Extracting quantitative features from CT and medical images to predict cancer treatment outcomes, toxicities, and biomarker statuses using machine learning pipelines.
Developing CNNs, Vision Transformers, U-Nets, and Foundation Models for image segmentation, classification, prognosis, and survival analysis in oncology.
Multi-modal data integration combining CT imaging, digital pathology, genomics, and clinical records for comprehensive predictive cancer models.
Computational identification of protein modification sites, disease-associated biomarkers, and AI-driven genomics pipelines for precision medicine.
Applying and evaluating large-scale pre-trained models for biomedical imaging interpretation, clinical NLP, and multimodal cancer research tasks.
Predicting PD-L1 expression and immune checkpoint inhibitor-induced toxicities such as pneumonitis from baseline CT scans using deep learning.
Anomaly detection and prognostics in high-dimensional sensor datasets using attention-based deep learning for industrial and healthcare applications.
Smart sensing systems, health monitoring platforms, and IoT-based intelligent infrastructure using edge computing and sensor fusion.