The sensitivity of computed tomographic virtual colonoscopy for detecting polyps varies widely in recently reported large clinical trials.
Dr Ronald Summers and his team determined whether a computer program is as sensitive as optical colonoscopy to detect adenomatous colonic polyps on computed thomographic virtual colonoscopy.
The data set was a cohort of 1186 screening patients at 3 medical centers.
All patients underwent same-day virtual and optical colonoscopy.
The research team's enhanced gold standard combined segmental unblinded optical colonoscopy and retrospective identification of precise polyp locations.
|Per-patient sensitivities for computer-aided polyp detection was 89%|
The data were randomized into 394 receiving separate training and 792 receiving test sets for analysis by a computer-aided polyp detection program.
The team found that per-polyp and per-patient sensitivities for computer-aided polyp detection were 89% for identifiable adenomatous polyps at least 1 cm in size.
The false-positive rate was 2 false polyps per patient.
Both carcinomas were detected by computer-aided polyp detection at a false-positive rate of 0.7 per patient.
The researchers noted that only 1 of 2 carcinomas was detected by optical colonoscopy before segmental unblinding.
At both 8-mm and 10-mm adenoma size thresholds, the per-patient sensitivities of computer-aided polyp detection did not differ from optical colonoscopy before segmental unblinding.
Dr Summers' team commented, “The per-patient sensitivity of computed tomograpic virtual colonoscopy computer-aided polyp detection in an asymptomatic screening population is comparable to that of optical colonoscopy for adenomas 8 mm or more.”
“It is generalizable to new computed tomographic virtual colonoscopy data”.