Clinical Evaluation of PolyDeep, A Computer-Aided Detection System: A Multicenter Randomized Tandem Colonoscopy Trial

Area of Science:

• Gastroenterology
• Medical Imaging
• Oncology

Background:

• Computer-aided detection (CADe) systems enhance lesion recognition during endoscopy.
• PolyDeep is a CADe/x tool previously evaluated in observational studies.
• The clinical utility of CADe systems in reducing adenoma miss rates requires further investigation.

Purpose of the Study:

• To determine if PolyDeep-assisted colonoscopy reduces the adenoma miss rate (AMR) compared to conventional colonoscopy.
• To evaluate the impact of PolyDeep on polyp miss rate (PMR), serrated lesion miss rate (SLMR), and advanced polyp miss rate (APMR).

Main Methods:

• A multicenter randomized controlled trial with a tandem colonoscopy design was conducted.
• Participants were recruited from a colorectal cancer screening program.
• Expert endoscopists performed colonoscopies, with patients randomized to either PolyDeep-assisted or conventional colonoscopy groups.

Main Results:

• A total of 240 patients were analyzed, with balanced baseline characteristics.
• No statistically significant differences were observed in AMR (21.3% vs. 18.1%, p=0.5) between the PolyDeep-assisted and conventional colonoscopy groups.
• Secondary endpoints, including PMR, SLMR, and APMR, also showed no significant differences between the groups.

Conclusions:

• PolyDeep-assisted colonoscopy did not demonstrate a reduction in the adenoma miss rate within this colorectal cancer screening program.
• The study did not find significant benefits of PolyDeep in detecting various types of polyps compared to standard colonoscopy.
• Further research may be needed to explore the role of CADe systems in different clinical settings or with improved algorithms.