Advances in low-dose spectral computed tomography imaging for colorectal cancer

Abstract

This systematic review examines the technological principles and clinical applications of low-dose spectral computed tomography (CT) in colorectal cancer (CRC). Although spectral CT provides significant functional and quantitative insights beyond conventional anatomical imaging, the associated high radiation exposure necessitates the development of low-dose imaging protocols. This review synthesizes the current evidence on methods used to achieve acceptable image quality with reduced radiation dose, using techniques such as automatic tube current modulation and high-pitch scanning. A summary of the reviewed studies indicates that these low-dose protocols can maintain adequate diagnostic performance for key clinical tasks in CRC, including vascular visualization, tumor delineation, and the development of radiomics and deep learning-based diagnostic models. The emerging use of advanced reconstruction techniques—particularly artificial intelligence-based iterative reconstruction and deep learning image reconstruction algorithms—shows promise in supporting substantial dose reduction without compromising diagnostic confidence. In addition to advances in reconstruction algorithms, photon-counting CT represents a promising future direction owing to its inherently higher dose efficiency and spatial resolution. Continued research in radiomics and deep learning models is also pivotal to the future of medical imaging, as these approaches hold strong potential to enhance diagnostic accuracy, support individualized treatment planning, and advance precision medicine in CRC imaging.