Evaluation of AI algorithms for biomedical research and sex and gender biases in AI

Training details

Location

Barcelona, Spain (On-site)

Date

26/01/2026

Time

14 : 00

Target Audiance

Scientist

Teaching language(s)

English

Organizing institution

BSC

Delivery mode

On-site

Level

Introductory

Format

Case Study Session, Lecture

Capacity or seats limit

47

Industrial domains

Health, Pharma and biotech

Topics / Keywords

Sex, gender, bias, diversity, responsible artificial intelligence.

Rapid advances in artificial intelligence (AI) are transforming biomedical research, but these systems can internalize and amplify historical biases present in clinical trial data, annotations, and experimental design. Without rigorous evaluation, we risk developing and using models that misdiagnose, recommend ineffective treatments, and perpetuate existing inequalities.

This course focuses on addressing sex and gender biases throughout the entire lifecycle of AI systems applied to biomedical research, from project conceptualization and data generation to model training, showcasing practical cases and providing tools to identify, evaluate, and mitigate these biases.

What You Will Learn

  • Understanding key concepts and why they matter in biomedical research. 
  • Detecting specific biases in biomedical data and understanding their impact on research quality. 
  • Critically evaluating AI models using contextualized metrics and frameworks. 
  • Designing bias-aware biomedical AI projects from the outset.

 

Agenda

Monday 26: Block 1 – Fundamentals and Concepts (4h)

  • 14:00-16:00 Part 1 
  • 16:00-16:30 Break
  • 16:30-18:30 Part 2 

Tuesday 27: Block 2 – Identifying Biases in Training Data for AI (4h)

  • 14:00-16:00 Part 1 
  • 16:00-16:30 Break 
  • 16:30-18:30 Part 2 

Thursday 29: Block 3 – Detection of Bias and Evaluation in AI Models (4h) 

  • 14:00-16:00 Part 1
  • 16:00-16:30 Break
  • 16:30-18:30 Part 2

Friday 30: Block 4 – Project Design and Implementation of Best Practices (3h) 

  • 14:00-16:00 Part 1
  • 16:00-16:30 Break
  • 16:30-17:30 Part 2

Instructor name(s)

  • María Flores
  • Carolina Belver
  • Simón Perera
  • Davide Cirillo
  • Claudia Rosas
  • Olivier Phillipe
  • Barbara Fuzi
  • Summer Devlin
  • Salvador Capella (TBC)

Instructor’s biography

  • Claudia Rosas

Claudia is a computer scientist and AI researcher at the Barcelona Supercomputing Center, specializing in performance evaluation now on AI within health and biomedicine related applications. With a background spanning software engineering, product management, and AI research, she brings both technical expertise and practical experience in building and assessing responsible AI systems.

Passionate about Diversity, Equity, and Inclusion in technology, she focuses on developing rigorous evaluation frameworks that ensure AI systems operate fairly and effectively across diverse populations. Through this course, she aims to equip participants with critical skills for identifying and addressing AI evaluation; bridging technical innovation with critical thinkin

  • Davide Cirillo

Davide Cirillo is the head of the Machine Learning for Biomedical Research Unit at the Life Sciences Department of the Barcelona Supercomputing Center (BSC) and co-founder of the OneCareAI spinoff of BSC. He received the MSc degree in Pharmaceutical Biotechnology from University of Roma ‘La Sapienza’, Italy, and the PhD degree in Biomedicine from Universitat Pompeu Fabra (UPF) and Center for Genomic Regulation (CRG) of Barcelona, Spain. His research focuses on computational methods for precision medicine with a special emphasis on machine learning, network science, and ethics of artificial intelligence. He is a member of the ELIXIR Machine Learning Focus Group, co-leads the research subgroup of BSC Bioinfo4Women initiative, and is a scientific advisor to the Swiss non-profit Women’s Brain Foundation. He is co-editor of the book “Sex and Gender Bias in Technology and Artificial Intelligence: Biomedicine and Healthcare Applications” (Elsevier Academic Press, 2022).

  • Carolina Belver Aguilar

Carolina Belver is a researcher at the Barcelona Supercomputing Centre. She holds a PhD in Physics and has worked at international research centres such as CERN and the University of Bern in the field of particle accelerators applied to medicine. She later joined the industry, working with X-ray-based imaging technologies and artificial intelligence. She currently combines the fields of health and artificial intelligence in the international and multidisciplinary AHEAD project, which focuses on evaluating AI models in health from technical, social, legal and ethical perspectives. At the same time, she coordinates the communication group at Bioinfo4Women, an initiative of the Life Sciences Department at the BSC, which promotes the integration of sex and gender dimensions in biocomputing research and other fields.

