MISM: A Medical Image Segmentation Metric for Evaluation of Weak Labeled Data

Hartmann, Dennis; Schmid, Verena; Meyer, Philip; Auer, Florian; Soto-Rey, Iñaki; Müller, Dominik; Kramer, Frank

doi:10.3390/diagnostics13162618

Open AccessBrief Report

MISM: A Medical Image Segmentation Metric for Evaluation of Weak Labeled Data

by

Dennis Hartmann

^1,†

,

Verena Schmid

^1,2,†,

Philip Meyer

²,

Florian Auer

¹

,

Iñaki Soto-Rey

²

,

Dominik Müller

^1,2,*,†

and

Frank Kramer

¹

IT-Infrastructure for Translational Medical Research, University of Augsburg, 86159 Augsburg, Germany

²

Medical Data Integration Center, Institute for Digital Medicine, University Hospital Augsburg, 86156 Augsburg, Germany

^*

Author to whom correspondence should be addressed.

^†

These authors contributed equally to this work.

Diagnostics 2023, 13(16), 2618; https://doi.org/10.3390/diagnostics13162618

Submission received: 20 June 2023 / Revised: 31 July 2023 / Accepted: 1 August 2023 / Published: 8 August 2023

(This article belongs to the Special Issue Deep Learning for Medical Imaging Diagnosis)

Download

Browse Figures

Versions Notes

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.

Keywords: medical image analysis; biomedical image segmentation; evaluation; performance assessment

Share and Cite

MDPI and ACS Style

Hartmann, D.; Schmid, V.; Meyer, P.; Auer, F.; Soto-Rey, I.; Müller, D.; Kramer, F. MISM: A Medical Image Segmentation Metric for Evaluation of Weak Labeled Data. Diagnostics 2023, 13, 2618. https://doi.org/10.3390/diagnostics13162618

AMA Style

Hartmann D, Schmid V, Meyer P, Auer F, Soto-Rey I, Müller D, Kramer F. MISM: A Medical Image Segmentation Metric for Evaluation of Weak Labeled Data. Diagnostics. 2023; 13(16):2618. https://doi.org/10.3390/diagnostics13162618

Chicago/Turabian Style

Hartmann, Dennis, Verena Schmid, Philip Meyer, Florian Auer, Iñaki Soto-Rey, Dominik Müller, and Frank Kramer. 2023. "MISM: A Medical Image Segmentation Metric for Evaluation of Weak Labeled Data" Diagnostics 13, no. 16: 2618. https://doi.org/10.3390/diagnostics13162618

APA Style

Hartmann, D., Schmid, V., Meyer, P., Auer, F., Soto-Rey, I., Müller, D., & Kramer, F. (2023). MISM: A Medical Image Segmentation Metric for Evaluation of Weak Labeled Data. Diagnostics, 13(16), 2618. https://doi.org/10.3390/diagnostics13162618

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Menu

MISM: A Medical Image Segmentation Metric for Evaluation of Weak Labeled Data

Abstract

Share and Cite

Article Metrics

Article Access Statistics

Further Information

Guidelines

MDPI Initiatives

Follow MDPI