Application of Artificial Intelligence Technology in Automatic Detec-tion of large Intestine Polyps
Abstract
Objective: to establish an automatic colonoscopy method based on artificial intelligence. Methods: a public database established by a university hospital was used, including colorectal fat and data collection. Initially, all frames in the video are normalized to reduce the high variability between databases. Then, the convolution neural network is used for full depth learning to complete the detection task of polyps. The network starts with the weights learned from millions of natural images in the ImageNet database. According to the fine-tuning technology, the colonoscopy image is used to update the network weight. Finally, the detection of polyps is performed by assigning the probability of containing Po ́ lipo to each table and determining the threshold defined when polyps appears in the table. Results: 1875 cases were collected from 5 public databases and databases established by university hospitals, with a total of 123046 forms. The method was trained and evaluated. Comparing the results with the scores of different colonoscopy experts, the accuracy was 0.77, the sensitivity was 0.89, the specificity was 0.71, and the ROC curve (re ceiver operation characteristics) was 0.87. Conclusion: compared with experienced gastrointestinal markers, this method overcomes the high variability of different types of lesions and different colonic light conditions (handle, folding or contraction), has very high sensitivity, and can reduce human errors, which is one of the main factors leading to the non detection or leakage of Po lipids in colonoscopy.
Keywords
Full Text:
PDFReferences
1. Bray F, Ferlay J, Soerjomataram I, et al. Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality for 36 cancers in 185 countries. CA Cancer J Clin 2018; 68(6): 394–424.
2. Data I, Method L. Globocan Colombia 2018. 2018; 380: 1–2.
3. Samadder NJ, Curtin K, Tuohy TM, et al. Charac-teristics of missed or interval colorectal cancer and patient survival: A population-based study. Gas-troenterology 2014; 146(4): 950–960.
4. Katenbach T, Sano Y, Friedland S, et al. American gastroenterological association. American Gas-troenterological Association (AGA) institute technology assessment on image-enhanced en-doscopy. Gastroenterology 2008; 134(1): 327–340.
5. Brown SR, Baraza W, Din S, et al. Chromoscopy versus conventional endoscopy for the detection of polyps in the colon and rectum. Cochrane Da-tabase Syst Rev 2016; 4: CD006439.
6. The Paris Endoscopic Classification of Superficial Tumor Lesions: Esophagus, Stomach and Colon: November 30 to December 1, 2002. Gastrointes-tinal 2003; 58 (6 supplement): s3-43.
7. Dinerson L, Chua TJ, Kaffees AJ. Meta-analysis of narrowband imaging versus conventional colon-oscopy for adenoma detection. Gastrointest En-dosc 2012; 75(3): 604–611.
8. Nagorni A, Bjelakovic G, Petrovic B. Narrow band imaging versus conventional white light colon-oscopy for the detection of colorectal polyps. Cochrane Database Syst Rev 2012; 1: CD008361.
9. Jin X, Chai S, Shi J, et al. Meta-analysis for eval-uating the accuracy of endoscopy with nar-row band imaging in detecting colorectal adeno-mas. Journal of Gastroenterology and Hepatology 2012; 27(5): 882–887.
10. Komenda Y, Suzuki N, Sarah M, et al. Factors associated with failed polyp retrieval at screening colonoscopy. Gastrointest Endosc 2013; 77(3): 395–400.
11. Choi HN, Kim HH, Oh JS, et al. Factors influ-encing the miss rate of polyps in a tandem colon-oscopy study. Korean J Gastroenterol 2014; 64(1): 24–30.
12. van Rijn JC, Reitsma JB, Stoker J, et al. Polyp miss rate determined by tandem colonoscopy: A systematic review. Am J Gastroenterol 2006; 101(2): 343–350.
13. Bernal J, Sanchez J, Vilarino F. Towards automatic polyp detection with a polyp appearance model. Pattern Recognition 2012; 45(9): 3166–3182.
14. Younghak S, Balasingham I. Comparison of hand-craft feature based SVM and CNN based deep learning framework for automatic polyp classification. Annu Int Conf IEEE Eng Med Biol Soc 2017; 2017: 3277–3280.
15. Urban G, Tripathi P, Alkayali T, et al. Deep learn-ing Localizes and identifies polyps in real yime with 96% accuracy in screening colonoscopy. Gastroenterology 2018; 155(4): 1069–1078.
16. Taha B, Werghi N, Dias J. Automatic polyp detec-tion in endoscopy videos: A survey. Biomed Eng 2017.
17. Szegedy C, Vanhoucke V, Ioffe S, et al. Rethinking the inception architecture for computer vision. Proc IEEE Comput Soc Conf Comput Vis Pattern Recognit. 2016; 2818–2826.
18. Simonyan K, Zisserman A. Very deep convolu-tional networks for large-scale image recognition. ICLR 2015; 1–14.
19. Zhang X, Zhang X, Li S, et al. Deep residual learning for image recognition. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2016. p. 770–778.
20. Tajbakhsh N, Gurudu SR, Liang J. Automated polyp detection in colonoscopy videos using shape and context information. IEEE Trans Med Imaging 2016; 35(2): 630–644.
21. Bernal J, Sanchez FJ, Fernandez-Esparrach G, et al. WM-DOVA maps for accurate polyp highlighting in Colonoscopy: Validation vs. Saliency maps from physicians. Comput Med Imaging Graph 2015; 43: 99–111.
22. Silva J, Histace A, Romain O, et al. Toward em-bedded detection of polyps in WCE images for early diagnosis of colorectal cancer. Int J Comput Assist Radiol Surg. 2014; 9(2): 283–293.
23. Freedman JS, Harari DY, Bamji ND, et al. The detection of premalignant colon polyps during colonoscopy is stable throughout the workday. Gastrointest Endosc 2011; 73(6): 1197–1206.
DOI: https://doi.org/10.32629/jai.v4i2.503
Refbacks
- There are currently no refbacks.
Copyright (c) 2021 Martin Alonso Gomez-Zuleta, iego Fernando Cano-Rosales, Diego Fernando Bravo-Higuera, Josue Andre Ruano-Balseca, Eduardo Romero-Castro
License URL: https://creativecommons.org/licenses/by-nc/4.0