Abstract

In this talk, Ikechukwu Okechi Kamalu presents an uncertainty-aware and fairness-informed machine learning framework designed for workplace mental health screening in the technology sector. Drawing from multi-year tech-industry survey data, the presentation highlights how combining psychometric modelling, Bayesian machine learning, and post-processing fairness optimisations can address demographic imbalances and subjective reporting dynamics. Attendees will gain insights into key workplace indicators of psychological safety—such as peer-level communication pathways—and explore an interactive decision-support dashboard built to support equitable, trustworthy mental health interventions in organisational settings.

Speaker Biography

Ikechukwu Okechi Kamalu is a biostatistician and machine learning researcher specialising in probabilistic modelling, Bayesian statistics, and trustworthy AI. He holds an MSc in Biostatistics with a CGPA of 4.00/4.00 from Near East University, North Cyprus, where his research focused on reproducible machine learning pipelines for high stakes predictive analytics. He also holds a First Class Honours degree in Microbiology from the University of Abuja, Nigeria.

His current research focuses on bridging the gap between advanced predictive modelling, epistemic uncertainty quantification, and algorithmic fairness frameworks for healthcare and mental health decision-support systems. His work integrates psychometrics, Bayesian machine learning, causal inference, and interpretable AI methods for real-world clinical and organisational applications.

Join using this Teams Link.

Please contact maudsley.brc@kcl.ac.uk if you have any questions.