Full Professor
University of Porto, Faculty of Engineering
My name is Jaime dos Santos Cardoso and I am a Full Professor at DEEC, in the Faculdade de Engenharia da Universidade do Porto (FEUP), Portugal.
Simultaneously, I'm also involved in research and development activities at INESC TEC in the Centre for Telecommunications and Multimedia. At INESC TEC, I am the co-founder of the Breast Research Group and of the Visual Computing and Machine Intelligence Group. At VCMI we focus our research on computer vision and pattern recognition.
If you are interested in pursuing a PhD in Computer Vision, Machine Learning, Medical Decision Support Systems, you are invited to email me. You may also want to enroll in the Doctoral Program in Electrical and Computer Engineering or in one of the Dual-degree Ph.D. Programs CMU | Portugal.
Learn here why you should learn and do research in Portugal!
University of Porto, Faculty of Engineering
University of Porto, Faculty of Engineering
University of Porto, Faculty of Engineering
University of Porto, Faculty of Engineering
University of Porto, Faculty of Engineering
Ph.D. in ECE (Computer Vision)
University of Porto, Faculty of Engineering
Title of the thesis: "Metadata Assisted Image Segmentation"
Master in Mathematical Engineering
University of Porto, Faculty of Sciences
Licentiate (5-year Licenciatura) in ECE
University of Porto, Faculty of Engineering
We pursue a never ending visual information learning system, to empower the next generation of intelligent systems with the capability of reasoning from visual data.
We perform research in both fundamental and applied problems in computer vision, image processing, machine learning, and decision support systems anchored in the automatic analysis of visual data.
Under these topics we favour more specific domains. Image and video processing focuses on medical images, biometrics and video object tracking for applications such as surveillance and sports. Our work on machine learning cares mostly with the adaptation of learning to the challenging conditions presented by visual data. The work on the development of intelligent decision support systems combines visual data understanding with any available additional information to enhance the analysis and the decision process.
Screening mammography is performed in the asymptomatic population to detect early signs of breast cancer such as masses, microcalcifications (MCs), bilateral asymmetry and architectural distortions (AD). Diagnostic mammography is performed on patients who have already demonstrated abnormal clinical findings. Both screening and diagnostic mammography are performed by radiologists who visually inspect mammograms. This is not an easy task: mammograms generally have low contrast. Mammograms show normal structures such as fat, fibroglandular tissue, breast ducts and nipples, as well as possible abnormalities. Although fat appears as black regions on mammograms, everything else appear as levels of white, making it hard to distinguish between normal and abnormal tissue.
The assumption made in this work is that, in screening, a substantial proportion of normal cases can be automatically detected, alleviating the human effort and giving the specialist more time to carefully evaluate more ambiguous cases.
We are interested in learning interpretable models for pre-CAD from weakly annotated data, both from mammograms and using the more recent tomosynthesis exam. For diagnosis, we aim at support the decision process by develping tools to identify clinically relevant findings.
When a woman faces a breast cancer diagnosis, and surgery is proposed, several options are available. The decision as to which type of surgery to offer patients is largely subjective and based almost exclusively on the judgment and experience of the clinician. 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. In breast-conserving surgery, there is evidence that approximately 30% of women receive a suboptimal or poor aesthetic outcome, however there is currently no standardised method of identifying these women.
In surgery planning, we aim to provide objective tools, tailored to the individual patient, to predict the aesthetic outcome of breast conserving surgery.
After the surgery, we aim to objectively evaluate the aesthetic outcome of the procedure.
The capability of (automatically) evaluating the aesthetic result depends strongly on our understanding of the main factors contributing to that outcome. Therefore, we study and identify the factors relevant to the aesthetic evaluation and perception of the result of the surgical and radiation therapy procedure. These will be the factors that can be derived from patients, tumour and treatment data and also from objective features extracted from digital photographs and 3D meshes of the breast surface. In spite of the differences between the available surgical and radiation therapy procedures, they share similarities that can be exploited when developing methods specific to each treatment type. The individual factors can be combined to provide an overall assessment of the aesthetic result. We adopt transfer and multitask learning techniques from the machine learning community to leverage the learning of models specific to each treatment type.
