TY - JOUR
T1 - Evaluating White Matter Lesion Segmentations with Refined Sørensen-Dice Analysis
AU - Carass, Aaron
AU - Roy, Snehashis
AU - Gherman, Adrian
AU - Reinhold, Jacob C.
AU - Jesson, Andrew
AU - Arbel, Tal
AU - Maier, Oskar
AU - Handels, Heinz
AU - Ghafoorian, Mohsen
AU - Platel, Bram
AU - Birenbaum, Ariel
AU - Greenspan, Hayit
AU - Pham, Dzung L.
AU - Crainiceanu, Ciprian M.
AU - Calabresi, Peter A.
AU - Prince, Jerry L.
AU - Roncal, William R.Gray
AU - Shinohara, Russell T.
AU - Oguz, Ipek
N1 - Funding Information:
This work was supported in part by the NIH, through NINDS grants R01-NS094456 (PI: I. Oguz), R01-NS085211 (PI: R.T. Shinohara), R21-NS093349 (Co-PI: R.T. Shinohara), R01-NS082347 (PI: P.A. Calabresi), and R01-NS070906 (PI: D.L. Pham), NIMH grant R24-MH114799 (PI: W.R. Gray Roncal), and NIBIB grant R01-EB017255 (PI: P.A. Yushkevich, Dept. of Radiology, Univ. of Pennsylvania). As well as National MS Society grants RG-1507-05243 (PI: D.L. Pham) and RG-1707-28586 (PI: R.T. Shinohara).
Publisher Copyright:
© 2020, The Author(s).
Copyright:
Copyright 2020 Elsevier B.V., All rights reserved.
PY - 2020/12/1
Y1 - 2020/12/1
N2 - The Sørensen-Dice index (SDI) is a widely used measure for evaluating medical image segmentation algorithms. It offers a standardized measure of segmentation accuracy which has proven useful. However, it offers diminishing insight when the number of objects is unknown, such as in white matter lesion segmentation of multiple sclerosis (MS) patients. We present a refinement for finer grained parsing of SDI results in situations where the number of objects is unknown. We explore these ideas with two case studies showing what can be learned from our two presented studies. Our first study explores an inter-rater comparison, showing that smaller lesions cannot be reliably identified. In our second case study, we demonstrate fusing multiple MS lesion segmentation algorithms based on the insights into the algorithms provided by our analysis to generate a segmentation that exhibits improved performance. This work demonstrates the wealth of information that can be learned from refined analysis of medical image segmentations.
AB - The Sørensen-Dice index (SDI) is a widely used measure for evaluating medical image segmentation algorithms. It offers a standardized measure of segmentation accuracy which has proven useful. However, it offers diminishing insight when the number of objects is unknown, such as in white matter lesion segmentation of multiple sclerosis (MS) patients. We present a refinement for finer grained parsing of SDI results in situations where the number of objects is unknown. We explore these ideas with two case studies showing what can be learned from our two presented studies. Our first study explores an inter-rater comparison, showing that smaller lesions cannot be reliably identified. In our second case study, we demonstrate fusing multiple MS lesion segmentation algorithms based on the insights into the algorithms provided by our analysis to generate a segmentation that exhibits improved performance. This work demonstrates the wealth of information that can be learned from refined analysis of medical image segmentations.
UR - http://www.scopus.com/inward/record.url?scp=85084961810&partnerID=8YFLogxK
U2 - 10.1038/s41598-020-64803-w
DO - 10.1038/s41598-020-64803-w
M3 - Journal articles
C2 - 32427874
AN - SCOPUS:85084961810
SN - 2045-2322
VL - 10
JO - Scientific Reports
JF - Scientific Reports
IS - 1
M1 - 8242
ER -
