Combined assessment using optical colonoscopy and computed tomographic colonography improves the determination of tumor location and invasion depth

Density clustering-based automatic anatomical section recognition in colonoscopy video using deep learning

Recognizing anatomical sections during colonoscopy is crucial for diagnosing colonic diseases and generating accurate reports. While recent studies have endeavored to identify anatomical regions of the colon using deep learning, the deformable anatomical characteristics of the colon pose challenges for establishing a reliable localization system. This study presents a system utilizing 100 colonoscopy videos, combining density clustering and deep learning. Cascaded CNN models are employed to estimate the appendix orifice (AO), flexures, and "outside of the body," sequentially. Subsequently, DBSCAN algorithm is applied to identify anatomical sections. Clustering-based analysis integrates clinical knowledge and context based on the anatomical section within the model. We address challenges posed by colonoscopy images through non-informative removal preprocessing. The image data is labeled by clinicians, and the system deduces section correspondence stochastically. The model categorizes the colon into three sections: right (cecum and ascending colon), middle (transverse colon), and left (descending colon, sigmoid colon, rectum). We estimated the appearance time of anatomical boundaries with an average error of 6.31 s for AO, 9.79 s for HF, 27.69 s for SF, and 3.26 s for outside of the body. The proposed method can facilitate future advancements towards AI-based automatic reporting, offering time-saving efficacy and standardization.

Location of colorectal cancer: colonoscopy versus surgery. Yield of colonoscopy in predicting actual location

BACKGROUND AND STUDY AIMS

 Recent studies suggest that differences in biological characteristics and risk factors across cancer site within the colon and rectum may translate to differences in survival. It can be challenging at times to determine the precise anatomical location of a lesion with a luminal view during colonoscopy. The aim of this study is to determine if there is a significant difference between the location of colorectal cancers described by gastroenterologists in colonoscopies and the actual anatomical location noted on operative and pathology reports after colon surgery.

PATIENTS AND METHODS

 A single-center retrospective analysis of colonoscopies of patient with reported colonic masses from January 2005 to April 2014 (n = 380) was carried. Assessed data included demography, operative and pathology reports. Findings were compared: between the location of colorectal cancers described by gastroenterologists in colonoscopies and the actual anatomical location noted on operative reports or pathology samples.

RESULTS

 We identified 380 colonic masses, 158 were confirmed adenocarcinomas. Of these 123 underwent surgical resection, 27 had to be excluded since no specific location was reported on their operative or pathology report. An absolute difference between endoscopic and surgical location was found in 32 cases (33 %). Of these, 22 (23 %) differed by 1 colonic segment, 8 (8 %) differed by 2 colonic segments and 2 (2 %) differed by 3 colonic segments.

CONCLUSION

 There is a significant difference between the location of colorectal cancers reported by gastroenterologists during endoscopy and the actual anatomical location noted on operative or pathology reports after colon surgery. Endoscopic tattooing should be used when faced with any luminal lesions of interest.