Colorectal Cancer Prevention

41 - Can Artificial Intelligence (AI) Achieve Real-Time 'Resect and Discard' Thresholds Independently of Device or Operator?

Tuesday, October 9
2:55 PM - 3:05 PM
Location: Terrace Ballroom 4 (level 400)

Category: Colorectal Cancer Prevention
Robin Zachariah, MD1, Andrew Ninh2, Tyler Dao3, James Requa2, William Karnes, MD1
1University of California Irvine Medical Center, Orange, CA; 2Docbot, Irvine, CA; 3Docbot, Orange, CA

Introduction: A scope- and operator-independent method to optically diagnose diminutive polyps could make "Resect and Discard" a reality and save over $1 billion/year in the US alone. The American Society for Gastrointestinal Endoscopy’s (ASGE) Preservation and Incorporation of Valuable endoscopic Innovations (PIVI) "Resect and Discard" guideline requires a > 90% negative predictive value (NPV) for diminutive adenomas and > 90% concordance in recommended surveillance intervals comparing optical pathology to histology. Convolutional neural networks (CNN) have potential to predict polyp pathology and meet PIVI guidelines independently of operator or scope manufacturer.

Methods: We developed an optical pathology (OP) model using a CNN built on Tensorflow and pre-trained on ImageNet. 5456 high quality images of unique adenomas and serrated polyps of known locations, size, and light source (white light [WL] or narrow band imaging [NBI]) were extracted from our endoscopic database.  Images were partitioned into 5 equal-sized subsamples for 5-fold cross validation with training (80%) and validation (20%). An Adam optimizer generates a probability between 0-0.5 (serrated) and 0.5-1 (adenoma). Surveillance intervals were calculated based on US Multi-Society Task Force guidelines, comparing OP vs. true pathology (TP).

Results: Among polyps throughout the colon, NPV for adenomas was 92% (WL) and 93% (NBI). Surveillance interval concordance between OP and TP for screening and surveillance cases was 93% and 96%, respectively. Among diminutive polyps (< 6 mm) throughout the colon, NPV for adenomas was 91% (WL) and 92% (NBI). Surveillance concordance was 93% and 96% for screening and surveillance cases, respectively. Among diminutive polyps in the left colon, NPV improved to 97% (WL) and 95% (NBI). The model processes > 90 frames per second and provides real-time feedback during colonoscopy using a conventional desktop and graphics processing unit.

Discussion: Without stringencies (no unclassifiable polyps), our optical pathology model meets "Resect and Discard" PIVI guidelines and provides operator-independent and real-time feedback during colonoscopy. Accuracy is unaffected by light source, suggesting it may work well with any scope manufacturer. Adenoma detection rate and surveillance recommendations at the time of colonoscopy are added benefits. Blinded multicenter studies utilizing multiple scope manufacturers are needed to validate these potentials.

Figure 1: Distribution of Polyp Pathology by AI Predictions (inclusive of all polyp sizes, locations and light sources). 90+% of cases utilized 180- and 190-series Olympus colonoscopes.
Table 1. Interval calculations are based on indication (family and/or personal history), size, number, location and polyp pathology (OP vs TP) according to US Multi-Society Task Force guidelines.

Disclosures:
Robin Zachariah indicated no relevant financial relationships.
Andrew Ninh: Docbot – Employee, Stockholder/Ownership Interest (excluding diversified mutual funds).
Tyler Dao: Docbot – Employee, Stockholder/Ownership Interest (excluding diversified mutual funds).
James Requa: Docbot – Consultant.
William Karnes: Docbot – Stockholder/Ownership Interest (excluding diversified mutual funds).

Robin Zachariah

Gastroenterology Fellow
University of California - Irvine Medical Center
Orange, California

Presentation(s):

Send Email for Robin Zachariah


Assets

41 - Can Artificial Intelligence (AI) Achieve Real-Time 'Resect and Discard' Thresholds Independently of Device or Operator?



Attendees who have favorited this

Please enter your access key

The asset you are trying to access is locked. Please enter your access key to unlock.

Send Email for Can Artificial Intelligence (AI) Achieve Real-Time 'Resect and Discard' Thresholds Independently of Device or Operator?