Abstract Details
Activity Number:
|
202
|
Type:
|
Contributed
|
Date/Time:
|
Monday, August 10, 2015 : 10:30 AM to 11:15 AM
|
Sponsor:
|
Section on Statistics in Imaging
|
Abstract #317861
|
|
Title:
|
Automated Intracerebral Hemorrhage Segmentation of CT Scans
|
Author(s):
|
John Muschelli* and Elizabeth Sweeney and Natalie L. Ullman and Daniel F. Hanley and Ciprian Crainiceanu
|
Companies:
|
The Johns Hopkins University and The Johns Hopkins University and Johns Hopkins Medical Institution and Johns Hopkins Medical Institution and The Johns Hopkins University
|
Keywords:
|
neuroimaging ;
CT ;
logistic regression ;
imaging ;
segmentation
|
Abstract:
|
Intracerebral hemorrhage (ICH) is a neurological condition that results from a blood vessel rupturing into the brain. Manual delineation of ICH is the gold standard but is time-consuming and has variability. We estimated the probability of ICH at a voxel-level using logistic regression with features extracted from X-ray computed tomography (CT) scans. We used 112 patients from MISTIE stroke trial, one scan per patient, for model estimation and validation. ICH was manually segmented by expert readers. Binary hemorrhage masks were created: voxels = 1 if the voxel was classified as hemorrhage, 0 otherwise. We derived a set of imaging predictors and used the first 10 scans to fit the model: logit(Y(v)) = ? + ?X(v), where X(v) is a set of features for voxel v and Y(v) is the binary presence of ICH. On 51 validation scans the model classified voxels as ICH or not and we used the Dice Similarity Index (DSI) to estimate model performance which ranges from 0 to 1 (highest). The mean (SD) validation scan DSI was 0.861 (0.05). These results indicate that the approach described can achieve accurate segmentation of ICH in a population of patients from a variety of imaging centers.
|
Authors who are presenting talks have a * after their name.
Back to the full JSM 2015 program
|
For program information, contact the JSM Registration Department or phone (888) 231-3473.
For Professional Development information, contact the Education Department.
The views expressed here are those of the individual authors and not necessarily those of the JSM sponsors, their officers, or their staff.
2015 JSM Online Program Home
ASA Meetings Department
732 North Washington Street, Alexandria, VA 22314
(703) 684-1221 • meetings@amstat.org
Copyright © American Statistical Association.