An automated method for segmenting white matter lesions through multi-level morphometric feature classification with application to lupus
We demonstrate an automated, multi-level method to segment white matter brain lesions and apply it to lupus. The method makes use of local morphometric features based on multiple MR sequences, including T1-weighted, T2-weighted, and Fluid Attenuated Inversion Recovery. After preprocessing, including...
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2010-04-01
|
| Series: | Frontiers in Human Neuroscience |
| Subjects: | |
| Online Access: | http://journal.frontiersin.org/Journal/10.3389/fnhum.2010.00027/full |
| _version_ | 1819145366130393088 |
|---|---|
| author | Mark Scully Mark Scully Blake Anderson Terran Lane Charles Gasparovic Charles Gasparovic Vince Magnotta Wilmer Sibbitt Carlos Roldan Ron Kikinis Henry Jeremy Bockholt Henry Jeremy Bockholt |
| author_facet | Mark Scully Mark Scully Blake Anderson Terran Lane Charles Gasparovic Charles Gasparovic Vince Magnotta Wilmer Sibbitt Carlos Roldan Ron Kikinis Henry Jeremy Bockholt Henry Jeremy Bockholt |
| author_sort | Mark Scully |
| collection | DOAJ |
| description | We demonstrate an automated, multi-level method to segment white matter brain lesions and apply it to lupus. The method makes use of local morphometric features based on multiple MR sequences, including T1-weighted, T2-weighted, and Fluid Attenuated Inversion Recovery. After preprocessing, including co-registration, brain extraction, bias correction, and intensity standardization, 49 features are calculated for each brain voxel based on local morphometry. At each level of segmentation a supervised classifier takes advantage of a different subset of the features to conservatively segment lesion voxels, passing on more difficult voxels to the next classifier. This multi-level approach allows for a fast lesion classification method with tunable trade-offs between sensitivity and specificity producing accuracy comparable to a human rater. |
| first_indexed | 2024-12-22T12:56:53Z |
| format | Article |
| id | doaj.art-eee98b2ae744467eb7b8271a6ff5789f |
| institution | Directory Open Access Journal |
| issn | 1662-5161 |
| language | English |
| last_indexed | 2024-12-22T12:56:53Z |
| publishDate | 2010-04-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Human Neuroscience |
| spelling | doaj.art-eee98b2ae744467eb7b8271a6ff5789f2022-12-21T18:25:06ZengFrontiers Media S.A.Frontiers in Human Neuroscience1662-51612010-04-01410.3389/fnhum.2010.000271176An automated method for segmenting white matter lesions through multi-level morphometric feature classification with application to lupusMark Scully0Mark Scully1Blake Anderson2Terran Lane3Charles Gasparovic4Charles Gasparovic5Vince Magnotta6Wilmer Sibbitt7Carlos Roldan8Ron Kikinis9Henry Jeremy Bockholt10Henry Jeremy Bockholt11Advanced Biomedical Informatics Group LLCThe University of New MexicoThe University of New MexicoThe University of New MexicoThe University of New MexicoThe University of New MexicoUniversity of Iowa Carver College of MedicineThe University of New MexicoThe University of New MexicoAdvanced Biomedical Informatics Group LLCAdvanced Biomedical Informatics Group LLCThe University of New MexicoWe demonstrate an automated, multi-level method to segment white matter brain lesions and apply it to lupus. The method makes use of local morphometric features based on multiple MR sequences, including T1-weighted, T2-weighted, and Fluid Attenuated Inversion Recovery. After preprocessing, including co-registration, brain extraction, bias correction, and intensity standardization, 49 features are calculated for each brain voxel based on local morphometry. At each level of segmentation a supervised classifier takes advantage of a different subset of the features to conservatively segment lesion voxels, passing on more difficult voxels to the next classifier. This multi-level approach allows for a fast lesion classification method with tunable trade-offs between sensitivity and specificity producing accuracy comparable to a human rater.http://journal.frontiersin.org/Journal/10.3389/fnhum.2010.00027/fullClassificationmachine learninglesionlupusMethodsegmentation |
| spellingShingle | Mark Scully Mark Scully Blake Anderson Terran Lane Charles Gasparovic Charles Gasparovic Vince Magnotta Wilmer Sibbitt Carlos Roldan Ron Kikinis Henry Jeremy Bockholt Henry Jeremy Bockholt An automated method for segmenting white matter lesions through multi-level morphometric feature classification with application to lupus Frontiers in Human Neuroscience Classification machine learning lesion lupus Method segmentation |
