Evaluating the effect of multiple sclerosis lesions on automatic brain structure segmentation
In recent years, many automatic brain structure segmentation methods have been proposed. However, these methods are commonly tested with non-lesioned brains and the effect of lesions on their performance has not been evaluated. Here, we analyze the effect of multiple sclerosis (MS) lesions on three...
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2017-01-01
|
| Series: | NeuroImage: Clinical |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2213158217301080 |
| _version_ | 1818922854384664576 |
|---|---|
| author | Sandra González-Villà Sergi Valverde Mariano Cabezas Deborah Pareto Joan C. Vilanova Lluís Ramió-Torrentà Àlex Rovira Arnau Oliver Xavier Lladó |
| author_facet | Sandra González-Villà Sergi Valverde Mariano Cabezas Deborah Pareto Joan C. Vilanova Lluís Ramió-Torrentà Àlex Rovira Arnau Oliver Xavier Lladó |
| author_sort | Sandra González-Villà |
| collection | DOAJ |
| description | In recent years, many automatic brain structure segmentation methods have been proposed. However, these methods are commonly tested with non-lesioned brains and the effect of lesions on their performance has not been evaluated. Here, we analyze the effect of multiple sclerosis (MS) lesions on three well-known automatic brain structure segmentation methods, namely, FreeSurfer, FIRST and multi-atlas fused by majority voting, which use learning-based, deformable and atlas-based strategies, respectively. To perform a quantitative analysis, 100 synthetic images of MS patients with a total of 2174 lesions are simulated on two public databases with available brain structure ground truth information (IBSR18 and MICCAI’12). The Dice similarity coefficient (DSC) differences and the volume differences between the healthy and the simulated images are calculated for the subcortical structures and the brainstem. We observe that the three strategies are affected when lesions are present. However, the effects of the lesions do not follow the same pattern; the lesions either make the segmentation method underperform or surprisingly augment the segmentation accuracy. The obtained results show that FreeSurfer is the method most affected by the presence of lesions, with DSC differences (generated − healthy) ranging from −0.11±0.54 to 9.65±9.87, whereas FIRST tends to be the most robust method when lesions are present (−2.40±5.54 to 0.44±0.94). Lesion location is not important for global strategies such as FreeSurfer or majority voting, where structure segmentation is affected wherever the lesions exist. On the other hand, FIRST is more affected when the lesions are overlaid or close to the structure of analysis. The most affected structure by the presence of lesions is the nucleus accumbens (from −1.12±2.53 to 1.32±4.00 for the left hemisphere and from −2.40±5.54 to 9.65±9.87 for the right hemisphere), whereas the structures that show less variation include the thalamus (from 0.03±0.35 to 0.74±0.89 and from −0.48±1.08 to −0.04±0.22) and the brainstem (from −0.20±0.38 to 1.03±1.31). The three segmentation approaches are affected by the presence of MS lesions, which demonstrates that there exists a problem in the automatic segmentation methods of the deep gray matter (DGM) structures that has to be taken into account when using them as a tool to measure the disease progression. Keywords: Brain structures, Multiple sclerosis lesions, Segmentation, Magnetic resonance imaging |
| first_indexed | 2024-12-20T02:00:09Z |
| format | Article |
| id | doaj.art-0d6c4f42e3ed45988c4fc00c8c766656 |
| institution | Directory Open Access Journal |
| issn | 2213-1582 |
| language | English |
| last_indexed | 2024-12-20T02:00:09Z |
| publishDate | 2017-01-01 |
| publisher | Elsevier |
| record_format | Article |
| series | NeuroImage: Clinical |
| spelling | doaj.art-0d6c4f42e3ed45988c4fc00c8c7666562022-12-21T19:57:20ZengElsevierNeuroImage: Clinical2213-15822017-01-0115228238Evaluating the effect of multiple sclerosis lesions on automatic brain structure segmentationSandra González-Villà0Sergi Valverde1Mariano Cabezas2Deborah Pareto3Joan C. Vilanova4Lluís Ramió-Torrentà5Àlex Rovira6Arnau Oliver7Xavier Lladó8Institute of Computer Vision and Robotics, University of Girona, Ed. P-IV, Campus Montilivi, 17073 Girona, Spain; Corresponding author.Institute of Computer Vision and Robotics, University of Girona, Ed. P-IV, Campus Montilivi, 17073 Girona, SpainInstitute of Computer Vision and Robotics, University of Girona, Ed. P-IV, Campus Montilivi, 17073 Girona, SpainMagnetic Resonance Unit, Dept of Radiology, Vall d'Hebron University Hospital, SpainGirona Magnetic Resonance Center, SpainMultiple Sclerosis and Neuroimmunology Unit, Dr. Josep Trueta University Hospital, SpainMagnetic