Artificial intelligence enabled tumor diagnosis and treatment: Status, breakthroughs and challenges

In recent years, cancer has emerged as a leading threat to global health, underscoring the critical need for early detection, precise diagnosis, and effective treatment to enhance patient outcomes. Against this backdrop, the rapid adoption of artificial intelligence (AI) in oncology has been driven by three key factors: the proliferation of large-scale datasets, breakthroughs in computational hardware and algorithms, and the development of innovative deep learning (DL) architectures. Specifically, AI applications now span the entire oncology workflow, including image-based screening, pathological diagnosis, intelligent decision support, treatment outcome prediction, and personalized therapy design. Hence, we aim to synthesize the recent explosive growth in AI applications across the oncology continuum and to critically examine the translational challenges that hinder clinical deployment. More importantly, this article provides a timely update on cutting-edge advancements (e.g., multimodal learning, explainable AI and AI-driven drug discovery) and offers a forward-looking perspective on future directions, consolidating actionable insights for researchers and clinicians and paving the way for next-generation AI-driven precision oncology.