New deep learning models have emerged as transformative tools in oncology, enhancing diagnostic precision and prognostic predictions. Studies have introduced AI frameworks that accurately predict 1p/19q co-deletion status in lower-grade gliomas and dynamically track carcinoembryonic antigen (CEA) trajectories to refine gastric cancer prognostics. These advancements demonstrate AI's potential to optimize personalized cancer treatment by enabling non-invasive, rapid, and accurate molecular assessments, paving the path for improved clinical intervention strategies.