Abstract
Image labeling and parcellation are critical tasks for the assessment of volumetric and morphometric features in medical imaging data. The process of image labeling is inherently error prone as images are corrupted by noise and artifact. Even expert interpretations are subject to subjectivity and the precision of the individual raters. Hence, all labels must be considered imperfect with some degree of inherent variability. One may seek multiple independent assessments to both reduce this variability as well as quantify the degree of uncertainty. Existing techniques exploit maximum a posteriori statistics to combine data from multiple raters. A current limitation with these approaches is that they require each rater to generate a complete dataset, which is often impossible given both human foibles and the typical turnover rate of raters in a research or clinical environment. Herein, we propose a robust set of extensions that allow for missing data, account for repeated label sets, and utilize training/catch trial data. With these extensions, numerous raters can label small, overlapping portions of a large dataset, and rater heterogeneity can be robustly controlled while simultaneously estimating a single, reliable label set and characterizing uncertainty. The proposed approach enables parallel processing of labeling tasks and reduces the otherwise detrimental impact of rater unavailability.
| Original language | English (US) |
|---|---|
| Title of host publication | Progress in Biomedical Optics and Imaging - Proceedings of SPIE |
| Volume | 7623 |
| Edition | PART 1 |
| DOIs | |
| State | Published - 2010 |
| Event | Medical Imaging 2010: Image Processing - San Diego, CA, United States Duration: Feb 14 2010 → Feb 16 2010 |
Other
| Other | Medical Imaging 2010: Image Processing |
|---|---|
| Country | United States |
| City | San Diego, CA |
| Period | 2/14/10 → 2/16/10 |
Fingerprint
Keywords
- analysis
- data fusion
- delineation
- labeling
- Parcellation
- STAPLE
- statistics
ASJC Scopus subject areas
- Atomic and Molecular Physics, and Optics
- Electronic, Optical and Magnetic Materials
- Biomaterials
- Radiology Nuclear Medicine and imaging
Cite this
Simultaneous truth and performance level estimation with incomplete, over-complete, and ancillary data. / Landman, Bennett A.; Bogovic, John A.; Prince, Jerry Ladd.
Progress in Biomedical Optics and Imaging - Proceedings of SPIE. Vol. 7623 PART 1. ed. 2010. 76231N.Research output: Chapter in Book/Report/Conference proceeding › Conference contribution
}
TY - GEN
T1 - Simultaneous truth and performance level estimation with incomplete, over-complete, and ancillary data
AU - Landman, Bennett A.
AU - Bogovic, John A.
AU - Prince, Jerry Ladd
PY - 2010
Y1 - 2010
N2 - Image labeling and parcellation are critical tasks for the assessment of volumetric and morphometric features in medical imaging data. The process of image labeling is inherently error prone as images are corrupted by noise and artifact. Even expert interpretations are subject to subjectivity and the precision of the individual raters. Hence, all labels must be considered imperfect with some degree of inherent variability. One may seek multiple independent assessments to both reduce this variability as well as quantify the degree of uncertainty. Existing techniques exploit maximum a posteriori statistics to combine data from multiple raters. A current limitation with these approaches is that they require each rater to generate a complete dataset, which is often impossible given both human foibles and the typical turnover rate of raters in a research or clinical environment. Herein, we propose a robust set of extensions that allow for missing data, account for repeated label sets, and utilize training/catch trial data. With these extensions, numerous raters can label small, overlapping portions of a large dataset, and rater heterogeneity can be robustly controlled while simultaneously estimating a single, reliable label set and characterizing uncertainty. The proposed approach enables parallel processing of labeling tasks and reduces the otherwise detrimental impact of rater unavailability.
AB - Image labeling and parcellation are critical tasks for the assessment of volumetric and morphometric features in medical imaging data. The process of image labeling is inherently error prone as images are corrupted by noise and artifact. Even expert interpretations are subject to subjectivity and the precision of the individual raters. Hence, all labels must be considered imperfect with some degree of inherent variability. One may seek multiple independent assessments to both reduce this variability as well as quantify the degree of uncertainty. Existing techniques exploit maximum a posteriori statistics to combine data from multiple raters. A current limitation with these approaches is that they require each rater to generate a complete dataset, which is often impossible given both human foibles and the typical turnover rate of raters in a research or clinical environment. Herein, we propose a robust set of extensions that allow for missing data, account for repeated label sets, and utilize training/catch trial data. With these extensions, numerous raters can label small, overlapping portions of a large dataset, and rater heterogeneity can be robustly controlled while simultaneously estimating a single, reliable label set and characterizing uncertainty. The proposed approach enables parallel processing of labeling tasks and reduces the otherwise detrimental impact of rater unavailability.
KW - analysis
KW - data fusion
KW - delineation
KW - labeling
KW - Parcellation
KW - STAPLE
KW - statistics
UR - http://www.scopus.com/inward/record.url?scp=79751523494&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=79751523494&partnerID=8YFLogxK
U2 - 10.1117/12.844182
DO - 10.1117/12.844182
M3 - Conference contribution
C2 - 20694058
AN - SCOPUS:79751523494
SN - 9780819480248
VL - 7623
BT - Progress in Biomedical Optics and Imaging - Proceedings of SPIE
ER -