RAML LAB

09/21/2026

The RAML Lab team recently came together for our Welcome Dinner in St. Catharines, a chance to welcome our new members and spend some time together as a team.

It was a lovely evening of conversation, team bonding, and getting to know one another better. We are glad to have our newest members with us and look forward to the research, ideas, and collaborations ahead.

With Dr. Blessing Ogbuokiri, Chioma Kamalu, Hridoy Rahman, Jacob Robitaille, David Shodipo, David Martin, Rimon Paul, Adrien Belcastro, Victoria Udechukwu, Rishi Modi, and Jose Henriquez.

09/03/2026

RAML Lab had a strong presence at the 23rd IEEE Conference on Computational Intelligence in Bioinformatics and Computational Biology (IEEE CIBCB 2026), held from August 31 to September 2 at the University of Piraeus in Athens, Greece.

Representing Brock University, our researchers Jose Henriquez , Cole Corbett , and David Martin, alongside our Director, Dr. Blessing Ogbuokiri , shared research spanning fairness, interpretability, medical imaging, and responsible AI, while engaging with the broader computational intelligence and bioinformatics community.

We are pleased to have contributed to these discussions and to continue advancing research toward more equitable, trustworthy, and clinically responsible AI systems.

09/01/2026

Today at the 2026 IEEE Conference on Computational Intelligence in Bioinformatics and Computational Biology (IEEE CIBCB 2026) at the University of Piraeus in Athens, Greece, Jose Henriquez and Cole Corbett presented our paper, “Evaluating Cross-Hospital Fairness in Self-Supervised Chest X-ray Representations.”

A collaboration between the Bio-inspired Computational Intelligence Lab and the Responsible and Applied Machine Learning Laboratory (RAML LAB), the study examines how fairly self-supervised chest X-ray representations transfer across hospitals with different patient populations and imaging conditions. The findings show why subgroup-level evaluation remains important even when overall model performance appears strong.

The paper is co-authored by Jose Henriquez, Cole Corbett, Dr. Beatrice Ombuki-Berman, and Dr. Blessing Ogbuokiri. We are pleased to see this work shared with the CIBCB community and contributing to research on equitable and trustworthy clinical AI.

09/01/2026

Today at the 2026 IEEE Conference on Computational Intelligence in Bioinformatics and Computational Biology (IEEE CIBCB 2026) at the University of Piraeus in Athens, Greece, David Martin presented our paper, “Fair and Interpretable Multi-Marker AI for Heart Failure Risk Detection from Chest X-Rays.”

Co-authored by David Martin, Hridoy Rahman, Victoria Udechukwu, and Dr. Blessing Ogbuokiri, the research examines fairness and interpretability in AI-based heart failure risk detection, with particular attention to model performance across patient age and s*x groups. The work also uses interpretability techniques to better understand what influences model predictions from chest X-rays.

We are proud to see this research presented to the CIBCB community today and to contribute to the broader effort toward developing healthcare AI systems that are effective, fair, interpretable, and responsible.

08/25/2026

🎉 Happy Birthday to our Director, Dr. Blessing Ogbuokiri!

Today, we celebrate your leadership, vision, and dedication to advancing responsible and applied machine learning. Thank you for your guidance, mentorship, and continued commitment to impactful research.

Wishing you a wonderful birthday and continued success in the year ahead! 🎂🎈

08/07/2026

First Runner-Up at Deep Learning Indaba 2026!

We are pleased to share that our poster, “Auditing Cross-Domain Stability of SHAP Explanations for Sentiment Classification: English Benchmarks and a Nigerian Case Study,” was awarded First Runner-Up at Deep Learning Indaba 2026 in Lagos, Nigeria.

The research examines the stability of SHAP explanations across domains and languages, including Nigerian Pidgin and Yoruba, contributing to ongoing work in explainable, responsible, and trustworthy AI.

Congratulations to the authors, Chioma Kamalu, Jennifer Iloekwe, and Dr. Blessing Ogbuokiri, on this recognition. We are grateful to Deep Learning Indaba and to the CIFAR Solution Network on AI Safety for supporting this research.

08/07/2026

08/07/2026

Media from today’s panel discussion at the 4th Trust AI Workshop at Deep Learning Indaba 2026 in Lagos, Nigeria.

Our Director, Dr. Blessing Ogbuokiri, joined Tobi Olatunji and Cecilia Mwende Mulu for the panel, “LLMs as Judges: How Trustworthy Are LLM-as-Judges? Implications for the African Continent,” moderated by Dalia Yousif.

The discussion explored the reliability of LLM-based evaluation, the role of human judgment and cultural context, and what trustworthy AI evaluation should look like in African settings. It was a thoughtful and engaging exchange with the audience.

08/03/2026

Today, our Director, Dr. Blessing Ogbuokiri, delivered the tutorial “Responsible and Sovereign AI: Fair, Trustworthy, and Context-Aware Machine Learning for Africa” at Deep Learning Indaba 2026 in Lagos, Nigeria.

The session brought together an engaged audience of researchers, practitioners, and students for a thoughtful discussion on fairness, trustworthiness, and the importance of developing AI systems that reflect African contexts and priorities.

We thank Deep Learning Indaba, CIFAR, and Brock University for supporting this important exchange, and we appreciate everyone who participated.