FADIA BAISSI

-Designed CNN-based detection models (malaria, pneumonia, bone age) with >95%+ performance metrics (AUC/F1) and strong cross-validation. -Built Flask model APIs and integrated with Node.js backend; productionized with MongoDB Atlas, Docker, and CI/CD. -Partnered with clinicians to define labeling protocols, validation criteria, and UI/UX that match medical workflows. -Implemented monitoring, logging, and model versioning; led security hardening and privacy guidance for healthcare use. -Earned innovative startup label; presented work at international venues.

-Led experiments with CNN architectures (custom + VGG16/19) for malaria cell classification; optimized with early stopping, checkpoints, and rigorous metrics. -Curated and balanced Red Blood Cells image datasets; standardized preprocessing pipelines. -Authored peer-reviewed publications; presented results to academic and clinical audiences.