Cervical cancer is one of the leading causes of cancer death in women, with most cases occurring in low to middle income countries. Screening tests such as cytology or colposcopy have been responsible for a strong decrease in cervical cancer deaths. In this research line we aim to achieve results that exceed the current state-of-the-art in cervical cancer screening and diagnosis, in colposcopy procedures, by creating a Computer Aided-Diagnosis (CADx) system that can be easily integrated in the conventional clinical workflow.
When a new patient arrives at the clinical consultation, a cytology, either conventional or Liquid-based Cytology (LBC), depending on the resource availability, is performed. Cytological testing involves collecting exfoliated cells from the cervix, which are stained, fixated, and then visually examined under a microscope by a cytotechnologist. Despite being recommended by the World Health Organization (WHO), this procedure has the disadvantage of requiring great resources in terms of quality control, training and time consumption. If the cytology reveals any suspicious findings, a colposcopy is done, which consists in a common and low cost diagnosis method for the visualization of the affected area using a colposcope. The WHO recommends a protocol to perform the diagnosis which includes the examination using different lens filters and applying several solutions sequentially, which gives different sources of data. Since the manual evaluation is highly error prone, some projects have attempted the automatic detection but with limited performance and including only data from part of the protocol.
The purpose of this research line is to achieve results that exceed the current state-of-the-art in cervical cancer screening and diagnosis, in colposcopy procedures, by creating a Computer Aided-Diagnosis (CADx) system that can be easily integrated in the conventional clinical workflow. The primary goals of the project include the exploration of low-cost image acquisition approaches for cervical data; quality assessment of cervical imaging; image processing and analysis of cytological and colposcopy data; and fundamental machine learning and computer vision strategies to take advantage of multimodal settings with interpretability requirements.
Lorem ipsum dolor sit amet, consectetur adipisicing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat. Duis aute irure dolor in reprehenderit in voluptate velit esse cillum dolore eu fugiat nulla pariatur. Excepteur sint occaecat cupidatat non proident, sunt in culpa qui officia deserunt mollit anim id est laborum.
Lorem ipsum dolor sit amet, consectetur adipisicing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat. Duis aute irure dolor in reprehenderit in voluptate velit esse cillum dolore eu fugiat nulla pariatur. Excepteur sint occaecat cupidatat non proident, sunt in culpa qui officia deserunt mollit anim id est laborum.
Lorem ipsum dolor sit amet, consectetur adipisicing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat. Duis aute irure dolor in reprehenderit in voluptate velit esse cillum dolore eu fugiat nulla pariatur. Excepteur sint occaecat cupidatat non proident, sunt in culpa qui officia deserunt mollit anim id est laborum.
Lorem ipsum dolor sit amet, consectetur adipisicing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat. Duis aute irure dolor in reprehenderit in voluptate velit esse cillum dolore eu fugiat nulla pariatur. Excepteur sint occaecat cupidatat non proident, sunt in culpa qui officia deserunt mollit anim id est laborum.
Lorem ipsum dolor sit amet, consectetur adipisicing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat. Duis aute irure dolor in reprehenderit in voluptate velit esse cillum dolore eu fugiat nulla pariatur. Excepteur sint occaecat cupidatat non proident, sunt in culpa qui officia deserunt mollit anim id est laborum.
Lorem ipsum dolor sit amet, consectetur adipisicing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat. Duis aute irure dolor in reprehenderit in voluptate velit esse cillum dolore eu fugiat nulla pariatur. Excepteur sint occaecat cupidatat non proident, sunt in culpa qui officia deserunt mollit anim id est laborum.
