Computer Vision For Computer Aided Arthroscopy


Arthroscopy is a modality of orthopeadic surgery for treatment of damaged joints in which instruments and endoscopic camera (the arthroscope) are inserted into the articular cavity through small incisions (the surgical ports). Arthroscopy is highly beneficial for the patient because it reduces trauma, risk of infection and recovery time. However, it is much more difficult to execute because of indirect visualization and limited manoeuvrability inside the joint, with novices having to undergo a long training period and experts often making mistakes of clinical consequences.
The VisArthro project aims to overcome these drawbacks by conducting research in computer vision techniques to build a system for Computer-Aided Arthroscopy (CAA). Such system will combine real-time video processing with augmented reality to guide the surgeon throughout the procedure, being the first of the kind not requiring additional intra-operative sensing modalities such as opto/magnetic tracking. Our research will be motivated by two specific procedures: the reconstruction of the Anterior Cruciate Ligament (ACL) in the knee and the removal of Femur Acetabular Impingements (FAI) in the hip.


Invited lecture in surgical navigation at the 31st edition of PADIS (Programa de Alta Direcção de Instituições de Saúde) organised by AESE Business School
VisArthro clinical team organizes a full day session of the Master Programme of EFORT Fellowship with an invited lecture by Joao Barreto on Surgical Navigation Technologies
Article reporting the first results in navigated arthroscopy of the knee is presented at MICCAI conference in Granada, Spain (ar: 25%)
Second cadaver experiment in knee arthroscopy successfully carried in INMLCF facility in Coimbra, Portugal.
Article in 3D registration of curves and surfaces is presented at CVPR conference in Salt Lake City, USA (ar: 25%)
Invited lecture in surgical navigation at the 30th edition of PADIS (Programa de Alta Direcção de Instituições de Saúde) organised by AESE Business School
P3D exhibits at the American Academy of Orthopeadic Surgeons (AAOS) in New Orleans, USA
First cadaver experiment in knee arthroscopy successfully carried in INMLCF facility in Coimbra, Portugal.
Article on 3D registration using normals presented at ICRA conference in Singapore
P3D exhibits a prototype of video-based navigation in arthroscopy at the American Academy of Orthopeadic Surgeons (AAOS) in San Diego, USA
Article presented at ECCV conference in Amesterdam, Netherlands (ar: 25%)
Article on using affine correspondences for SfM presented as "spotlight paper" at CVPR conference in Las Vegas, USA (ar: 10%)
Official kickoff of project VisArthro


First Cadaver Experiment in Navigated Arthroscopy

Video of the first cadaver experiment in knee arthroscopy performed at the "Instituto Nacional de Medicina Legal e Ciência Forense" (INMLCF) that kindly joined the consortium to make these experiments possible. The video shows the view acquired by the arthroscopic camera. The surgeon starts by attaching the base marker to anatomy using a specially designed tool and then digitalises the intercondylar surface with the help of an instrumented touch probe. The 3D reconstruction results are used as input by a registration algorithm that overlays a pre-operative CT of the femur over the patient anatomy. The CT includes a plan for femural tunnel which enables to guide the surgeon using augmented reality techniques.

Functionalities of video-based Computer Aided Arthroscopy (CAA)

The video shows the key functionalities of the video-based Computer-Aided Arthroscopy (CAA) developed in project VisArthro. The technology, dubbed in.nav by P3D, is demonstrated in the context of ACL reconstruction in the knee. It enables metric measurements, 3D reconstruction of points and contours, overlay of pre-operative 3D models and guidance either using augmented or virtual reality


International Conferences

Submitted / Under review





The VisArthro consortium comprises three institutions: the Institute for Systems and Robotics (ISR-UC) that is affiliated with the University of Coimbra, the Perceive3D (P3D) that is a startup company operating in CAS, and the "Centro Hospitalar e Universitário de Coimbra" (CHUC) that is the clinical partner. The project also counts with the participation of the University College London (UCL) that advises and collaborates with the consortium.

Team at ISR-UC

Pedro Rodrigues

Diogo Vaz

Former Members

Bruna Nogueira

Ricardo Simões

Team at P3D

Rui Melo

Cristóvão Sousa

Michel Antunes

Team at CHUC

Joao Oliveira

Team at UCL