Category: Colorectal Cancer Prevention
Shahab R. Khan
Rush University Medical Center
Adenoma detection rate (ADR) is a quality indicator of screening colonoscopy. Artificial intelligence (AI) using Convolutional neural network (CNN), is a type of machine learning algorithm that uses convolutions of the input image to extract relevant information and classify it into different entities. In this analysis, we aim to quantitatively appraise the reported data on ADR during colonoscopy in the presence of CNN based computer-aided detection (CADe) from prospectively conducted parallel RCTs in real-life scenarios compared to standard colonoscopy (SC).
Methods: Multiple databases were searched(from inception to May 2020),and parallel RCTs that compared deep CNN based CADe assisted colonoscopy to SC were included for this analysis. Using a random-effects model, pooled risk ratios (RR) and mean difference (MD) were calculated.Heterogeneity was assessed by I2 % values.
6 prospective studies we analyzed using a CNN based machine learning algorithm with the capability of detecting lesions in real-time. The total number of patients analyzed was 4962,with 2480 in CADe arm and 2482 in the SC group.Baseline age range(50-52 vs 51), male gender(50% vs 51%) and screening/surveillance indication(13% vs 14%) were comparable between the CADe and SC arms.
The pooled ADR with the use of CADe endoscopy was significantly greater when compared to SC(RR=1.5, 95% CI 1.3-1.72, p=0.0001, I2 =56%).The pooled proportion of ADR with CADe was 32.8%(95% CI 24.2- 42.7)and the pooled proportion of ADR with SC was 21.1%(95% CI 14.5-29.7).Additionally, the pooled RR of polyp detection rate was significantly greater with CADe when compared to SC(1.42, 95% CI 1.33-1.51, p=0.0001, I2 =9%)and the MD of scope withdrawal time was statistically lesser with CADe(0.38 minutes, 95% CI 0.05-0.72, p=0.02, I2 =97%).
The pooled RR of advanced ADR(1, 95% CI 0.74-1.36, p=0.93) and sessile serrated ADR(1.29, 95% CI 0.89-1.89, p=0.18)were comparable between CADe and SC;however the mean adenoma detected per colonoscopy was significantly better with CADe colonoscopy(MD=0.19, 95% CI 1.16-0.21, p=0.001;The pooled proportion of false positives on CADe colonoscopy was 10.3%(95% CI 6.1-16.8),with comparable cecal intubation time(MD=0.04, 95% CI 0.29-0.38, p=0.8) between CADe and SC.
CNN based CADe system significantly increases ADR during real-time colonoscopy,with faster withdrawal time and no increase in cecal insertion time.Future studies are warranted to study the impact of AI exclusively in screening colonoscopy.
Forest plot: Figure 1 - Overall ADR Outcomes
Forest plot: Figure 2 - Withdrawal Time
Antonio Facciorusso indicated no relevant financial relationships.
Babu Mohan indicated no relevant financial relationships.
Saurabh Chandan indicated no relevant financial relationships.
Shahab Khan indicated no relevant financial relationships.
Lena Kassab indicated no relevant financial relationships.
Paraskevas Gkolfakis indicated no relevant financial relationships.
Georgios Tziatzios indicated no relevant financial relationships.
Konstantinos Triantafyllou indicated no relevant financial relationships.
Douglas Adler: Boston Scientific, USA – Consultant.