Using uncertainty estimation to reduce false positives in liver lesion detection

Ishaan Bhat, Hugo J. Kuijf, Veronika Cheplygina, Josien P.W. Pluim

Onderzoeksoutput: Hoofdstuk in Boek/Rapport/CongresprocedureConferentiebijdrageAcademicpeer review

6 Citaten (Scopus)

Samenvatting

Despite the successes of deep learning techniques at detecting objects in medical images, false positive detections occur which may hinder an accurate diagnosis. We propose a technique to reduce false positive detections made by a neural network using a SVM classifier trained with features derived from the uncertainty map of the neural network prediction. We demonstrate the effectiveness of this method for the detection of liver lesions on a dataset of abdominal MR images. We find that the use of a dropout rate of 0.5 produces the least number of false positives in the neural network predictions and the trained classifier filters out approximately 90% of these false positives detections in the test-set.

Originele taal-2Engels
Titel2021 IEEE 18th International Symposium on Biomedical Imaging, ISBI 2021
UitgeverijIEEE Computer Society
Pagina's663-667
Aantal pagina's5
ISBN van elektronische versie9781665412469
DOI's
StatusGepubliceerd - 13 apr. 2021
Evenement18th IEEE International Symposium on Biomedical Imaging, ISBI 2021 - Nice, Frankrijk
Duur: 13 apr. 202116 apr. 2021

Congres

Congres18th IEEE International Symposium on Biomedical Imaging, ISBI 2021
Land/RegioFrankrijk
StadNice
Periode13/04/2116/04/21

Bibliografische nota

Publisher Copyright:
© 2021 IEEE.

Vingerafdruk

Duik in de onderzoeksthema's van 'Using uncertainty estimation to reduce false positives in liver lesion detection'. Samen vormen ze een unieke vingerafdruk.

Citeer dit