Related Experiment Video
Updated: Jul 18, 2025
Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
Published on: November 30, 2022
MISM: A Medical Image Segmentation Metric for Evaluation of Weak Labeled Data
Dennis Hartmann1, Verena Schmid1,2, Philip Meyer2
1IT-Infrastructure for Translational Medical Research, University of Augsburg, 86159 Augsburg, Germany.
Abstract:
Performance measures are an important tool for assessing and comparing different medical image segmentation algorithms. Unfortunately, the current measures have their weaknesses when it comes to assessing certain edge cases. These limitations arise when images with a very small region of interest or without a region of interest at all are assessed. As a solution to these limitations, we propose a new medical image segmentation metric: MISm. This metric is a composition of the Dice similarity coefficient and the weighted specificity. MISm was investigated for definition gaps, an appropriate scoring gradient, and different weighting coefficients used to propose a constant value. Furthermore, an evaluation was performed by comparing the popular metrics in the medical image segmentation and MISm using images of magnet resonance tomography from several fictitious prediction scenarios. Our analysis shows that MISm can be applied in a general way and thus also covers the mentioned edge cases, which are not covered by other metrics, in a reasonable way. In order to allow easy access to MISm and therefore widespread application in the community, as well as reproducibility of experimental results, we included MISm in the publicly available evaluation framework MISeval.
Insights
A new medical image segmentation metric, MISm (Medical Image Segmentation metric), effectively addresses limitations of current measures, particularly for edge cases like small or absent regions of interest, ensuring broader applicability.
Area of Science:
- Medical Imaging
- Computer Vision
- Algorithm Evaluation
Background:
- Current performance measures for medical image segmentation algorithms have limitations.
- These limitations are evident when evaluating images with very small or absent regions of interest.
Purpose of the Study:
- To introduce and evaluate a novel medical image segmentation metric, MISm.
- To address the shortcomings of existing metrics in handling edge cases.
Main Methods:
- MISm is proposed as a composite metric combining the Dice similarity coefficient and weighted specificity.
- The metric was analyzed for definition gaps, scoring gradient, and optimal weighting coefficients.
- Evaluation involved comparing MISm with popular metrics using simulated MRI data across various scenarios.
Main Results:
- MISm demonstrates general applicability and effectively handles edge cases not covered by other metrics.
- The proposed metric provides a reasonable assessment in challenging segmentation scenarios.
Conclusions:
- MISm offers a more robust and comprehensive evaluation of medical image segmentation algorithms.
- The inclusion of MISm in the MISeval framework promotes its community adoption and ensures result reproducibility.
