Summary
Burhan Hussein is an R&D engineer with a decade of hands-on experience in AI and medical imaging, currently building production-grade deep learning pipelines for mammography at Hera-MI. He combines research rigour from a PhD and postdoc—where he developed novel MS lesion segmentation methods—with practical engineering skills in TensorFlow/PyTorch, Docker, MLflow and GitLab CI to deliver reproducible, auditable solutions. At Hera-MI he has driven multitask and multimodal models that boosted internal ROC AUC from ~0.84 to ~0.93 and implemented Bayesian post-processing that cut false positives by 30% with minimal sensitivity loss. Comfortable across the full ML lifecycle, he pairs dataset curation and failure-mode augmentation with principled evaluation for imbalanced clinical data. Based in Rennes, he brings a cross-disciplinary mindset—bridging academia, industry and clinical collaborators—and a continual-learning ethos evident from diverse projects ranging from herbarium digitization to banking systems.
10 years of coding experience
6 years of employment as a software developer
Bachelor of Science Computer Science, Computer Software Engineering, upper second class, Bachelor of Science Computer Science, Computer Software Engineering, upper second class at ruaha catholic university
Artificial Intelligence, Completed, Artificial Intelligence, Completed at Coursera
Master in computing and information system, Computer Science, Master in computing and information system, Computer Science at Universiti Teknologi Brunei
Doctor of Philosophy - PhD, Computer Science, Awarded, Doctor of Philosophy - PhD, Computer Science, Awarded at Universiti Brunei Darussalam
English, Arabic, Swahili, French