P2221 - Comparing Polyp Detection Rate Between Water Exchange and Air Insufflation by a Computer-aided Detection Algorithm: An Analysis of Withdrawal Phase Videos from a Randomized Controlled Trial

Tuesday, October 26, 2021

10:30 AM – 4:00 PM PT

Location: Shoreline Exhibit Hall

Has Audio

Presenting Author(s)

CT

Chia-Pei Tang, MD

Buddhist Dalin Tzu Chi General Hospital
Chiayi, Chiayi, Taiwan (Republic of China)

Chia-Pei Tang, MD¹, Yu-Hsi Hsieh, MD², Tu-Liang Lin, PhD³, Chih-Wei Tseng, MD⁴, Felix Leung, MD⁵
¹Buddhist Dalin Tzu Chi General Hospital, Chiayi, Chiayi, Taiwan; ²Dalin Tzu Chi Hospital, Tzu Chi Medical Foundation, Chiayi, Chiayi, Taiwan; ³National Chiayi University, Chiayi, Chiayi, Taiwan; ⁴Dalin Tzu Chi Hospital, Tzu Chi Medical Foundation, Chiayi, Chiayi, Taiwan; ⁵Sepulveda Ambulatory Care Center, VAGLAHS, North Hills, CA

Introduction: Water exchange (WE) improves polyp (PDR) and adenoma detection rate (ADR) compared with air insufflation [GIE 2021;93:1411], but still misses polyps due to human limitations [JCG 2021, in press]. A computer-aided detection (CADe) algorithm can overcome human omissions, but is limited by false positives (FPs) [Lancet Gastroenterol & Hepatol 2019;4:71], which might be reduced by salvage-cleaning effect provided by insertion WE [Tzu-Chi Med J 2021;33:108]. We hypothesized that the strengths of CADe and WE can complement the weaknesses of each other. Their combination can enhance PDR and ADR and reduce false positive rate (FPR).

Methods: We used a locally developed convolutional neural network with YOLOv4 to build a CADe model. Images of polyps (n=15961) were divided in 80% and 20% for training and testing. Prior to real-time application, we validated the model in detecting polyps, using edited and coded withdrawal phase videos of the right colon from a RCT comparing right colon adenoma detection between WE and air insufflation. The polyps detected by the CADe were confirmed by the consensus of 2 blinded reviewers (experienced endoscopists). We also developed a computer-aided diagnosis (CADx) model based on YOLOv4 to characterize the detected polyps as adenomas or not. The primary and secondary outcomes were PDR and the FPR, respectively.

Results: A total of 245 coded videos of colonoscopies inserted with WE (n=123) or air insufflation (n=122) were analyzed. CADe achieved significantly higher PDR in the WE group (61.8%) than in the air insufflation group (40.2%), P=0.0009. CADe also had significantly higher additional PDR in the WE group (30.1% vs. 12.3%, P=0.001). The CADx assessed ADR was significantly higher in the WE group than that in the air insufflation group (46.3% vs. 30.3%, P=0.018), after the detected polyps were characterized by the CADx. (Table 1a) The mean numbers of FPs related to feces [1.78 (1.67) vs. 2.09 (2.09), P=0.007] and bubbles [0.53 (0.89) vs. 1.25 (2.45), P=0.001] were significantly lower in the WE group than in the air insufflation group. (Table 1b)

Discussion: The data confirmed the hypothesis that the strengths of CADe and WE complemented the weaknesses of each other, enhanced PDR and ADR and reduced FPR. The locally developed CADe and CADx are validated for polyp detection and adenoma assignment for real-time application. Furthermore, WE provides an optimal platform for the computer-assisted algorithms.

Table 1a: Polyp detection characteristics
Variables		WE N=123		Air insufflation N=122		*P value*
Per patient analysis
PDR by endoscopist, n (%)		59 (47.9)		41 (33.6)		0.027⁺⁺
PDR by CADe, n (%)		76 (61.8)		49 (40.2)		0.0009⁺⁺
Additional polyp detection rate by CADe, n (%)		37 (30.1)		15 (12.3)		0.001⁺⁺
ADR by endoscopist, n (%)		43 (34.9)		33 (27.0)		0.214⁺⁺
ADR by CADe, n (%)		57 (46.3)		37 (30.3)		0.018⁺⁺
Additional adenoma detect rate by CADe, n (%)		14 (11.4)		5 (4.1)		0.054⁺⁺
Per polyp analysis
Number of polyps found by endoscopist, mean (SD)		0.87 (1.10)		0.59 (1.21)		0.022⁺
Number of polyps found by CADe, mean (SD)		0.80 (1.07)		0.57 (1.21)		0.015⁺
Number of Additional polyps found by CADe, mean (SD)		0.41 (0.70)		0.12 (0.33)		< 0.001⁺
Total number of polyps, mean (SD)		1.20 (1.43)		0.70 (1.30)		0.002⁺
Proportion of additional polyps found by CADe, n/N (%)		50/149 (33.3)		15/86 (17.4)		0.010⁺⁺
Overall CADe polyp detection sensitivity		169/170 (99.4%)
Overall CADe adenoma detection sensitivity		122/126 (96.8%)
Table 1b: False positives characteristics
Causes of false positives	WE		Air		P value
	n=123		n=122
Feces and Debris, mean (SD)	1.78 (1.67)		2.09 (2.09)		0.007†
Bubbles, mean (SD)	0.53 (0.89)		1.25 (2.45)		0.001†
Winkled Folds, mean (SD)	3.14 (2.10)		2.94 (2.23)		0.349†
Others, mean (SD)	0.84 (1.08)		0.70 (0.99)		0.246†
False positives appearance time, sec
≤ 0.5, mean (SD)	5.98 (3.12)		7.04 (4.62)		0.187†
0.6-1.0, mean (SD)	0.15 (0.39)		0.20 (0.44)		0.368†
1.1-2.0, mean (SD)	0.02 (0.13)		0.03 (0.22)		0.640†
>2.0, mean (SD)	0.13 (0.38)		0.08 (0.30)		0.281†
Total, mean (SD)	6.28 (3.11)		7.36 (4.76)		0.205†

Table: ADR, adenoma detection rate; All the percentages calculated according to respective total number of patients for each group; PDR, polyp detection rate; Total numbers of polyps were based on those found by the performing endoscopist plus those detected by the CADe; Proportion of additional polyps found by CADe equals numbers of additional polyp (adenomas) found by CADe divided by total numbers of polyps (adenomas).
Others: ileocecal valve, cecum, polypectomy site, suction marks and drug pills. † Mann-Whitney test; ††Fisher exact test.

Disclosures:
Chia-Pei Tang indicated no relevant financial relationships.
Yu-Hsi Hsieh indicated no relevant financial relationships.
Tu-Liang Lin indicated no relevant financial relationships.
Chih-Wei Tseng indicated no relevant financial relationships.
Felix Leung indicated no relevant financial relationships.

Chia-Pei Tang, MD¹, Yu-Hsi Hsieh, MD², Tu-Liang Lin, PhD³, Chih-Wei Tseng, MD⁴, Felix Leung, MD⁵. P2221 - Comparing Polyp Detection Rate Between Water Exchange and Air Insufflation by a Computer-aided Detection Algorithm: An Analysis of Withdrawal Phase Videos from a Randomized Controlled Trial, ACG 2021 Annual Scientific Meeting Abstracts. Las Vegas, Nevada: American College of Gastroenterology.