Cookies Policy
The website need some cookies and similar means to function. If you permit us, we will use those means to collect data on your visits for aggregated statistics to improve our service. Find out More
Accept Reject
  • Menu
    About

    About

    Hélder P. Oliveira Hélder P. Oliveira was born in Porto, Portugal, in 1980. He graduated in Electrical and Computer Engineering in 2004, received the M.Sc. degree in Automation, Instrumentation and Control in 2008 and the Ph.D. degree in Electrical and Computer Engineering in 2013 at the Faculty of Engineering of the University of Porto (FEUP), Portugal. He is currently working as Senior Researcher at INESC TEC - Institute for Systems and Computer Engineering, Technology and Science, a R&D institute affiliated to the University of Porto. He is the Leader of the Visual Computing and Machine Intelligence Area, member of the coordination council of the Centre for Telecommunications and Multimedia, and takes part of the Breast Research Group. He is also one of the coordinators of the Data Science Hub at INESC TEC. He is also working at the Computer Science Department of the Faculty of Sciences of the University of Porto as an Invited Assistant Professor. Between 2014 and 2016 he was contracted as Invited Assistant Professor at Informatics Engineering Department of FEUP. Previously between 2008 and 2011 was working as Invited Assistant in the same Faculty and Department. Hélder Oliveira is the principal investigator in 2 funded research projects (LuCaS, MICOS), project member in 4 projects (S-MODE, HEMOSwimmers, LEGEM and TAMI). In the past was also project member in 5 other funded projects (one European and 4 National) and 3 other as research assistant. He was also responsible at INESC TEC for other 2 projects related with technological transfer with industry, the project Evo3DModel with Adapttech - Adaptation Technologies and the project FollicleCounter with Saúde Viável. He was the founder member and coordinator (between 2010 and 2013) of the Bio-related Image Processing and Analysis Student’s Group (BioStar) at FEUP. Since 2007 I have co-authored 20 peer-reviewed papers and 8 journal abstracts. I have 1 patent conceded (Europe, China, Japan), 3 book chapters and also 64 works in international conferences, 40 articles in national refereed conferences and participated in the creation of 3 public datasets. In total, these publications have attracted 748 citations, with h-index of 14 according to Harzing’s Publish or Perish application on March 30, 2021. He was one of the mentors and belonged to the organizer committee of the VISion Understanding and Machine Intelligence (VISUM) summer school in 6 editions of the event. He also participated in the organization of other 12 events and was invited as keynote speaker in 3 international events. Hélder Oliveira is currently supervising 6 PhD Students, and has 1 Phd Student concluded as supervisor in 2018. During his career supervised (or co-supervised) 56 MSc students. Currently supervises 4 research fellows in projects at INESC TEC. Hélder Oliveira participated as principal jury in 2 PhD and 15 MSc defences as principal examiner. Hélder Oliveira is member of Portuguese Association of Pattern Recognition (APRP) and was been elected for president of the fiscal council in 2017. His research interests include medical image analysis, bio-image analysis, computer vision, image and video processing, machine learning, data science, computer science, programming, and 3D modelling.

    Interest
    Topics
    Explore Interest Topics
    Details

    Details

    • Name

      Hélder Filipe Oliveira

    • Role

      Assistant Researcher

    • Since

      01st November 2008
    023
    Publications
    View all Publications

    2026

    Optimizing Medical Image Captioning with Conditional Prompt Encoding

    Authors
    Fernandes, RF; Oliveira, HS; Ribeiro, PP; Oliveira, HP;

    Publication
    PATTERN RECOGNITION AND IMAGE ANALYSIS, IBPRIA 2025, PT II

    Abstract
    Medical image captioning is an essential tool to produce descriptive text reports of medical images. One of the central problems of medical image captioning is their poor domain description generation because large pre-trained language models are primarily trained in non-medical text domains with different semantics of medical text. To overcome this limitation, we explore improvements in contrastive learning for X-ray images complemented with soft prompt engineering for medical image captioning and conditional text decoding for caption generation. The main objective is to develop a softprompt model to improve the accuracy and clinical relevance of the automatically generated captions while guaranteeing their complete linguistic accuracy without corrupting the models' performance. Experiments on the MIMIC-CXR and ROCO datasets showed that the inclusion of tailored soft-prompts improved accuracy and efficiency, while ensuring a more cohesive medical context for captions, aiding medical diagnosis and encouraging more accurate reporting.

    CloseRead Abstract

    2026

    Enhancing Medical Image Analysis: A Pipeline Combining Synthetic Image Generation and Super-Resolution

    Authors
    Sousa, P; Campai, D; Andrade, J; Pereira, P; Goncalves, T; Teixeira, LF; Pereira, T; Oliveira, HP;

    Publication
    PATTERN RECOGNITION AND IMAGE ANALYSIS, IBPRIA 2025, PT II

    Abstract
    Cancer is a leading cause of mortality worldwide, with breast and lung cancer being the most prevalent globally. Early and accurate diagnosis is crucial for successful treatment, and medical imaging techniques play a pivotal role in achieving this. This paper proposes a novel pipeline that leverages generative artificial intelligence to enhance medical images by combining synthetic image generation and super-resolution techniques. The framework is validated in two medical use cases (breast and lung cancers), demonstrating its potential to improve the quality and quantity of medical imaging data, ultimately contributing to more precise and effective cancer diagnosis and treatment. Overall, although some limitations do exist, this paper achieved satisfactory results for an image size which is conductive to specialist analysis, and further expands upon this field's capabilities.

    CloseRead Abstract

    2026

    Abnormal Human Behaviour Detection Using Normalising Flows and Attention Mechanisms

    Authors
    Nogueira, AFR; Oliveira, HP; Teixeira, LF;

    Publication
    PATTERN RECOGNITION AND IMAGE ANALYSIS, IBPRIA 2025, PT I

    Abstract
    The aim of this work is to explore normalising flows to detect anomalous behaviours which is an essential task mainly for surveillance systems-related applications. To accomplish that, a series of ablation studies were performed by varying the parameters of the Spatio-Temporal Graph Normalising Flows (STG-NF) model [3] and combining it with attention mechanisms. Out of all these experiments, it was only possible to improve the state-of-the-art result for the UBnormal dataset by 3.4 percentual points (pp), for the Avenue by 4.7 pp and for the Avenue-HR by 3.2 pp. However, further research remains urgent to find a model that can give the best performance across different scenarios. The inaccuracies of the pose tracking and estimation algorithm seems to be the main factor limiting the models' performance. The code is available at https://github.com/AnaFilipaNogueira/Abnormal-Human-Behaviour-Detection- using-Normalising-Flows-and- Attention-Mechanisms.

    CloseRead Abstract

    2025

    Radiogenomic Insights from a Portuguese Lung Cancer Cohort: Foundations for Predictive Modeling

    Authors
    Neves, I; Freitas, C; Lemos, C; Oliveira, HP; Hespanhol, V; França, M; Pereira, T;

    Publication
    Measurement and Evaluations in Cancer Care

    Abstract

    2025

    MYCN-Amplified Neuroblastoma Detection Radiomics Vs. Trainable Features

    Authors
    Malafaia, M; Silva, F; Carvalho, DC; Martins, R; Dias, SC; Torrão, H; Oliveira, HP; Pereira, T;

    Publication
    2025 IEEE 25th International Conference on Bioinformatics and Bioengineering (BIBE)

    Abstract