TY  - JOUR
AU  - Gloria Fernandez Esparrach
AU  - Jorge Bernal
AU  - Maria Lopez Ceron
AU  - Henry Cordova
AU  - Cristina Sanchez Montes
AU  - Cristina Rodriguez de Miguel
AU  - F. Javier Sanchez
PY  - 2016//
TI  - Exploring the clinical potential of an automatic colonic polyp detection method based on the creation of energy maps
T2  - END
JO  - Endoscopy
SP  - 837
EP  - 842
VL  - 48
IS  - 9
N2  - Background and aims: Polyp miss-rate is a drawback of colonoscopy that increases significantly in small polyps. We explored the efficacy of an automatic computer vision method for polyp detection.Methods: Our method relies on a model that defines polyp boundaries as valleys of image intensity. Valley information is integrated into energy maps which represent the likelihood of polyp presence.Results: In 24 videos containing polyps from routine colonoscopies, all polyps were detected in at least one frame. Mean values of the maximum of energy map were higher in frames with polyps than without (p<0.001). Performance improved in high quality frames (AUC= 0.79, 95%CI: 0.70-0.87 vs 0.75, 95%CI: 0.66-0.83). Using 3.75 as maximum threshold value, sensitivity and specificity for detection of polyps were 70.4% (95%CI: 60.3-80.8) and 72.4% (95%CI: 61.6-84.6), respectively.Conclusion: Energy maps showed a good performance for colonic polyp detection. This indicates a potential applicability in clinical practice.
L1  - http://refbase.cvc.uab.es/files/FBL2016.pdf
UR  - http://dx.doi.org/10.1055/s-0042-108434
N1  - MV;
ID  - Gloria Fernandez Esparrach2016
ER  -