???global.info.a_carregar???
My name is Diogo Araújo, and I am a dedicated researcher currently affiliated with the Signal and Image Processing Group at the Institute of Systems and Robotics, University of Lisbon. I completed my Master's degree in Electrical and Computer Engineering at Instituto Superior Técnico, University of Lisbon, specializing in Control, Robotics, and Artificial Intelligence. Under the guidance of Professors Catarina Barata and Carlos Santiago, my master’s thesis introduced innovative methods using Vision Transformers and Multiple Instance Learning to identify key patches in dermoscopy images. The success of my thesis and its continuation in the MIA-Breast project under the supervision of Professor Jacinto Nascimento led to a paper accepted for the DEF-MIA-AI workshop at CVPR2024, further validating our research approach. Currently, we are advancing this work by investigating advanced multimodal diagnostic techniques for breast cancer, with the aim of presenting our new findings at NeurIPS2024. My academic and professional journey has been marked by consistent engagement with complex challenges, from top placements in programming competitions to the development of machine learning frameworks for business efficiency in various internships. These experiences have honed my skills in machine learning, particularly in medical image processing, which I am eager to continue developing.
Identification

Personal identification

Full name
Diogo José Pereira Araújo

Citation names

  • Araújo, Diogo
  • Diogo José Pereira Araújo
  • Araújo, D. J. P.

Author identifiers

Ciência ID
C51A-0E43-E3E1
ORCID iD
0009-0006-7709-0704

Email addresses

  • diogoparaujo@tecnico.ulisboa.pt (Professional)
  • diogo.jpa.99@gmail.com (Personal)

Telephones

Mobile phone
  • 964193234 (Personal)

Websites

Knowledge fields

  • Engineering and Technology - Electrotechnical Engineering, Electronics and Informatics - Robotics and Automatic Control

Languages

Language Speaking Reading Writing Listening Peer-review
Portuguese (Mother tongue)
English Advanced (C1) Advanced (C1) Advanced (C1) Advanced (C1) Advanced (C1)
Education
Degree Classification
2021/09 - 2023/12
Concluded
Engenharia Electrotécnica e de Computadores (Mestrado)
Major in Control, Robotics, and Artificial Intelligence.
Universidade de Lisboa Instituto Superior Técnico, Portugal
"Automatic Identification of Regions Of Interest Using Vision Transformers and Multiple Instance Learning." (THESIS/DISSERTATION)
17/20
2018/09 - 2021/07
Concluded
Engenharia Electrotécnica e de Computadores (Licenciatura)
Universidade de Lisboa Instituto Superior Técnico, Portugal
15/20
2017/09 - 2018/07
Attended
Gestão (Licenciatura)
ISCTE Business School, Portugal
15/20
Affiliation

Science

Category
Host institution
Employer
2023/09/01 - Current Researcher (Research) Universidade de Lisboa Instituto de Sistemas e Robótica, Portugal
Universidade de Lisboa Instituto de Sistemas e Robótica, Portugal

Others

Category
Host institution
Employer
2023/03 - 2023/09 Intern: Machine Learning Research and Development GLS-Portugal, Germany
GLS-Portugal, Germany
2023/01 - 2023/03 Intern: Full-stack Developer Premium Minds, Portugal
Premium Minds, Portugal
2022/06 - 2022/09 Intern: Quality Assurance (QA) Sky UK Limited, United Kingdom
Sky UK Limited, United Kingdom
2021/06 - 2021/09 Intern: IT & Programming. Celfinet, Portugal
Celfinet, Portugal
Projects

Grant

Designation Funders
2023/09 - Current Multiple Instance Attention Learning for Multimodal Breast Cancer Diagnosis (MIA-BREAST)
BL228
1801P.01295.1.01
Research Fellow
Universidade de Lisboa Instituto de Sistemas e Robótica, Portugal
Fundação para a Ciência e a Tecnologia
Ongoing

Contract

Designation Funders
2023/09 - Current Multiple Instance Attention Learning for Multimodal Breast Cancer Diagnosis (MIA-BREAST)
1801P.01295.1.01
Research Fellow
Associação do Instituto Superior Técnico para a Investigação e Desenvolvimento, Portugal
Fundação para a Ciência e a Tecnologia
Ongoing
Outputs

