Lisa Reissner, Jonas Hein, Silvan Beeler, Arnd Viehöfer, Stephan Wirth, Lilian Calvet, Philipp Fürnstahl

Computers in Biology and Medicine, 2025

An open dataset and automated labeling workflow for bone surface extraction from ultrasound.

Paper Code Dataset

Augmented reality orthopedic simulator with instrument guidance — AR surgical training · Source ↗

CBM2025

A novel augmented reality-based simulator for enhancing orthopedic surgical training

Luohong Wu, Matthias Seibold, Nicola A. Cavalcanti, Jonas Hein, Tatiana Gerth, Roni Lekar, Armando Hoch, Lazaros Vlachopoulos, Helmut Grabner, Patrick Zingg, Mazda Farshad, Philipp Fürnstahl

Computers in Biology and Medicine, vol. 185, 109536, 2025

Augmented reality simulation for orthopedic surgical training.

Paper Project

Prototype demo · Source ↗

UCAmI2019

The PRISMA: A Visual Feedback Display for Learning Scenarios

Bernardo Tabuenca, Luohong Wu, Edmundo Tovar

13th International Conference on Ubiquitous Computing and Ambient Intelligence (UCAmI), 2019

Paper

Datasets

Hugging Face ↗

UltraBones100k

Over 100,000 annotated ultrasound images of human lower limbs, with CT-derived bone models and tracking data for bone segmentation, surface reconstruction, and registration.

Dataset Code

UltraBonesHip

Paired CT and tracked ultrasound of the femur and pelvis from five cadaveric specimens, with code for reconstructing 3D ultrasound point clouds.

Dataset

SpineDepthSegmented

A preprocessed version of SpineDepth (Liebmann et al., 2021), with vertebra-level CT meshes and intraoperative RGB-D point clouds for lumbar spine registration.

Dataset

Open-H-Embodiment

A community-driven, multi-institution dataset of paired video and robot kinematics for medical robotics, spanning surgical manipulation, robotic ultrasound, and simulation. Contributed as part of the Open-H-Embodiment Consortium.

Dataset Project

Academic Service

Conference reviewer: NeurIPS, MICCAI.
Journal reviewer: Computer Methods in Biomechanics and Biomedical Engineering, Scientific Data, Scientific Reports.
Community: MONAI Ultrasound Working Group.

Teaching

Teaching assistant at the University of Zurich: Programming (2025), Introduction to Digital Health (2024), and Introduction to Data Science (2024).