Resonance Unit, Dept of Radiology, Vall d'Hebron University Hospital, SpainInstitute of Computer Vision and Robotics, University of Girona, Ed. P-IV, Campus Montilivi, 17073 Girona, SpainInstitute of Computer Vision and Robotics, University of Girona, Ed. P-IV, Campus Montilivi, 17073 Girona, SpainIn recent years, many automatic brain structure segmentation methods have been proposed. However, these methods are commonly tested with non-lesioned brains and the effect of lesions on their performance has not been evaluated. Here, we analyze the effect of multiple sclerosis (MS) lesions on three well-known automatic brain structure segmentation methods, namely, FreeSurfer, FIRST and multi-atlas fused by majority voting, which use learning-based, deformable and atlas-based strategies, respectively. To perform a quantitative analysis, 100 synthetic images of MS patients with a total of 2174 lesions are simulated on two public databases with available brain structure ground truth information (IBSR18 and MICCAI’12). The Dice similarity coefficient (DSC) differences and the volume differences between the healthy and the simulated images are calculated for the subcortical structures and the brainstem. We observe that the three strategies are affected when lesions are present. However, the effects of the lesions do not follow the same pattern; the lesions either make the segmentation method underperform or surprisingly augment the segmentation accuracy. The obtained results show that FreeSurfer is the method most affected by the presence of lesions, with DSC differences (generated − healthy) ranging from −0.11±0.54 to 9.65±9.87, whereas FIRST tends to be the most robust method when lesions are present (−2.40±5.54 to 0.44±0.94). Lesion location is not important for global strategies such as FreeSurfer or majority voting, where structure segmentation is affected wherever the lesions exist. On the other hand, FIRST is more affected when the lesions are overlaid or close to the structure of analysis. The most affected structure by the presence of lesions is the nucleus accumbens (from −1.12±2.53 to 1.32±4.00 for the left hemisphere and from −2.40±5.54 to 9.65±9.87 for the right hemisphere), whereas the structures that show less variation include the thalamus (from 0.03±0.35 to 0.74±0.89 and from −0.48±1.08 to −0.04±0.22) and the brainstem (from −0.20±0.38 to 1.03±1.31). The three segmentation approaches are affected by the presence of MS lesions, which demonstrates that there exists a problem in the automatic segmentation methods of the deep gray matter (DGM) structures that has to be taken into account when using them as a tool to measure the disease progression. Keywords: Brain structures, Multiple sclerosis lesions, Segmentation, Magnetic resonance imaginghttp://www.sciencedirect.com/science/article/pii/S2213158217301080 |
| spellingShingle | Sandra González-Villà Sergi Valverde Mariano Cabezas Deborah Pareto Joan C. Vilanova Lluís Ramió-Torrentà Àlex Rovira Arnau Oliver Xavier Lladó Evaluating the effect of multiple sclerosis lesions on automatic brain structure segmentation NeuroImage: Clinical |
| title | Evaluating the effect of multiple sclerosis lesions on automatic brain structure segmentation |
| title_full | Evaluating the effect of multiple sclerosis lesions on automatic brain structure segmentation |
| title_fullStr | Evaluating the effect of multiple sclerosis lesions on automatic brain structure segmentation |
| title_full_unstemmed | Evaluating the effect of multiple sclerosis lesions on automatic brain structure segmentation |
| title_short | Evaluating the effect of multiple sclerosis lesions on automatic brain structure segmentation |
| title_sort | evaluating the effect of multiple sclerosis lesions on automatic brain structure segmentation |
| url | http://www.sciencedirect.com/science/article/pii/S2213158217301080 |
| work_keys_str_mv | AT sandragonzalezvilla evaluatingtheeffectofmultiplesclerosislesionsonautomaticbrainstructuresegmentation AT sergivalverde evaluatingtheeffectofmultiplesclerosislesionsonautomaticbrainstructuresegmentation AT marianocabezas evaluatingtheeffectofmultiplesclerosislesionsonautomaticbrainstructuresegmentation AT deborahpareto evaluatingtheeffectofmultiplesclerosislesionsonautomaticbrainstructuresegmentation AT joancvilanova evaluatingtheeffectofmultiplesclerosislesionsonautomaticbrainstructuresegmentation AT lluisramiotorrenta evaluatingtheeffectofmultiplesclerosislesionsonautomaticbrainstructuresegmentation AT alexrovira evaluatingtheeffectofmultiplesclerosislesionsonautomaticbrainstructuresegmentation AT arnauoliver evaluatingtheeffectofmultiplesclerosislesionsonautomaticbrainstructuresegmentation AT xavierllado evaluatingtheeffectofmultiplesclerosislesionsonautomaticbrainstructuresegmentation |