Bouthaina Slika is a PhD student at the University of the Basque Country (UPV/EHU), affiliated with Ho Chi Minh City Open University. She is currently pursuing a PhD under the supervision of Dr. Fadi Dornaika. Her research focuses on deep learning methods for biomedical image analysis, with a particular emphasis on lung infection severity quantification using CXRs and CT scans. She is expected to defend her PhD by mid-2025.

During her doctoral studies, Bouthaina has focused on developing generalizable, AI-driven models for medical image analysis, with an emphasis on pulmonary disease severity assessment. Her work integrates diverse deep learning strategies, including Vision Transformer (ViT)-based and Mamba-based encoders, parallel architectures, multi-task learning, attention mechanisms, multi-modality training, and advanced data augmentation techniques. She has proposed several models for severity score prediction—such as ViTReg-IP, MViTReg-IP, PViTGAtt-IP, QCross-Att-PVT, and MPAttCXR-IP, along with various enhanced variants. These architectures are designed to perform deep regression tasks that predict clinically meaningful severity scores, helping assess patient condition objectively. Furthermore, her use of data augmentation addresses the challenge of limited annotated datasets, enhancing the robustness and generalizability of the models in real-world clinical scenarios.

In addition to her research, Bouthaina serves as a lab instructor in the Department of Biomedical Engineering at the Lebanese International University, where she teaches laboratory courses including acquisition and microcontrollers, signal processing, image processing, and medical and laboratory instrumentation. She engages in collaborative research with international institutions and contributes to the broader deep learning and biomedical engineering community through conference presentations and peer-reviewed publications. Her work reflects a strong commitment to advancing medical imaging and healthcare through scalable, intelligent, and data-driven solutions.