  • María Morales Martínez

María Morales is a Research Scientist in the Social Link Analytics group at Barcelona Supercomputing Center. She holds a Master’s degree in Biochemistry and a Master’s degree in Biotechnology from the University of Málaga and the International University of Andalucia, as well as a Master’s degree in Bioinformatics Applied to Personalized Medicine and Health from the Carlos III Health Institute.

She has developed her professional career in the private sector, working at companies specialized in genetic diagnostics such as Genetaq and Imegen, where she focused on pipelines development, bioinformatic analysis, and web development. Subsequently, she worked as a researcher at the Josep Carreras Institute, focusing on sex-aware genomic analysis and research application development.

Currently, she is part of the Bioinfo4Women coordination group, dedicated to raising awareness and mitigating sex and gender biases in health. Her work focuses on evaluating artificial intelligence models from a multidisciplinary approach and analyzing health data and integrating technical, social, and gender dimensions.

  • Simón Perera del Rosario

BSc in Biotechnology and MSc in Biological Anthropology. Business Development Director at ProtoQSAR (SME specializing in computational chemistry, drug design and development, structural bioinformatics, and computational toxicology). Currently a PhD candidate at UPF (Lab of Genomics of Individuality, Institute of Evolutionary Biology). Co-director and curator of the cultural project “Una Mirada LGTBIQA+” (“A Queer Gaze”), with exhibitions, interventions, and trainings, in many museums and research centers, and codirector of the outreach event “BCNspiracy”. Director of the science outreach event “BCNspiracy” (since 2017, 1000 yearly attendants). Responsible for the European project “INCLUDE” (about intersectional gender equality plants) and former secretary general (2018-2025) at PRISMA “LGBTIAQ+ Science”. Previously served as a project manager and business developer at Anaxomics (bioinformatics, systems biology), associate professor at UAB (Dept. of Biological Anthropology), responsible for European projects and coordinator of Bioinformatics Barcelona (a consortium related to bioinformatics), and scientific communication technician at UPF (project “La ciència al teu món”). Board member of the Spanish Association for Science Communication (AEC2), the Catalan Association for Science Communication (ACCC) and the Biotechnology Communicators Association (ComunicaBiotec), and collaborator with the science outreach associations “Hablando de Ciencia” and “Scenio”.

  • Olivier Phillipe

Olivier Philippe is a Research Scientist at the Barcelona Supercomputing Center (BSC), specializing in computational techniques applied to social phenomena, with a focus on social networks and disinformation. His work seeks to understand human behavior through the lenses of psychology, sociology, and social networks.

Since 2025, he has been part of the Social Link Unit within the Life Sciences Department at BSC. His current projects include studying the dissemination of fake news related to health within international collaborations, analyzing gender issues in biological literature using large language models (LLMs), and investigating the impact of climate change in Catalonia on mental health. These interdisciplinary efforts integrate computational techniques with social and environmental sciences to address pressing global challenges.

From 2021 to 2024, Olivier worked as a Research Engineer in the same unit at BSC, focusing on developing data science approaches to explore and counteract the spread of disinformation.

In 2020 and 2021, he served as a Data Scientist for DATAPOP in the Department of Communication at the University of Pompeu Fabra. His work involved data mining and creating innovative methodologies to analyze social network communications on various political debates. He also deployed fine-tuned transformer models and clustering methods to derive insights from large datasets.

Course Description

This course will be divided into four blocks covering various aspects from conceptual foundations to the implementation of bias mitigation strategies:

Block 1: Fundamentals and Concepts. We will establish the theoretical basis by differentiating between biological sex and gender, exploring intersectionality, and introducing the types of AI models in health and how biases arise at each stage of their life cycle. 

Block 2: Identifying Biases in Training Data for AI. We will analyse specific case studies that may arise in biomedical research, incorporating a gender and LGBTIAQ+ perspective. We will address biases in clinical trials, pharmacology, cohort design, and problems of sex/gender annotation and inference in databases. We will also examine real projects that illustrate some of these challenges in biomedical data. 

Block 3: Detection of bias and Evaluation in A modeIs. We will learn statistical and equity metrics, interpretability tools, standardised evaluation frameworks, and their application.

Block 4: Project Design and Implementation of Best Practices. We will integrate a gender perspective into proposal design using regulatory guidelines and frameworks and learn how to write research projects. We will present a specific example applying the guidelines we have developed to consider these biases from the outset of the project, and we will conclude with a space for reflection on the course.

Prerequisites

<p><span style="font-weight: 400">Basic knowledge of artificial intelligence is desirable but not mandatory. Previous education or experience in biological sciences is not necessary, although it may facilitate engaging with the course content. </span></p>

Certificate/badge details

<p>Certificate of Achievement</p>