Wilson Silva #
Research Interest #
Development of AI models for multi-center trials and prospective validation (generalizability, privacy, and explainable AI). Leading the research line in AI for multi-center data.
Education #
- 2011-2014 BSc in Electrical and Computer Engineering, University of Porto;
- 2014-2016 MSc in Electrical and Computer Engineering, University of Porto;
- 2017-2022 PhD in Electrical and Computer Engineering (Machine Learing and Medical Image Analysis), University of Porto. Supervisors: Prof. Jaime S. Cardoso University of Porto, Prof. Maria Joao Cardoso (Champalimaud Foundation).
Work experience #
- 2016-2017 IT Advisor, KPMG Portugal, KPMG Portugal;
- 2018-2019 Invited assistant, Dept. of Electrical and Computer Engineering, University of Porto;
- 2019-2020 Visiting PhD Student, ARTORG Center for Biomedical Engineering Research University of Bern
- 2021-2022 Invited assistant, Dept. of Electrical and Computer Engineering, University of Porto;
- 2017-2022 PhD Student, University of Porto / INESC TEC;
- 2022-on Postdoctoral Fellow, Department of Radiology, nki.nl.
- 2023-on Assistant Professor in Explainable AI for Life, AI Technology for Life, Utrecht University
Selected Publications #
(Full list on Google Scholar)
Silva et al. “Attention-based Regularisation for Improved Generalisability in Medical Multi-Centre Data”. IEEE International Conference on Machine Learning and Applications (2023)
Corbetta et al. “Interpretability-guided Data Augmentation for Robust Segmentation in Multi-centre Colonoscopy Data”. Workshop on Machine Learning in Medical Imaging at MICCAI 2023
Silva et al. “Computer-aided diagnosis through medical image retrieval in radiology”. Nature Scientific Reports (2022)
Montenegro et al. “Disentangled Representation Learning for Privacy-preserving Case-based Explanations”. Workshop on Medical Applications with Disentanglements at (MICCAI 2022)
Montenegro et al. “Privacy-preserving Case-based Explanations: Enabling Visual Interpretability by Protecting Privacy”. IEEE Access (2022)
Silva et al. “Interpretability-Guided Content-Based Medical Image Retrieval”. International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI 2020)
Silva et al. “Towards Complementary Explanations Using Deep Neural Networks”. Workshop on Interpretability of Machine Intelligence in Medical Image Computing at (MICCAI 2018)
Cardoso et al. “Evolution, current challenges, and future possibilities in the objective assessment of aesthetic outcome of breast cancer locoregional treatment”. The Breast (2020)
Supervised PhD students, past and present #
PhD in Artificial Intelligence (Maastricht University) #
- Valentina Corbetta is working to improve generalizability of segmentation and classification algorithms in the context of medical multi-center data.
- Daan Boeke is exploring AI in the context of immune micro-environment characterization and treatment response prediction.
Supervised MSc students, past and present #
MSc in Artificial Intelligence (VU Amsterdam) #
- Melanie Groeneveld ongoing
- Laura Latorre “Towards Case-based Interpretability for Federated Learning Models”
MSc in Bioengineering (University of Porto) #
- Daniel Silva “Disentanglement Representation Learning for Generalisability in Medical Multi-centre Data”
- Isabela Miranda “Integrating Anatomical Prior Knowledgde for Increased Generalisability in Breast Cancer Multi-centre Data”
- Tiago Goncalves “Deep Aesthetic Assessment of Breast Cancer Surgery Outcomes” MSc Thesis
- Maria Carvalho “Towards Biometrically-morphed Medical Case-based Explanations” MSc Thesis
MSc in Informatics and Computer Engineering (University of Porto) #
- Filipe Campos ongoing
- Rita Mendes ongoing
- Margarida Vieira ongoing
- Helena Montenegro “A privacy-preserving framework for case-based interpretability in machine learning” MSc Thesis, CTM best Master’s Thesis award, APRP best Master’s Thesis award
MSc in Biomedical Engineering (University of Porto) #
- Diogo Mata “Biomedical multimodal explanations – increasing diversity and complementarity in Explainable Artificial Intelligence”, CTM best Master’s Thesis award
Links #
Email - Twitter - GitHub - ResearchGate - Scholar