A study, published in the Annals of Internal Medicine, examined how computer-aided diagnosis is improving rectosigmoid adenoma detection during colonoscopies, Medpage Today reports.
Researchers examined 791 patients who had a colonoscopy at Yokohama, Japan-based Showa University Northern Yokohama Hospital. Twenty-three endoscopists performed colonoscopies from June 2017 to September 2017. Researchers were examining whether CAD-stained analysis would produce a greater than 90 percent negative predictive value for diagnosing diminutive rectosigmoid adenomas.
Here's what you should know:
1. About 461 patients had 838 diminutive polyps of <5 mm. In total, 466 diminutive polyps from 325 patients were assessed by CAD.
2. Researchers found the pathologic prediction rate of CAD was 98.1 percent. The negative predictive value ranged from 96.5 percent to a worst-case scenario of 93.7 percent for the stained mode.
3. For narrow band imaging, the best- and worst-case scenarios were 96.5 percent and 95.2 percent, respectively.
"We found CAD designed for endocytoscopy offers performance levels that meet the clinical threshold for using the diagnose-and-leave strategy for diminutive, non-neoplastic rectosigmoid polyps," researchers said.