VCMI is organizing 2 summer schools:


MAP Breast


Site Stats
Visits Total561
Visits Today0
Counted days2825



PICTURE (Patient Information Combined for the Assessment of Specific Surgical Outcomes in Breast Cancer) is a FP7 funded by the European Commission with a budget of 2.2M EUR for 2013/15.


Breast cancer is an increasingly treatable disease, and 10-year survival now exceeds 80%. When a woman faces a breast cancer diagnosis, and surgery is proposed, two options are available: breast-conserving surgery or mastectomy. Given the high breast cancer survival rate, many women will live for many years with the potentially disfiguring aesthetic consequences of their surgical and therapeutic treatment. The cosmetic outcome of surgery is a function of many factors including tumour size and location, the volume of the breast, its density, and the dose and distribution of radiotherapy. A good aesthetic outcome is an important endpoint for breast cancer treatment and is closely related to psychosocial recovery and quality of life.

The PICTURE project aims to address these issues by providing objective tools, tailored to the individual patient, to predict the aesthetic outcome of breast conserving surgery. Using a combination of 3D photography and routinely acquired radiological images (i.e. mammography, ultrasound and MRI, when available), together with information about the tumour (size, location, shape etc.) we will develop techniques to biomechanically model the anatomy of the breast and the effect of surgical removal of cancerous tissue. This digital patient representation and associated predictive tools will enable alternative surgical strategies to be explored and the consequences of the available options, with respect to the appearance of the breast, to be visualised. This will aid communication with the patient of the type of breast surgery recommended by the surgeon, and will empower patients to take an active role in a shared decision making process.

We will also develop tools to enable the patient's aesthetic appearance after treatment to be objectively evaluated. Current techniques use subjective methods, such as assessment by an expert panel, or computer analysis of 2-dimensional photography to estimate, for instance, breast asymmetry. By adopting recent developments in low cost 3D photography and depth sensing technology, we will develop a standardised, reproducible analysis tool which will base the aesthetic outcome evaluation on both the 3-dimensional shape of the reconstructed breast and its volume. This will establish standardised quality assurance and evaluation procedures, enabling institutions across Europe to be compared and factors that have a positive or negative impact on surgical outcome identified.


In PICTURE, VCMI is leading the Workpackages of 'Aesthetic Quantification', 'Image Processing and Image Analysis', and 'Knowledge Management'.


SARA (Asset Management System for Road Networks), funded by the Portuguese Agency for Innovation (ADI), with a budget of 1.4M EUR for 2012/14.

The project SARA – System to Manage Assets in Road Networks is a joint research initiative that benefits from the expertise of two INESC TEC Units – the Information and Computer Graphics Systems Unit (USIG) and the Telecommunications and Multimedia Unit (UTM). The aim is to create an innovative solution to efficiently manage road network assets.

The project SARA will provide a system to represent, register and update assets in a spatio-temporal context. With this system, it will be possible to integrate and develop decision-support tools and methodologies, thus contributing to a significant evolution of current activities in this area.

The main objective is to provide the necessary instruments to road network managers that will help them implement the most suitable conservation strategies according to the available economic resources and predefined quality standards. All this is based on the knowledge of the evolution of the network’s elements throughout time.

As part of the project, UTM will be creating computational vision algorithms to detect and classify road networks and pavement pathologies based on the semi-automatic recognition of images. Additionally, the Unit will develop graphic tools for result visualisation and to correct these elements and pathologies.

USIG, on the other hand, will be responsible for spatio-temporal modelling and for developing tools to edit instances in geographical entities (assets) and the occurrence of pathologies. The Unit will also implement decision-support modules to help create intervention scenarios and to help define conservation strategies and policies for the road networks.

With a duration of two years, the SARA has recently applied to NSRF funding through company MonteAdriano – Engenharia e Construção, S.A, a group in the construction sector which participates in businesses related to road concession and parking lots.

3d BCT

3d BCT (3D Models for Aesthetic Evaluation and Prediction of Breast Cancer Interventions) project (PTDC/SAU-ENB/114951/2009), funded by the Portuguese Foundation for Science and Technology (FCT), with a budget of 69K EUR for 2011/14.
In 3d BCT we investigate new methods to reconstruct 3D data from one or more uncalibrated views of the breast of the patient. We research a model fitting method, allowing the system to automatically fit a generic deformable model to patient specific three-dimensional (3D) breast surface measurements using a physically-based framework. This can be used to quantitatively and reliably assess the aesthetic outcome of breast reconstructive surgery. In addition this will also allow the surgeon to quantitatively analyze the degrees of various deformities and asymmetries in the shape of the breast. Finally, a model creation mode will allow a surgeon to interactively adjust the shape of the breast by varying key shape variables, analogous to the aesthetic and structural elements surgeons inherently vary manually during breast reconstruction. Our contribution will be a set of global deformations with very intuitive parameters that a physician can apply to a generic geometric primitive in order to model the breast of a patient for pre-operative planning purposes and for communicating and demonstrating this plan to the patient.


SINPATCO (Sistema Inteligente para Diagnóstico de Patologias da Coluna Vertebral) project, for 2010/11, under the Programme MOBILE CNPq - FEUP.
Project of international scientific collaboration with the Universidade Federal do Ceará. The focus of the project is in the application of machine learning techniques, specifically neural networks and kernel methods, in Traumato-Orthopedics clinic.


INCT MACC (Instituto Nacional de Ciência e Tecnologia Medicina Assistida por Computação Científica), for 2009/12.
Project led by Laboratório Nacional de Computação Científica - LNCC, Brazil and involving 33 institutions, designed to strengthen scientific and technological excellence on scientific computation for medicine (http://macc.lncc.br/contexto.php). 

Semantic PACS 

Semantic PACS (Picture Archiving and Communication System with Semantic Search Engine) project (project nº 003472), funded by the Portuguese Agency for Innovation (ADI), with a budget of 320K EUR for 2009/11.
Semantic PACS aims to develop a software module to integrate with PACS that supports automatic, semantic based, description and search methods directly over medical images. In opposition with existent systems, this solution will make possible to generate on-the-fly diagnosis reports based on the similarity of medical images archived in the system.


NeTS (Next Generation Network Operations and Management) project (CMU-PT/RNQ/0029/2009), funded by the Portuguese Foundation for Science and Technology (FCT), with a budget of 390K EUR for 2011/13 under the Cooperation Agreement between Portugal and Carnegie Mellon University.
The goal of NeTS is to develop a novel network operation and management framework that departs from conventional approaches through a cross-disciplinary research collaboration based on hierarchical network abstraction modeling, structure learning of probabilistic graphical models for machine learning, and wavelet and kernel-based signal processing technologies.


ASSIST is a contracted project that started in the end of 2012. It aims to study and develop algorithms and technologies to use video for the analysis of patterns in the movement of groups of people in limited spaces. The project aims to develop an accessible and flexible system for the national social economical context, but with high export potential.