Pedro C. Neto

Pedro C. Neto

AI scientist at Unilabs and Invited Assistant Professor at FEUP, University of Porto

I take medical AI from research to production, real-world use: models trained on real data, built to meet MDR, IVDR and GDPR. My research covers diagnostic AI for pathology and face recognition that is explainable, fair and private.

Publications on Google Scholar Invite me to speak or advise

About

Pedro C. Neto (Pedro Carneiro Neto) is a Portuguese artificial intelligence scientist. At Unilabs he builds AI models for the medical ecosystem, from data and training through regulatory compliance and privacy, and he is an Invited Assistant Professor at the Faculty of Engineering of the University of Porto (FEUP), where he teaches machine learning.

He holds a PhD in Electrical and Computer Engineering from FEUP, with a thesis on explainability, fairness and privacy in AI for biometrics, an MSc in Computer Science from Aalto University in Finland, and a BSc in Informatics Engineering from ISEP. Before Unilabs, he spent four years as a researcher at INESC TEC, where he co-developed an AI system for colorectal cancer diagnosis from whole-slide images that was covered by national and international press.

In 2024 the European Association for Biometrics gave him the Max Snijder Award for his work on interpretable face-matching confidence.

Em português

Pedro C. Neto é cientista de inteligência artificial na Unilabs e Professor Auxiliar Convidado na Faculdade de Engenharia da Universidade do Porto. Doutorado pela FEUP, investiga IA aplicada à saúde, em particular patologia digital, e reconhecimento facial explicável e justo. Escreve e fala regularmente sobre IA na saúde para o grande público.

At a glance

Now
AI Scientist, Unilabs
Invited Assistant Professor, FEUP
PhD
FEUP, 2025
Award
EAB Max Snijder Award, 2024
Open to
Talks, advisory roles, media commentary, research collaboration
Fields
Computational pathology, biometrics, explainable and fair AI
Based in
Porto, Portugal
Languages
Portuguese, English

Medical AI in practice

At Unilabs I work where most medical AI projects stall: turning a model that works on a benchmark into one that can be used safely in a regulated diagnostic network. The specific products are confidential, but the work spans three areas.

Models at population scale

Training and validating models on real data, across imaging, pathology and other clinical data. Generative AI with strong guardrails on non-clinical domains.

Regulation built in

Developing models with the evidence the EU Medical Device Regulation (MDR) and In Vitro Diagnostic Regulation (IVDR) require, from design documentation to performance evaluation.

Privacy by design

Data protection impact assessments (DPIAs) and GDPR-compliant data pipelines for sensitive health data. Self-Hosted Solutions for Large-Language Models deployment and integration on workflows.

Research

Medical AI and computational pathology

Diagnostic models for whole-slide images that work with the annotations pathologists can realistically provide.

Trustworthy biometrics

Face recognition that can explain its decisions, treats demographic groups fairly and can be trained without harvesting personal data.

More than 25 peer-reviewed papers, cited 906 times with an h-index of 18 on Google Scholar (October 2026). PhD thesis: Explainability, Fairness and Privacy in AI for Biometrics, University of Porto, 2025.

Recognition

European Biometrics Max Snijder Award, 2024

Awarded by the European Association for Biometrics in Darmstadt for PIC-Score, a comparison score that states the probability that a face match is correct and can combine several images optimally.

Read the EAB announcement

In the news

Writing and talks

Experience

  1. 2024 to present

    AI Scientist, Unilabs

    Medical AI at population scale, MDR and IVDR compliance, DPIAs.

  2. 2022 to present

    Invited Assistant Professor, FEUP

    Machine learning teaching and PhD supervision.

  3. 2020 to 2024

    Research Scientist, INESC TEC

    Colorectal cancer diagnosis (CADPath.AI) and explainable biometrics.

  4. 2020

    Summer Research Intern, Feedzai

  5. 2019 to 2020

    Research Assistant, Aalto University

Education

  1. 2025

    PhD, Electrical and Computer Engineering

    FEUP, University of Porto

  2. MSc

    Computer Science

    Aalto University, Finland

  3. BSc

    Informatics Engineering

    ISEP, Polytechnic of Porto

Tools

Python, PyTorch, HuggingFace, Transformers, LLM, scikit-learn, C++, SQL, Docker, Git and LaTeX, applied to whole-slide imaging, medical imaging and face biometrics.

Speaking and advisory

I have given talks on medical AI, explainable AI and fairness in face recognition to universities, clinicians, industry and the general public, in Portuguese and English.

I'm open to:

Talks
Keynotes, masterclasses, panels and seminars
Advisory
Medical AI strategy and responsible AI
Media
Commentary on AI in healthcare, biometrics and AI policy
Research
Academic and clinical collaborations

Send an enquiry

Teaching and supervision

Contact

For talks, advisory roles, media requests or research collaborations, email me. I usually reply within a few days.

pedro.carneiro.neto.21@gmail.com

For Unilabs matters: pedro.neto@unilabs.com