Machine Learning Advancements for Lung Cancer Detection: An In-Depth Review and Future Prospects

Abstract

Lung cancer is a major cause of cancer-related mortality globally, underscoring the necessity for efficient diagnostic instruments. The review paper summarizes the role of ML and DL in detection, staging, and prognostication of lung cancer. We evaluate the relative efficacy of several models, such as CNNs, SVMs, and ensemble approaches, by analyzing publically accessible imaging and molecular information. We emphasize difficulties including class imbalance, model interpretability, and generalizability across clinical environments. We also talk about new trends that could improve clinical translation, like vision transformers, explainable AI, and federated learning. This interdisciplinary approach highlights the revolutionary potential of AI-driven techniques in lung cancer therapy and delineates critical future research directions to enhance clinical integration.