Recommendation:
In adults undergoing colonoscopy, AGA makes no recommendation on the use of CADe-assisted colonoscopy.
Remarks:
- This recommendation underscores the overall very low certainty of evidence on long-term outcomes deemed critical for decision–making (ie, CRC incidence, CRC mortality, and PCCRC). The panel acknowledged the evidence for CADe in improving ADR as an important surrogate outcome.
- As an iterative tool, the panel highlights the likely potential for CADe as an AI application to continually improve with further training and validation over time. As this technology improves, and with additional studies evaluating long-term critical outcomes (ie, CRC incidence, CRC mortality, and PCCRC), an update of this recommendation is planned.
- Evidence gaps to be addressed include data from diverse patient populations and settings (e.g., community and academic endoscopists), studies evaluating impact on colon polyps of greater clinical significance (e.g., advanced adenomas), or more difficult to detect polyps (e.g., sessile serrated lesions), information on resource implications, access to colonoscopy, patient values and preferences, cost-effectiveness, and impact on long-term patient important outcomes.