This is not according to the standards of Reproducible Research but... it is a step forward on that direction :-)
Some code related with published papers:
@inproceedings{JaimeICIP2017, author = "Jaime S. Cardoso and Nuno Marques and Neeraj Dhungel and Gustavo Carneiro and Andrew Bradley", title = "Mass Segmentation in Mammograms: a Cross-Sensor comparison of deep and tailored features", booktitle = "Proceedings of the IEEE International Conference on Image Processing (ICIP)", url = "http://www.inescporto.pt/~jsc/publications/conferences/2017JaimeICIP.pdf", year = "2017", }.
@book{CardosoMICCAIDLMIA2016, year="2016", title="Deep Learning and Data Labeling for Medical Applications", subtitle = "First International Workshop, LABELS 2016, and Second International Workshop, DLMIA 2016, Held in Conjunction with MICCAI 2016, Athens, Greece, October 21, 2016, Proceedings", volume="10008", editor = "Gustavo Carneiro and Diana Mateus and Loïc Peter and Andrew Bradley and João Manuel R. S. Tavares and Vasileios Belagiannis and João Paulo Papa and Jacinto C. Nascimento and Marco Loog and Zhi Lu and Jaime S. Cardoso and Julien Cornebise" }
@article{KelwinNC2016, author = "Kelwin Fernandes and Jaime S. Cardoso", title = "Discriminative Directional Classifiers", journal = "Neurocomputing", year = "2016", type = "article", pages = "141--149", url = "http://www.inescporto.pt/~jsc/publications/journals/2016KelwinNeuroComputing.pdf", }
@inbook{JoaoCMonteiro2014, author="Joao C. Monteiro and Ana F. Sequeira and Helder P. Oliveira and Jaime S. Cardoso", editor= "Sebastiano Battiano and Sabine Coquillart and Robert Laramee and Andreas Kerren and Jose Braz", title = "Computer Vision, Imaging and Computer Graphics: Theory and Applications", chapter= "Robust Iris Localisation in Challenging Scenarios", pages = "146--162", publisher = "Springer", year= "2014", url = "http://www.inescporto.pt/~jsc/publications/conferences/2014JMonteiroCVICG.pdf", }
FEUP
Office: I335
Phone: 220413399
INESC TEC
Office: 3.14
Phone extension: 4228
email: jaime.cardoso at inesctec.pt
1st semester, FEUP/DEEC, MIEEC, 2 year, T and P
1st semester, FEUP/DEEC, PDEEC
1st semester, FEUP/DEEC, MIEEC, 2 year, T and P
1st semester, FEUP/DEEC, PDEEC
2nd semester, FEUP/DEEC, MIEEC, 1 year, L
2nd semester, FEUP/DEEC, PDEEC
1st semester, FEUP/DEEC, MIEEC, 2 year, T and P
1st semester, FEUP/DEEC, PDEEC
2nd semester, FEUP/DEEC, PDEEC
2nd semester, FEUP/DEEC, MEINF
2nd semester, FEUP/DEEC, MEINF
1st semester, FEUP/DEEC, MIEEC, 2 year, T and P
1st semester, FEUP/DEEC, PDEEC
2nd semester, FEUP/DEEC, PDEEC
2nd semester, FEUP/DEEC, MEINF
2nd semester, FEUP/DEEC, MEINF
1st semester, FEUP/DEEC, MIEEC, 2 year, T and P
1st semester, FEUP/DEEC, PDEEC
Lorem ipsum dolor sit amet, consectetur adipisicing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat. Duis aute irure dolor in reprehenderit in voluptate velit esse cillum dolore eu fugiat nulla pariatur. Excepteur sint occaecat cupidatat non proident, sunt in culpa qui officia deserunt mollit anim id est laborum.
I would be happy to talk to you.
You can find me at my office located at University of Porto.
I am at my office every day, but you may consider a call to fix an appointment.
You can find me at my Work located at University of Porto.
I am at my office every day, but you may consider a call to fix an appointment.
You can find me at my office located at University of Porto.
I am at my office every day, but you may consider a call to fix an appointment.