Small and diminutive polyps: No cancer, no risk!

WCE polyp detection based on novel feature descriptor with normalized variance locality-constrained linear coding

PURPOSE

Wireless capsule endoscopy (WCE) has become an effective facility to detect digestive tract diseases. To further improve the accuracy and efficiency of computer-aided diagnosis system in the detection of intestine polyps, a novel algorithm is proposed for WCE polyp detection in this paper.

METHODS

First, by considering the rich color information of endoscopic images, a novel local color texture feature called histogram of local color difference (LCDH) is proposed for describing endoscopic images. A codebook acquisition method which is based upon positive samples is also proposed, generating more balanced visual words with the LCDH features. Furthermore, based on locality-constrained linear coding (LLC) algorithm, a normalized variance regular term is introduced as NVLLC algorithm, which considers the dispersion degree between k nearest visual words and features in the approximate coding phase. The final image representations are obtained from using the spatial matching pyramid model. Finally, the support vector machine is employed to classify the polyp images.

RESULTS

The WCE dataset including 500 polyp and 500 normal images is adopted for evaluating the proposed method. Experimental results indicate that the classification accuracy, sensitivity and specificity have reached 96.00%, 95.80% and 96.20%, which performances better than traditional ways.

CONCLUSION

A novel method for WCE polyp detection is developed using LCDH feature descriptor and NVLLC coding scheme, which achieves a promising performance and can be implemented in clinical-assisted diagnosis of intestinal diseases.

Leaving colorectal polyps in place can be achieved with high accuracy using blue light imaging (BLI)

Objectives

A negative predictive value of more than 90% is proposed by the American Society of Gastrointestinal Endoscopy Preservation and Incorporation of Valuable Endoscopic Innovations (PIVI) statement for a new technology in order to leave distal diminutive colorectal polyps in place without resection. To our knowledge, no prior prospective study has yet evaluated the feasibility of the most recently introduced blue light imaging (BLI) system for real-time endoscopic prediction of polyp histology for the specific endpoint of leaving hyperplastic polyps in place.

Aims

Prospective assessment of real-time prediction of colorectal polyps by using BLI.

Material and methods

In total, 177 consecutive patients undergoing screening or surveillance colonoscopy were included. Colorectal polyps were evaluated in real-time by using high-definition endoscopy and the BLI technology without optical magnification. Before resection, the endoscopist described each polyp according to size, shape and surface characteristics (pit and vascular pattern, colour and depression), and histology was predicted with a level of confidence (high or low).

Results

Histology was predicted with high confidence in 92.5% of polyps. Sensitivity of BLI for prediction of adenomatous histology was 92.68%, with a specificity and accuracy of 94.87 and 93.75%, respectively. Following the recommendation of the PIVI statement, positive and negative predictive values were calculated with values of 95 and 92.5%, respectively. Prediction of surveillance based on both US and European guidelines was correctly predicted in 91% of patients.

Conclusion

The most recently introduced BLI technology is accurate enough to leave distal colorectal polyps in place without resection. BLI also allowed for assignment of postpolypectomy surveillance intervals. This approach therefore has the potential to reduce costs and risks associated with the redundant removal of diminutive colorectal polyps.