Shanghai Jiaotong University

Institute of Medical Robotics

Center for Image-guided Therapy and Interventions



The Center for Image-Guided Therapy and Interventions is one of the ten centers of the Institute of Medical Robotics (IMR), Shanghai Jiao Tong University (SJTU), and was established on January 1, 2019. The center focuses on the development and transformation of minimally invasive precision medical technology for major diseases such as tumor, cardiovascular and stroke. There are three PIs in the center, including a "Cross-century Outstanding Talent of the Ministry of Education" and "Shanghai Shuguang Scholar". Its projects include the construction of a public platform for the medical imaging in SJTU and a public computing platform for deep learning in IMR. The center currently purchases equipment worth 25 million yuan, including a 1.5T magnetic resonance scanner, a digital blood vessel silhouette machine, a three-dimensional Doppler ultrasound scanner and a DGX-2 deep learning computing platform.

Research Goals: The center develops a multi-mode image navigation intervention platform to realize the segmentation, registration, visualization and fusion of high-dimensional multi-mode image data, as well as the system integration of intelligent navigation and surgical robots based on multi-source information. For minimally invasive precision medical treatment for major diseases such as tumors, cardiovascular disease, and stroke, it carries out image navigation and robotic surgery applications for orthopedic surgery, abdominal puncture, and cardiovascular interventional surgery.


Research News: Five Papers Accepted to MICCAI 2022
June 3, 2022

Title: Handling Imbalanced Data: Uncertainty-guided Virtual Adversarial Training with Batch Nuclear-norm Optimization for Semi-supervised Medical Image Classification

Title: BIDMIR: Bi-Directional Medical Image Registration with Symmetric Attention and Cyclic Consistency Regularization

Title: Entropy and distance maps-guided segmentation of articular cartilage: data from the Osteoarthritis Initiative

Title: CyCMIS: Cycle-consistent Cross-domain Medical Image Segmentation via diverse image augmentation

Title: MCG-Net: End-to-end Fine-grained Delineation and Diagnostic Classification of Cardiac Events from Magnetocardiographs

Title: Spine-transformers: Vertebra labeling and segmentation in arbitrary field-of-view spine CTs via 3D transformers

Research News: Two Papers Accepted to MICCAI 2021
September 27-October 1, 2021

Title: Disentangled Representation Learning For Deep MR To CT Synthesis Using Unpaired Data

Title: Nonlinear Regression via Deep Negative Correlation Learning

Title: Entropy Guided Unsupervised Domain Adaptation for Cross-Center Hip Cartilage Segmentation from MRI