Segmentation of nodular medulloblastoma using Random Walker and Hierarchical Normalized Cuts

Author

Tchikindas, Lev ; Sparks, Rachel ; Baccon, Jennifer ; Ellison, David ; Judkins, Alexander R. ; Madabhushi, Anant

fYear

2011

fDate

1-3 April 2011

Firstpage

1

Lastpage

2

Abstract

Medulloblastoma (MB) is the most common brain tumor in children. Recent studies have demonstrated a relationship between specific signaling pathway abnormalities, a tendency to more favorable outcomes, and a histopathological feature: nodular growth patterns. In this work we present a new segmentation scheme which requires minimal user interaction to segment nodules on MB histopathological sections. Our segmentation scheme consists of two steps: (1) color reduction using Hierarchical Normalized Cuts (HNCut), (2) Random Walker (RW) segmentation within the reduced HNCut color space. Across a cohort of 18 nodular MB images, our integrated HNCut and RW scheme yielded nodule segmentations with a Dice coefficient of 83:55 ± 12:4% and Predictive Positive Value (PPV) of 93:71 ± 9:0%.

Keywords

biomedical optical imaging; brain; cancer; image colour analysis; image segmentation; medical image processing; paediatrics; random processes; tumours; Dice coefficient; Predictive Positive Value; brain tumor; children; color reduction; hierarchical normalized cuts; histopathological feature; nodular growth patterns; nodular medulloblastoma; random walker; reduced HNCut color space; segmentation; specific signaling pathway abnormalities; Cancer; Image color analysis; Image edge detection; Image segmentation; Pixel; Shape; Tumors;

fLanguage

English

Publisher

ieee

Conference_Titel

Bioengineering Conference (NEBEC), 2011 IEEE 37th Annual Northeast

Conference_Location

Troy, NY

ISSN

2160-7001

Print_ISBN

978-1-61284-827-3

Type

conf

DOI

10.1109/NEBC.2011.5778640

Filename

5778640