Colorectal cancer (CRC) is the predominant condition of gastrointestinal malignancy and is a leading cause of morbidity and mortality. It is commonly acknowledged that adenomatous polyps (APs) cause 60–90 % of these malignancies [1]. Sessile serrated polyps (SSPs), also known as sessile serrated adenomas (SSA), and identified in the past decade as significant premalignant lesions, constitute 15–30 % of CRC cases [2]. Colonoscopy detection with polypectomy is the most important method for colon cancer prevention. An increase of 1.0 % in the adenoma detection rate correlates with a reduction of 3.0 % in cancer risk [3] The removal of adenoma polyps can reduce the incidence of CRC by up to 89 % [4]. However, white light image (WLI) endoscopy has a high polyp miss rate, especially for adenomas, ranging between 26 % and 47.2 % [5]. Many measures have been applied to improve adenoma detection rates, including optimal colon preparation before endoscopy and using image-enhanced endoscopy (narrow-band imaging [NBI], flexible spectral imaging color enhancement [FICE], and blue-light imaging [BLI]), wide-field-of-view lenses, or devices such as caps, Endocuffs, and EndoRings to observe behind mucosal folds [6]. Despite these efforts, the endoscopy technique still has notable disadvantages, such as a high polyp miss rate, especially for small polyps, polyps located in the right colon, and SSPs [7].

Detecting right-sided polyps can be more challenging due to the shape of the colonic folds and the need for a complete colonoscopy. Right-sided lesions can indicate an increased risk of recurrence of advanced adenomas. Right-sided colon polyps are 1.2 times likelier to be adenomatous, whereas left-sided polyps are 2.5 times likelier to be hyperplastic. Right-sided polyps are predominantly sessile, making their removal more challenging compared to pedunculated polyps, which are more frequently found on the left side. Right-sided polyps exhibit a more aggressive nature than left-sided polyps due to their poorer histological characteristics and sessile morphology [8]. Serrated polyps are more common in the right colon and are often sessile or flat. They can be covered with mucus, and can have a benign appearance. All of these factors make right-sided colon polyp detection challenging for endoscopists.

Recently, a few studies have shown the application of artificial intelligence (AI) in supporting the detection of colorectal polyps, with an accuracy of 96 % [9]. Although the polyp detection rate has increased, the most important issue is to avoid unnecessarily removing hyperplastic polyps, as it increases medical costs, wastes time, and increases the risk of bleeding [10]. Therefore, prediction of the histopathology of colorectal polyps during real-time endoscopy is of significant clinical importance, as it guides the selection of appropriate treatments.

Advanced endoscopy systems that include linked-color imaging (LCI) and computer-aided detection have been applied to increase the polyp detection rate and predict histopathology during real-time endoscopy [11]. LCI is an image-enhanced endoscopy technique that helps endoscopists observe the colon mucosa more clearly by enhancing the color contrast of the lesion with the surrounding normal mucosa and is based on the distribution of vascularity in the lesion [12,13]. Neoplastic lesions involve the development of vessels in the submucosa, and with LCI, the red light will not be absorbed but will be reflected instead, allowing us to see images with enhanced red signals, which indicate neoplastic polyps. Moreover, when magnifying lesions, LCI facilitates the observation of the microstructure and microvascularity of polyps, thereby permitting AI to predict the histopathology of those polyps. Recently, a novel computer-aided diagnostic (CAD) system for endoscopy, termed CAD EYE, has been developed utilizing AI deep learning technology to facilitate real-time detection of colon polyps [11].

In Vietnam, colorectal cancer is one of the five most common cancers and is predicted to be the most common cancer in men and the third most common cancer in women by 2025 [14]. The optimization of endoscopic detection and polypectomy is important for preventing this type of cancer. Research on the application of LCI combined with CAD EYE in the detection and prediction of the histology of colorectal polyps is still limited. We conducted this study to evaluate the diagnostic value of LCI combined with CAD EYE in the detection and characterization of right-sided colon polyps in comparison with WLI colonoscopy.