Low Rates of Cancer or High-Grade Dysplasia in Colorectal Polyps Collected From Computed Tomography Colonography Screening

This chapter presents a comprehensive scheme for automated detection of colorectal polyps in computed tomography colonography (CTC) with particular emphasis on robust learning algorithms that differentiate polyps from non-polyp shapes. The authors’ automated CTC scheme introduces two orientation independent features which encode the shape characteristics that aid in classification of polyps and non-polyps with high accuracy, low false positive rate, and low computations making the scheme suitable for colorectal cancer screening initiatives. Experiments using state-of-the-art machine learning algorithms viz., lazy learning, support vector machines, and naïve Bayes classifiers reveal the robustness of the two features in detecting polyps at 100% sensitivity for polyps with diameter greater than 10 mm while attaining total low false positive rates, respectively, of 3.05, 3.47 and 0.71 per CTC dataset at specificities above 99% when tested on 58 CTC datasets. The results were validated using colonoscopy reports provided by expert radiologists.

Download Full-text

1182 OUTCOMES OF ENDOSCOPIC RESECTION FOR COLORECTAL POLYPS WITH HIGH GRADE DYSPLASIA OR INTRAMUCOSAL CANCER

Gastrointestinal Endoscopy ◽

10.1016/j.gie.2020.03.897 ◽

2020 ◽

Vol 91 (6) ◽

pp. AB108-AB109

Author(s):

Erik Almazan ◽

Sonmoon Mohapatra ◽

Kevan Salimian ◽

Saowanee Ngamruengphong

Keyword(s):

Endoscopic Resection ◽

Colorectal Polyps ◽

High Grade Dysplasia ◽

High Grade ◽

Intramucosal Cancer

Download Full-text

Small Colorectal Polyps

Digestive Diseases ◽

10.1159/000366034 ◽

2014 ◽

Vol 33 (1) ◽

pp. 38-41 ◽

Cited By ~ 3

Author(s):

Rainer Schoefl ◽

Alexander Ziachehabi ◽

Friedrich Wewalka

Keyword(s):

White Light ◽

Colorectal Polyps ◽

High Grade Dysplasia ◽

High Grade ◽

Hyperplastic Polyps ◽

High Definition ◽

White Light Endoscopy ◽

Virtual Chromoendoscopy ◽

Experienced Endoscopists

Small (<10 mm) and diminutive (<6 mm) polyps harbour high-grade dysplasia or cancer in 0.3-5% of cases. The potential to grow and develop advanced histology is low. Traditional guidelines still recommend the removal of all polyps. Visual characterisation with modern endoscopic technology could enable us to leave diminutive hyperplastic polyps in situ and remove but discard small polyps. In expert hands, high-definition white-light endoscopy and virtual chromoendoscopy can reach an accuracy of more than 90% in distinguishing between hyperplastic and adenomatous pathology. For less experienced endoscopists the values are lower and therefore the concept is not yet fit for routine use. Polyps can be removed completely with snares but not with forceps. The cold snaring technique in particular has proved safe and effective for small polyps. With more experience in the future a ‘cut and discard' strategy for small polyps and a ‘do not resect' strategy for diminutive polyps will save money and time to deal with more advanced lesions.

Download Full-text

Squamous Metaplasia in Colonic Tubulovillous Adenoma: A Three Case Series and review of literature

American Journal of Clinical Pathology ◽

10.1093/ajcp/aqab191.137 ◽

2021 ◽

Vol 156 (Supplement_1) ◽

pp. S66-S66

Author(s):

D Duhoki ◽

D Allison ◽

E Wiley

Keyword(s):

Squamous Metaplasia ◽

Case Series ◽

Colonic Polyps ◽

Colorectal Polyps ◽

High Grade Dysplasia ◽

High Grade ◽

Tubulovillous Adenoma ◽

Polyp Size ◽

Eosinophilic Cytoplasm

Abstract Introduction/Objective Although squamous metaplasia of colonic tubulovillous adenoma is a rare phenomenon, colorectal polyps can show focal squamous metaplasia. While the exact cause in unknown, it has been suggested that mechanical irritation, torsion and chronic inflammation may predispose to squamous metaplasia within large colonic polyps. Methods/Case Report We observed this finding in two men and one woman with colorectal polyps who underwent endoscopic mucosal resection in a three-week timeframe. In two cases the polyps were rectal, while in the third case, it was located in the descending colon. Polyp size ranged from 1-5 cm and were semi-pedunculated. Histologically, all three lesions were tubulovillous adenomas with focal high-grade dysplasia and multiple foci of squamous metaplasia characterized by cells with benign-appearing nuclei, eosinophilic cytoplasm, and solid growth pattern. Some metaplastic foci approached the muscularis mucosae, mimicking adenocarcinoma at low power magnification. Interestingly, these metaplastic foci were reactive against p16 immunohistochemical stain. Results (if a Case Study enter NA) NA Conclusion We present these three cases to add to the few previously reported. Pathologists should be aware of this finding to avoid the overdiagnosis of invasive carcinoma in the setting of nested and pseudo invasive squamous morules, especially in polyps with high grade dysplasia. In addition, squamous metaplasia could be a precursor lesion for colorectal squamous cell carcinoma.

