Liver Couinaud Segmentation: A Review on Existing Segmentation Techniques of the Couinaud Classification of Liver

Abstract

Liver Couinaud segmentation is crucial in biomedical image processing for treatment planning and surgical interventions of the liver parenchyma. This is because the liver Couinaud segments help provide clinicians with a clearer and specific diagnosis of the location of the diseased areas within the liver tissue. Technological advancements in the form of machine learning (ML) and deep learning (DL) models have greatly enhanced the process of liver Couinaud segmentation. The efficiency of ML and DL models showcases their performance with incredible results. Hence, this article discusses the state-of-the-art techniques for liver Couinaud segmentation from 2006 to 2024. The segmentation techniques are divided into semi-automatic and automatic. Additionally, the online medical imaging datasets suitable for developing algorithms and frameworks for liver Couinaud segmentation are presented with their descriptions, advantages, and limitations. Furthermore, the performance metrics that are commonly used for evaluating the performance of segmentation algorithms are thoroughly discussed. Finally, the research challenges and recommendations are accentuated for future researchers who intend to bridge the highlighted gaps in liver Couinaud segmentation.