TY - GEN

T1 - Lesion-wise evaluation for effective performance monitoring of small object segmentation

AU - Groothuis, Irme

AU - Sudre, Carole H.

AU - Ingala, Silvia

AU - Barnes, Jo

AU - Gispert, Juan Domingo

AU - Sørensen, Lauge

AU - Pai, Akshay

AU - Nielsen, Mads

AU - Ourselin, Sebastien

AU - Cardoso, M. Jorge

AU - Barkhof, Frederik

AU - Modat, Marc

N1 - Publisher Copyright: © 2021 SPIE.

PY - 2021

Y1 - 2021

N2 - Object detection in medical images using deep learning is a challenging task, due to the imbalance often present in the data. Deep learning algorithms require large amount of balanced data to achieve optimal performance, as well as close monitoring and ne-tuning of hyper parameters. For most applications, such performance monitoring is done by simply feeding unseen data trough the network, and then using the loss function for evaluation. In the case of small or sparse objects, the loss function might not able to describe the features needed, but such features can be hard to capture in a loss function. In this paper we introduce a lesion-wise whole volume validation tool, which allows more a more accurate performance monitoring of segmentation of small and sparse objects. We showcase the efficacy of our tool by applying it to the task of microbleed segmentation, and compare the behaviour of lesionwise-whole volume validation compared to well known segmentation loss functions. Microbleeds are visible as small (less than 10 mm), ovoid hypo-intensities on T2∗-weighted and susceptibility weighted magnetic resonance images. Detection of microbleeds is clinically relevant, as microbleeds can indicate the risk of recurrent stroke, and are used as imaging biomarker for various neurodegenerative diseases. Manual detection or segmentation is time consuming and error prone, and suffers from high inter- and intraobserver variability. Due to the sparsity and small size of the lesions, the data is severely imbalanced.

AB - Object detection in medical images using deep learning is a challenging task, due to the imbalance often present in the data. Deep learning algorithms require large amount of balanced data to achieve optimal performance, as well as close monitoring and ne-tuning of hyper parameters. For most applications, such performance monitoring is done by simply feeding unseen data trough the network, and then using the loss function for evaluation. In the case of small or sparse objects, the loss function might not able to describe the features needed, but such features can be hard to capture in a loss function. In this paper we introduce a lesion-wise whole volume validation tool, which allows more a more accurate performance monitoring of segmentation of small and sparse objects. We showcase the efficacy of our tool by applying it to the task of microbleed segmentation, and compare the behaviour of lesionwise-whole volume validation compared to well known segmentation loss functions. Microbleeds are visible as small (less than 10 mm), ovoid hypo-intensities on T2∗-weighted and susceptibility weighted magnetic resonance images. Detection of microbleeds is clinically relevant, as microbleeds can indicate the risk of recurrent stroke, and are used as imaging biomarker for various neurodegenerative diseases. Manual detection or segmentation is time consuming and error prone, and suffers from high inter- and intraobserver variability. Due to the sparsity and small size of the lesions, the data is severely imbalanced.

KW - Data imbalance

KW - Evaluation

KW - Microbleed

UR - http://www.scopus.com/inward/record.url?scp=85103674563&partnerID=8YFLogxK

U2 - 10.1117/12.2580734

DO - 10.1117/12.2580734

M3 - Article in proceedings

AN - SCOPUS:85103674563

T3 - Progress in Biomedical Optics and Imaging - Proceedings of SPIE

SP - 1

EP - 8

BT - Medical Imaging 2021

A2 - Isgum, Ivana

A2 - Landman, Bennett A.

PB - SPIE

T2 - Medical Imaging 2021: Image Processing

Y2 - 15 February 2021 through 19 February 2021

ER -