Publications

Conference paper
  1. Araújo, Diogo; Verdelho, MR; Bissoto, Alceu; Jacinto C Nascimemto; Santiago, Carlos; Barata, Catarina. "Key Patches Are All You Need: A Multiple Instance Learning Framework For Robust Medical Diagnosis". Paper presented in CVPR 2024: Workshop on Domain adaptation, Explainability, Fairness in AI for Medical Image Analysis (DEF-MIA-AI 2024), Seattle, Washington, USA, 2024.
    Accepted
  2. Mota, Tiago; Verdelho, MR; Araújo, Diogo; Bissoto, Alceu; Santiago, Carlos; Barata, Catarina. "MMIST-ccRCC: A Real World Medical Dataset for the Development of Multi-Modal Systems". Paper presented in Data Curation and Augmentation in Enhancing Medical Imaging Applications Workshop @CVPR 2024 (DCAMI 2024), Seattle, Washington, USA, 2024.
    Accepted
Conference poster
  1. Araújo, Diogo. "Automatic Identification of Regions of Interest in Dermoscopy Images Using Vision Transformers and Weakly Supervised Learning". Paper presented in Laboratory of Robotics and Engineering Systems (LARSyS) Annual Meeting, 2023.
Thesis / Dissertation
  1. Diogo José Pereira Araújo. "Automatic Identification of Regions of Interest in Dermoscopy Images Using Vision Transformers and Weakly Supervised Learning". Master, Associação do Instituto Superior Técnico para a Investigação e Desenvolvimento, 2023. https://scholar.tecnico.ulisboa.pt/records/nz2nB67Wy3eo9isg4iNSHBzXdTaMOhgB1xBm.

Other

Other output
  1. WattAI. In collaboration with two colleagues, I am co-developing a smart HVAC system employing Deep Reinforcement Learning to dynamically balance energy efficiency and occupant comfort in smart buildings. 2023. Araújo, Diogo; Mesquita, Gonçalo; Silva, Duarte. https://github.com/GoncaloMesquita/WattAI.
  2. Image Classification Using Deep Learning Models. Engineered a training pipeline for deep learning models, achieving 92% balanced accuracy on the ISIC2019 dataset for image classification. 2023. Araújo, Diogo. https://github.com/diogojpa99/Image-Classification-Using-Deep-Models.
  3. Monte Carlo Localization of Pioneer 3D-X robot. Implemented a Monte Carlo localization algorithm using Python to autonomously locate a Pioneer 3D-X robot within a known map. 2022. Araújo, Diogo; Mesquita, Gonçalo; Melo, Sebastião; Cruz, Paulo. https://github.com/diogojpa99/Monte-Carlo-Localization-for-Pioneer-3D-X-Robot.
  4. BGP Stable State Simulator. Developed a modified version of the Bellman-Ford Algorithm to address BGP’s commercial and topological constraints. 2021. Araújo, Diogo. https://github.com/diogojpa99/BGP-Stable-State-Simulator.
Activities

Event participation

Activity description
Type of event
Event name
Institution / Organization
2023/07/17 - 2023/07/18 The LARSyS 2023 Annual Meeting was an institutional conference with a two-day program.
Meeting
2023 LARSyS Annual Meeting
Laboratório de Robótica e Sistemas de Engenharia, Portugal

Mentoring / Tutoring

Topic Student name
2023/09 - Current Academic Support for Undergraduate Students. Rita Pereira
2023/09 - Current Academic Support for Undergraduate Students. Rui Daniel
2023/09 - Current Academic Support for Undergraduate Students. Carolina Filipe
2023/09 - Current Academic Support for Undergraduate Students. Bernardo Santos
Distinctions

Award

2023 1st place at the 37th AI Cloudflight Coding Contest
Cloudflight Coding Contest (CCC), Germany
2022 3rd place in the 36th AI Cloudflight Coding Contest
Cloudflight Coding Contest (CCC), Germany

Other distinction

2021 Academic Merit Diploma 2020/2021
Universidade de Lisboa Instituto Superior Técnico, Portugal