<p>Coastlines are vital geographical features that interact dynamically with land and sea, making their accurate mapping essential for coastal management, environmental monitoring and climate adaptation. Recent advances in remote sensing technology and the increasing availability of various data sources, such as satellite imagery, unmanned aerial vehicle (UAV) data and synthetic aperture radar (SAR) imagery, have enabled more accurate and dynamic extraction of the coastline. In this context, deep learning models have proven to be a powerful tool that can handle the complexity of shoreline detection in different coastal geomorphologies. This review provides a comprehensive analysis of the current literature and focuses on the development and application of deep learning techniques for shoreline extraction using remote sensing data. In particular, the study highlights the performance of models such as U-Net variants, Convolutional Neural Networks (CNNs) and hybrid approaches and emphasizes the impact of different input data types on model accuracy. The study also addresses important challenges such as the need for high-resolution datasets, the difficulty of dealing with different coastal environments, and the limitations of generalizing across different geographical regions. By analyzing the strengths and limitations of these models, this systematic review offers practical insights into existing methods and highlights key directions for future research.</p>