A family of boundary overlap metrics for the evaluation of medical image segmentation
All medical image segmentation algorithms need to be validated and compared, yet no evaluation framework is widely accepted within the imaging community. None of the evaluation metrics which are popular in the literature are consistent in the way they rank segmentation results: they tend to be sensi...
| Main Authors: | , |
|---|---|
| Format: | Journal article |
| Published: |
Society of Photo-optical Instrumentation Engineers
2018
|
| _version_ | 1797078056500723712 |
|---|---|
| author | Yeghiazaryan, V Voiculescu, I |
| author_facet | Yeghiazaryan, V Voiculescu, I |
| author_sort | Yeghiazaryan, V |
| collection | OXFORD |
| description | All medical image segmentation algorithms need to be validated and compared, yet no evaluation framework is widely accepted within the imaging community. None of the evaluation metrics which are popular in the literature are consistent in the way they rank segmentation results: they tend to be sensitive to one or another type of segmentation error (size, location, shape) but no single metric covers all error types. We introduce a new family of metrics, with hybrid characteristics. These metrics quantify the similarity or difference of segmented regions by considering their average overlap in fixed-size neighbourhoods of points on the boundaries of those regions. Our metrics are more sensitive to combinations of segmentation error types than other metrics in the existing literature. We compare the metric performance on collections of segmentation results sourced from carefully compiled 2D synthetic data and 3D medical images. We show that our metrics: (1) penalize errors successfully, especially those around region boundaries; (2) give a low similarity score when existing metrics disagree, thus avoiding overly inflated scores; and (3) score segmentation results over a wider range of values. We analyze a representative metric from this family and the effect of its free parameter on error sensitivity and running time. |
| first_indexed | 2024-03-07T00:26:59Z |
| format | Journal article |
| id | oxford-uuid:7e78a014-24cd-4195-8383-aa305870c397 |
| institution | University of Oxford |
| last_indexed | 2024-03-07T00:26:59Z |
| publishDate | 2018 |
| publisher | Society of Photo-optical Instrumentation Engineers |
| record_format | dspace |
| spelling | oxford-uuid:7e78a014-24cd-4195-8383-aa305870c3972022-03-26T21:10:24ZA family of boundary overlap metrics for the evaluation of medical image segmentationJournal articlehttp://purl.org/coar/resource_type/c_dcae04bcuuid:7e78a014-24cd-4195-8383-aa305870c397Symplectic Elements at OxfordSociety of Photo-optical Instrumentation Engineers2018Yeghiazaryan, VVoiculescu, IAll medical image segmentation algorithms need to be validated and compared, yet no evaluation framework is widely accepted within the imaging community. None of the evaluation metrics which are popular in the literature are consistent in the way they rank segmentation results: they tend to be sensitive to one or another type of segmentation error (size, location, shape) but no single metric covers all error types. We introduce a new family of metrics, with hybrid characteristics. These metrics quantify the similarity or difference of segmented regions by considering their average overlap in fixed-size neighbourhoods of points on the boundaries of those regions. Our metrics are more sensitive to combinations of segmentation error types than other metrics in the existing literature. We compare the metric performance on collections of segmentation results sourced from carefully compiled 2D synthetic data and 3D medical images. We show that our metrics: (1) penalize errors successfully, especially those around region boundaries; (2) give a low similarity score when existing metrics disagree, thus avoiding overly inflated scores; and (3) score segmentation results over a wider range of values. We analyze a representative metric from this family and the effect of its free parameter on error sensitivity and running time. |
| spellingShingle | Yeghiazaryan, V Voiculescu, I A family of boundary overlap metrics for the evaluation of medical image segmentation |
| title | A family of boundary overlap metrics for the evaluation of medical image segmentation |
| title_full | A family of boundary overlap metrics for the evaluation of medical image segmentation |
| title_fullStr | A family of boundary overlap metrics for the evaluation of medical image segmentation |
| title_full_unstemmed | A family of boundary overlap metrics for the evaluation of medical image segmentation |
| title_short | A family of boundary overlap metrics for the evaluation of medical image segmentation |
| title_sort | family of boundary overlap metrics for the evaluation of medical image segmentation |
| work_keys_str_mv | AT yeghiazaryanv afamilyofboundaryoverlapmetricsfortheevaluationofmedicalimagesegmentation AT voiculescui afamilyofboundaryoverlapmetricsfortheevaluationofmedicalimagesegmentation AT yeghiazaryanv familyofboundaryoverlapmetricsfortheevaluationofmedicalimagesegmentation AT voiculescui familyofboundaryoverlapmetricsfortheevaluationofmedicalimagesegmentation |