| title | An automated method for segmenting white matter lesions through multi-level morphometric feature classification with application to lupus |
| title_full | An automated method for segmenting white matter lesions through multi-level morphometric feature classification with application to lupus |
| title_fullStr | An automated method for segmenting white matter lesions through multi-level morphometric feature classification with application to lupus |
| title_full_unstemmed | An automated method for segmenting white matter lesions through multi-level morphometric feature classification with application to lupus |
| title_short | An automated method for segmenting white matter lesions through multi-level morphometric feature classification with application to lupus |
| title_sort | automated method for segmenting white matter lesions through multi level morphometric feature classification with application to lupus |
| topic | Classification machine learning lesion lupus Method segmentation |
| url | http://journal.frontiersin.org/Journal/10.3389/fnhum.2010.00027/full |
| work_keys_str_mv | AT markscully anautomatedmethodforsegmentingwhitematterlesionsthroughmultilevelmorphometricfeatureclassificationwithapplicationtolupus AT markscully anautomatedmethodforsegmentingwhitematterlesionsthroughmultilevelmorphometricfeatureclassificationwithapplicationtolupus AT blakeanderson anautomatedmethodforsegmentingwhitematterlesionsthroughmultilevelmorphometricfeatureclassificationwithapplicationtolupus AT terranlane anautomatedmethodforsegmentingwhitematterlesionsthroughmultilevelmorphometricfeatureclassificationwithapplicationtolupus AT charlesgasparovic anautomatedmethodforsegmentingwhitematterlesionsthroughmultilevelmorphometricfeatureclassificationwithapplicationtolupus AT charlesgasparovic anautomatedmethodforsegmentingwhitematterlesionsthroughmultilevelmorphometricfeatureclassificationwithapplicationtolupus AT vincemagnotta anautomatedmethodforsegmentingwhitematterlesionsthroughmultilevelmorphometricfeatureclassificationwithapplicationtolupus AT wilmersibbitt anautomatedmethodforsegmentingwhitematterlesionsthroughmultilevelmorphometricfeatureclassificationwithapplicationtolupus AT carlosroldan anautomatedmethodforsegmentingwhitematterlesionsthroughmultilevelmorphometricfeatureclassificationwithapplicationtolupus AT ronkikinis anautomatedmethodforsegmentingwhitematterlesionsthroughmultilevelmorphometricfeatureclassificationwithapplicationtolupus AT henryjeremybockholt anautomatedmethodforsegmentingwhitematterlesionsthroughmultilevelmorphometricfeatureclassificationwithapplicationtolupus AT henryjeremybockholt anautomatedmethodforsegmentingwhitematterlesionsthroughmultilevelmorphometricfeatureclassificationwithapplicationtolupus AT markscully automatedmethodforsegmentingwhitematterlesionsthroughmultilevelmorphometricfeatureclassificationwithapplicationtolupus AT markscully automatedmethodforsegmentingwhitematterlesionsthroughmultilevelmorphometricfeatureclassificationwithapplicationtolupus AT blakeanderson automatedmethodforsegmentingwhitematterlesionsthroughmultilevelmorphometricfeatureclassificationwithapplicationtolupus AT terranlane automatedmethodforsegmentingwhitematterlesionsthroughmultilevelmorphometricfeatureclassificationwithapplicationtolupus AT charlesgasparovic automatedmethodforsegmentingwhitematterlesionsthroughmultilevelmorphometricfeatureclassificationwithapplicationtolupus AT charlesgasparovic automatedmethodforsegmentingwhitematterlesionsthroughmultilevelmorphometricfeatureclassificationwithapplicationtolupus AT vincemagnotta automatedmethodforsegmentingwhitematterlesionsthroughmultilevelmorphometricfeatureclassificationwithapplicationtolupus AT wilmersibbitt automatedmethodforsegmentingwhitematterlesionsthroughmultilevelmorphometricfeatureclassificationwithapplicationtolupus AT carlosroldan automatedmethodforsegmentingwhitematterlesionsthroughmultilevelmorphometricfeatureclassificationwithapplicationtolupus AT ronkikinis automatedmethodforsegmentingwhitematterlesionsthroughmultilevelmorphometricfeatureclassificationwithapplicationtolupus AT henryjeremybockholt automatedmethodforsegmentingwhitematterlesionsthroughmultilevelmorphometricfeatureclassificationwithapplicationtolupus AT henryjeremybockholt automatedmethodforsegmentingwhitematterlesionsthroughmultilevelmorphometricfeatureclassificationwithapplicationtolupus |