Advancements in real-time oncology diagnosis: harnessing AI and image fusion techniques

Abstract

Real-time computer-aided diagnosis using artificial intelligence (AI), with images, can help oncologists diagnose cancer with high accuracy and in an early phase. It explores various real-time techniques, encompassing technical solutions, AI-based imaging, and image fusion diagnosis. Techniques such as computer-aided surgical navigation systems and augmented reality platforms have improved the precision of minimally invasive procedures once combined with real-time imaging and robotic assistance. The integration of modalities like ultrasound into image fusion workflows improves procedural guidance, reduces radiation exposure, and provides high cross-modality interpretation. Optical imaging techniques—such as diffuse reflectance spectroscopy, Raman spectroscopy, fluorescence endoscopy, and hyperspectral imaging—are emerging as powerful diagnostic tools for tumor detection, margin assessment, and intraoperative decision-making. Promising methods like fluorescence confocal microscopy and shear wave elastography offer practical, real-time diagnostic capabilities. However, regarding these technologies, there are technical challenges including tissue motion, registration variability, and data imbalance. The incorporation of AI-based landmark detection and the development of robust algorithms will be key to overcoming these barriers. We close by offering a more futuristic overview to solve existing problems in real-time image-based cancer diagnosis. The reviewed technologies altogether mark that continued research, multi-center validation, and providing hardware accelerators will be crucial to their full clinical potential and usage.