Benchmarking definitions of false-positive alerts during computer-aided polyp detection in colonoscopy
-
- Erik A. Holzwanger
- Division of Gastroenterology and Hepatology, Tufts Medical Center, Boston, Massachusetts, United States
-
- Mohammad Bilal
- Center for Advanced Endoscopy, Division of Gastroenterology, Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, Massachusetts, United States
-
- Jeremy R. Glissen Brown
- Center for Advanced Endoscopy, Division of Gastroenterology, Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, Massachusetts, United States
-
- Shailendra Singh
- West Virginia University Health Sciences Center Charleston Division, Charleston, West Virginia, United States
-
- Aymeric Becq
- Sorbonne Université, Centre d’Endoscopie Digestive, Hôpital Saint Antoine, APHP, Paris, France
-
- Kenneth Ernest-Suarez
- Gastroenterology Department, Hospital México, University of Costa Rica, San Jose, Costa Rica
-
- Tyler M. Berzin
- Center for Advanced Endoscopy, Division of Gastroenterology, Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, Massachusetts, United States
説明
<jats:title>Abstract</jats:title><jats:p> Background The occurrence of false-positive alerts is an important outcome measure in computer-aided colon polyp detection (CADe) studies. However, there is no consensus definition of a false positive in clinical trials evaluating CADe in colonoscopy. We aimed to study the diagnostic performance of CADe based on different threshold definitions for false-positive alerts. </jats:p><jats:p> Methods A previously validated CADe system was applied to screening/surveillance colonoscopy videos. Different thresholds for false-positive alerts were defined based on the time an alert box was continuously traced by the system. Primary outcomes were false-positive results and specificity using different threshold definitions of false positive.</jats:p><jats:p> Results 62 colonoscopies were analyzed. CADe specificity and accuracy were 93.2 % and 97.8 %, respectively, for a threshold definition of ≥ 0.5 seconds, 98.6 % and 99.5 % for a threshold definition of ≥ 1 second, and 99.8 % and 99.9 % for a threshold definition of ≥ 2 seconds. </jats:p><jats:p> Conclusion Our analysis demonstrated how different threshold definitions of false positive can impact the reported diagnostic performance of CADe for colon polyp detection.</jats:p>
収録刊行物
-
- Endoscopy
-
Endoscopy 53 (09), 937-940, 2020-11-02
Georg Thieme Verlag KG