Related Experiment Video
Updated: May 3, 2026
06:48
On-Site Sampling and Extraction of Brain Tumors for Metabolomics and Lipidomics Analysis
Published on: May 31, 2020
6.5K
COLLABORATIVE LABELING OF MALIGNANT GLIOMA
Zhoubing Xu1, Andrew J Asman1, Eesha Singh1
1Electrical Engineering, Vanderbilt University, Nashville, TN, 37235.
Proceedings. IEEE International Symposium on Biomedical Imaging
|January 25, 2014
Summary
This study introduces a novel internet-based collaborative labeling method to accurately quantify malignant glioma tumor regions on MRI scans. This efficient technique offers a cost-effective solution for medical image analysis.
Area of Science:
- Neuro-oncology
- Medical Imaging Analysis
- Computational Biology
Background:
- Malignant gliomas are aggressive brain tumors requiring precise surgical resection.
- Accurate quantification of tumor extent on MRI is crucial for treatment planning.
- Current manual labeling methods can be time-consuming and resource-intensive.
Purpose of the Study:
- To develop and validate an internet-based, collaborative labeling approach for quantifying malignant gliomas on MRI.
- To assess the reliability and efficiency of this novel labeling technique.
Main Methods:
- Utilized clinically acquired MRI scans of patients with malignant gliomas.
- Employed teams of minimally trained human raters for collaborative labeling of tumor regions.
- Focused on characterizing the gadolinium-enhancing core and edema tumor components.
- Quantified accuracy using the Dice similarity coefficient.
Main Results:
- Demonstrated reliable characterization of gadolinium-enhancing core and edema tumor regions with a Dice score of approximately 0.9.
- The collaborative approach proved highly parallel and efficient in terms of time and resources.
- Minimal training and no specialized hardware were required for participants.
Conclusions:
- Internet-based collaborative labeling is a promising, cost-effective technique for manual medical image annotation.
- This method facilitates efficient and reliable quantification of malignant glioma extent on MRI.
- Potential for wide applicability in medical imaging data analysis.
