Artificial Intelligence-Assisted Cancer Screening Advances in Radiology, Histopathology, and Liquid Biopsy

Abstract

Background: Cancer screening plays a crucial role in early diagnosis and improved survival, yet conventional screening methods are limited by false-positive findings, false-negative results, interobserver variability, delayed reporting, and unequal access to expert interpretation. Objective: To review recent advances in artificial intelligence-assisted cancer screening, focusing on applications in radiology, histopathology, and liquid biopsy. Methodology: This narrative review summarized published evidence on the role of artificial intelligence, machine learning, and deep learning in cancer screening. The review focused on three major domains: AI-assisted radiology, AI-assisted histopathology, and AI-assisted liquid biopsy. Results: AI-assisted radiology has shown value in lesion detection, image classification, risk stratification, and workflow prioritization, particularly in breast and lung cancer screening. In histopathology, AI supports tumor detection, grading, biomarker quantification, margin assessment, and reduction of interobserver variability. In liquid biopsy, AI improves interpretation of circulating tumor DNA, circulating tumor cells, methylation markers, microRNAs, and other molecular biomarkers for early cancer detection and tumor-origin prediction. Conclusion: Artificial intelligence has considerable potential to strengthen cancer screening by improving early detection, diagnostic accuracy, efficiency, and individualized risk assessment.