Download Full-text

The indirect role of site distribution in high-grade dysplasia in adenomatous colorectal polyps

Journal of Cancer Research and Therapeutics ◽

10.4103/0973-1482.19587 ◽

2005 ◽

Vol 1 (4) ◽

pp. 204 ◽

Cited By ~ 3

Author(s):

SA Ziaee ◽

N Rahbar ◽

S Molanie ◽

L Arefian ◽

SA Fanaie ◽

...

Keyword(s):

Colorectal Polyps ◽

High Grade Dysplasia ◽

High Grade ◽

Site Distribution

Download Full-text

Diagnostic performance of computed tomography colonography and colonoscopy: a prospective and validated analysis of 231 paired examinations

Acta Radiologica ◽

10.1080/02841850701422096 ◽

2007 ◽

Vol 48 (8) ◽

pp. 831-837 ◽

Cited By ~ 7

Author(s):

R. B. Arnesen ◽

E. von Benzon ◽

S. Adamsen ◽

L. B. Svendsen ◽

H. O. Raaschou ◽

...

Keyword(s):

Computed Tomography ◽

Diagnostic Performance ◽

Helical Ct ◽

Cost Effective ◽

Colorectal Polyps ◽

Low Grade ◽

Ct Scanner ◽

Computed Tomography Colonography ◽

First Choice ◽

Effective Use

Background: Detection of colorectal tumors with computed tomography colonography (CTC) is an alternative to conventional colonoscopy (CC), and clarification of the diagnostic performance is essential for cost-effective use of both technologies. Purpose: To evaluate the diagnostic performance of CTC compared with CC. Material and Methods: 231 consecutive CTCs were performed prior to same-day scheduled CC. The radiologist and endoscopists were blinded to each other's findings. Patients underwent a polyethylene glycol bowel preparation, and were scanned in prone and supine positions using a single-detector helical CT scanner and commercially available software for image analysis. Findings were validated (matched) in an unblinded comparison with video-recordings of the CCs and re-CCs in cases of doubt. Results: For patients with polyps ⩾5 mm and ⩾10 mm, the sensitivity was 69% (95% CI 58–80%) and 81% (68–94%), and the specificity was 91% (84–98%) and 98% (93–100%), respectively. For detection of polyps ⩾5 mm and ⩾10 mm, the sensitivity was 66% (57–75%) and 77% (65–89%). A flat, elevated low-grade carcinoma was missed by CTC. One cancer relapse was missed by CC, and a cecal cancer was missed by an incomplete CC and follow-up double-contrast barium enema. Conclusion: CC was superior to CTC and should remain first choice for the diagnosis of colorectal polyps. However, for diagnosis of lesions ⩾10 mm, CTC and CC should be considered as complementary methods.

Download Full-text

Machine Learning for Automated Polyp Detection in Computed Tomography Colonography

Advances in Bioinformatics and Biomedical Engineering - Biomedical Image Analysis and Machine Learning Technologies ◽

10.4018/978-1-60566-956-4.ch003 ◽

2010 ◽

pp. 54-77

Author(s):

Abhilash Alexander Miranda ◽

Olivier Caelen ◽

Gianluca Bontempi

Keyword(s):

Machine Learning ◽

Computed Tomography ◽

False Positive ◽

False Positive Rate ◽

Learning Algorithms ◽

Colorectal Polyps ◽

Machine Learning Algorithms ◽

Computed Tomography Colonography ◽

Positive Rate ◽

Independent Features

This chapter presents a comprehensive scheme for automated detection of colorectal polyps in computed tomography colonography (CTC) with particular emphasis on robust learning algorithms that differentiate polyps from non-polyp shapes. The authors’ automated CTC scheme introduces two orientation independent features which encode the shape characteristics that aid in classification of polyps and non-polyps with high accuracy, low false positive rate, and low computations making the scheme suitable for colorectal cancer screening initiatives. Experiments using state-of-the-art machine learning algorithms viz., lazy learning, support vector machines, and naïve Bayes classifiers reveal the robustness of the two features in detecting polyps at 100% sensitivity for polyps with diameter greater than 10 mm while attaining total low false positive rates, respectively, of 3.05, 3.47 and 0.71 per CTC dataset at specificities above 99% when tested on 58 CTC datasets. The results were validated using colonoscopy reports provided by expert radiologists.

Download Full-text