Artificial Intelligence in Histopathological Diagnosis: Achievements and Challenges of Digital Pathology

Artificial intelligence (AI) and digital pathology (DP) are transforming the landscape of diagnostic histopathology. This review provides an overview of how traditional pathology workflows-long reliant on physical slides and subjective interpretation-are being reshaped by digital slide acquisition, machine learning algorithms, and multi-omics integration. The article outlines the historical foundations of diagnostic pathology and the shift toward digitization, followed by an exploration of the practical challenges in implementing AI tools, including infrastructure needs, training requirements, and regulatory considerations. The clinical applications of AI-such as automated cancer detection, mutation prediction, and deep learning-based classification-are discussed alongside recent developments in federated learning and explainable AI (XAI). Particular attention is given to the integration of genomics and digital morphology, which allows for more accurate prognostic and predictive modeling. While the implementation of AI raises concerns regarding cost, standardization, and interpretability, the conclusion emphasizes that AI will serve as a powerful tool to support-not replace-the pathologist. By enhancing diagnostic reproducibility, reducing workload, and expanding the analytical scope of histopathology, AI has the potential to advance precision medicine and improve patient care.