Differentiating Small Polyp Histologies Using Real-Time Screening Colonoscopy With Fuji Intelligent Color Enhancement

419 Background: Early detection of colorectal cancer (CRC) has been shown to decrease mortality, although compliance to CRC screening is low. Availability of a blood-based test is expected to improve CRC screening compliance. Specific detection of CRC using the Septin9 biomarker (mSEPT9) in a large prospective trial of an average-risk CRC screening population exhibited 67% sensitivity for CRC with 88% specificity. Laboratory-developed tests detecting mSEPT9 in plasma are now available in North America and a 2nd generation molecular diagnostic blood test for mSEPT9 is available as a CE-marked kit in Europe. The current research evaluated the clinical performance of the 2nd generation mSEPT9 assay. Methods: Bisulfite-converted DNA (bisDNA) was prepared from 3.5 mL human plasma using the 2ndgeneration plasma DNA preparation kit. Resulting bisDNA was analyzed in triplicate on the ABI7500 Fast Dx (Life Technologies, Inc.) using proprietary HeavyMethyl real-time PCR technology for mSEPT9 and the 2nd generation real-time PCR kit. In a case – control design, plasma from 98 CRC patients (n = 87 stages I - III) and 99 age-matched, colonoscopy-verified normal individuals were processed with the mSEPT9 assay. In addition, plasma from 150 prospectively enrolled average risk individuals scheduled for screening colonoscopy was tested. mSEPT9 was qualitatively analyzed such that any detection of mSEPT9 in a PCR was called “positive”. Results: The revised mSEPT9 assay exhibited 95% sensitivity (95% CI: 89-98%) for CRC. Sensitivity for stage I was 89% (95% CI: 72-96%, n = 27) and sensitivity for stage II was 93% (95% CI: 78-98%, n = 29). The control group was positive at a rate of 16% (95% CI: 10-25%). Specificity of the mSEPT9 assay in the screening cohort was 85% (95% CI: 78-89%). Conclusions: The 2nd generation mSEPT9 assay demonstrated improved sensitivity for CRC without significant impact on specificity. The enhanced design and robustness of the assay will further facilitate its standardized use in routine laboratory settings. Finally, the increased sensitivity of the revised mSEPT9 assay improves the detection of early stage disease and demonstrates the feasibility of a blood-based CRC screening technology.

Download Full-text

AI based colorectal disease detection using real-time screening colonoscopy

Precision Clinical Medicine ◽

10.1093/pcmedi/pbab013 ◽

2021 ◽

Author(s):

Jiawei Jiang ◽

Qianrong Xie ◽

Zhuo Cheng ◽

Jianqiang Cai ◽

Tian Xia ◽

...

Keyword(s):

Real Time ◽

Diagnostic Performance ◽

Characteristic Curve ◽

External Validation ◽

High Accuracy ◽

Colorectal Disease ◽

Screening Colonoscopy ◽

Validation Dataset ◽

Early Screening ◽

Colorectal Diseases

Abstract Colonoscopy is an effective tool for early screening of colorectal diseases. However, the application of colonoscopy in distinguishing different intestinal diseases still faces great challenges of efficiency and accuracy. Here we constructed and evaluated a deep convolution neural network (CNN) model based on 117,055 images from 16,004 individuals, which achieved a high accuracy of 0.933 in the validation dataset in identifying patients with polyp, colitis, colorectal cancer (CRC) from normal. The proposed approach was further validated on multi-center real-time colonoscopy videos and images, which achieved accurate diagnostic performance on detecting colorectal diseases with high accuracy and precision to generalize across external validation datasets. The diagnostic performance of the model was further compared to the skilled endoscopists and the novices. In addition, our model has potential in diagnosis of adenomatous polyp and hyperplastic polyp with an area under the receiver operating characteristic curve of 0.975. Our proposed CNN models have potential in assisting clinicians in making clinical decisions with efficiency during application.

Download Full-text