I joined Université Paris-Saclay in September 2026 as a Machine Learning Research Scientist through the France-Ghana FEF-AI4SD project. My research focuses on making medical image segmentation aware of both the dataset and the hardware on which it runs.
My work includes HALO-nnUAP, an automated framework for 2D, 3D, and 4D medical image segmentation. The experimental pipeline covers evaluation across multiple datasets, cross-validation, computational complexity analysis, GPU memory profiling, inference latency, and reproducible experiment tracking with MLflow.
I also develop computational spectroscopy models using FFT and power spectral density analysis to characterize wavelength-dependent spectral signatures. This work is supervised by Professor Daniel Adjei.