This editorial discusses the transformative potential of integrating dynamic full-field optical coherence tomography (D-FFOCT) with artificial intelligence (AI) in intraoperative breast cancer diagnosis. Traditional methods, such as frozen pathology, often face limitations in speed and accuracy, which can impact surgical outcomes. D-FFOCT, offering high-resolution, real-time imaging of tissue microstructures without ionizing radiation, presents a non-destructive alternative that maintains specimen integrity. Coupled with AI, particularly deep learning algorithms, this technology has demonstrated impressive diagnostic accuracy and speed, significantly reducing intraoperative margin evaluation time. Despite challenges in implementing these innovations, such as the need for high-quality datasets and addressing algorithmic bias, the integration of D-FFOCT and AI promises to enhance decision-making, alleviate the burden on pathologists, and improve patient outcomes. This approach not only aims to optimize breast cancer surgeries but also has broader implications for the diagnosis and treatment of other tumor types, highlighting the importance of ethical considerations and collaborative efforts in advancing clinical practice.
Keywords: Artificial intelligence; D-FFOCT; Intraoperative diagnosis; Machine Learning; breast cancer.