Related Experiment Video
Updated: Jul 19, 2025
Brain Infarct Segmentation and Registration on MRI or CT for Lesion-symptom Mapping
Published on: September 25, 2019
A standardized protocol for manually segmenting stroke lesions on high-resolution T1-weighted MR images
Bethany P Lo1, Miranda R Donnelly1, Giuseppe Barisano2
1Chan Division of Occupational Science and Occupational Therapy, University of Southern California, Los Angeles, CA, United States.
Insights
This study presents a standardized protocol for manual stroke lesion segmentation on MRI scans. The detailed method ensures consistent tracing by multiple individuals, improving reproducibility in stroke research.
Area of Science:
- Neurology
- Medical Imaging
Background:
- Manual segmentation remains the gold standard for stroke lesion analysis despite advancements in automated methods.
- Accurate lesion segmentation is crucial for understanding stroke pathophysiology and treatment efficacy.
Purpose of the Study:
- To describe a detailed, standardized protocol for manual stroke lesion tracing on 3D T1-weighted MRI.
- To provide a reproducible method for training individuals in consistent lesion segmentation.
- To enhance the reliability of manual stroke lesion segmentation in research.
Main Methods:
- Development of a step-by-step protocol for manual lesion tracing on 3D T1-weighted MRI.
- Training of six individuals ('tracers') using the standardized protocol.
- Calculation of inter-rater and intra-rater reliability using intraclass correlation and Dice similarity coefficient.
Main Results:
- High inter-rater reliability (intraclass correlation: 0.98) and intra-rater reliability (intraclass correlation: 0.99).
- Strong agreement between raters (Dice similarity coefficient: 0.727) and within raters (Dice similarity coefficient: 0.839).
- Demonstrated consistency in lesion tracing across multiple trained individuals.
Conclusions:
- The developed protocol provides a standardized guideline for manual stroke lesion segmentation in T1-weighted MRI.
- The protocol effectively promotes reproducibility and consistency in stroke research.
- This standardized approach supports reliable manual analysis of stroke lesions.
Abstract:
Although automated methods for stroke lesion segmentation exist, many researchers still rely on manual segmentation as the gold standard. Our detailed, standardized protocol for stroke lesion tracing on high-resolution 3D T1-weighted (T1w) magnetic resonance imaging (MRI) has been used to trace over 1,300 stroke MRI. In the current study, we describe the protocol, including a step-by-step method utilized for training multiple individuals to trace lesions ("tracers") in a consistent manner and suggestions for distinguishing between lesioned and non-lesioned areas in stroke brains. Inter-rater and intra-rater reliability were calculated across six tracers trained using our protocol, resulting in an average intraclass correlation of 0.98 and 0.99, respectively, as well as a Dice similarity coefficient of 0.727 and 0.839, respectively. This protocol provides a standardized guideline for researchers performing manual lesion segmentation in stroke T1-weighted MRI, with detailed methods to promote reproducibility in stroke research.
