Automated identification of brain tumours from single MR images based on segmentation with refined patient-specific priors
Brain tumours can have different shapes or locations, making their identification very challenging. In functional MRI, it is not unusual that patients have only one anatomical image due to time and financial constraints. Here, we provide a modified automatic lesion identification (ALI) procedure whi...
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2013-12-01
|
| Series: | Frontiers in Neuroscience |
| Subjects: | |
| Online Access: | http://journal.frontiersin.org/Journal/10.3389/fnins.2013.00241/full |
| _version_ | 1819087584895172608 |
|---|---|
| author | Ana eSanjuán Ana eSanjuán Cathy J Price Laura eMancini Goulven eJosse Alice eGrogan Adam K Yamamoto Sharon eGeva Alex P Leff Alex P Leff Tarek A Yousry Mohamed L Seghier |
| author_facet | Ana eSanjuán Ana eSanjuán Cathy J Price Laura eMancini Goulven eJosse Alice eGrogan Adam K Yamamoto Sharon eGeva Alex P Leff Alex P Leff Tarek A Yousry Mohamed L Seghier |
| author_sort | Ana eSanjuán |
| collection | DOAJ |
| description | Brain tumours can have different shapes or locations, making their identification very challenging. In functional MRI, it is not unusual that patients have only one anatomical image due to time and financial constraints. Here, we provide a modified automatic lesion identification (ALI) procedure which enables brain tumour identification from single MR images. Our method rests on (A) a modified segmentation-normalisation procedure with an explicit extra prior for the tumour and (B) an outlier detection procedure for abnormal voxel (i.e. tumour) classification. To minimise tissue misclassification, the segmentation-normalisation procedure requires prior information of the tumour location and extent. We therefore propose that ALI is run iteratively so that the output of Step B is used as a patient-specific prior in Step A. We test this procedure on real T1-weighted images from 18 patients, and the results were validated in comparison to two independent observers’ manual tracings. The automated procedure identified the tumours successfully with an excellent agreement with the manual segmentation (area under the ROC curve = 0.97 ± 0.03). The proposed procedure increases the flexibility and robustness of the ALI tool and will be particularly useful for lesion-behaviour mapping studies, or when lesion identification and/or spatial normalisation are problematic. |
| first_indexed | 2024-12-21T21:38:28Z |
| format | Article |
| id | doaj.art-6116669badac4c478cbaf8f91f41349f |
| institution | Directory Open Access Journal |
| issn | 1662-453X |
| language | English |
| last_indexed | 2024-12-21T21:38:28Z |
| publishDate | 2013-12-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Neuroscience |
| spelling | doaj.art-6116669badac4c478cbaf8f91f41349f2022-12-21T18:49:26ZengFrontiers Media S.A.Frontiers in Neuroscience1662-453X2013-12-01710.3389/fnins.2013.0024164274Automated identification of brain tumours from single MR images based on segmentation with refined patient-specific priorsAna eSanjuán0Ana eSanjuán1Cathy J Price2Laura eMancini3Goulven eJosse4Alice eGrogan5Adam K Yamamoto6Sharon eGeva7Alex P Leff8Alex P Leff9Tarek A Yousry10Mohamed L Seghier11University College of LondonUniversitat Jaume IUniversity College of LondonNational Hospital for Neurology and NeurosurgeryHôpital de la Pitié-SalpêtrièreUniversity College of LondonNational Hospital for Neurology and NeurosurgeryUniversity College of London, Institute of Child HealthUniversity College of LondonUniversity College of LondonNational Hospital for Neurology and NeurosurgeryUniversity College of LondonBrain tumours can have different shapes or locations, making their identification very challenging. In functional MRI, it is not unusual that patients have only one anatomical image due to time and financial constraints. Here, we provide a modified automatic lesion identification (ALI) procedure which enables brain tumour identification from single MR images. Our method rests on (A) a modified segmentation-normalisation procedure with an explicit extra prior for the tumour and (B) an outlier detection procedure for abnormal voxel (i.e. tumour) classification. To minimise tissue misclassification, the segmentation-normalisation procedure requires prior information of the tumour location and extent. We therefore propose that ALI is run iteratively so that the output of Step B is used as a patient-specific prior in Step A. We test this procedure on real T1-weighted images from 18 patients, and the results were validated in comparison to two independent observers’ manual tracings. The automated procedure identified the tumours successfully with an excellent agreement with the manual segmentation (area under the ROC curve = 0.97 ± 0.03). The proposed procedure increases the flexibility and robustness of the ALI tool and will be particularly useful for lesion-behaviour mapping studies, or when lesion identification and/or spatial normalisation are problematic.http://journal.frontiersin.org/Journal/10.3389/fnins.2013.00241/fullMRIsegmentationspatial normalizationfuzzy clusteringAutomated lesion identification |
| spellingShingle | Ana eSanjuán Ana eSanjuán Cathy J Price Laura eMancini Goulven eJosse Alice eGrogan Adam K Yamamoto Sharon eGeva Alex P Leff Alex P Leff Tarek A Yousry Mohamed L Seghier Automated identification of brain tumours from single MR images based on segmentation with refined patient-specific priors Frontiers in Neuroscience MRI segmentation spatial normalization fuzzy clustering Automated lesion identification |
| title | Automated identification of brain tumours from single MR images based on segmentation with refined patient-specific priors |
| title_full | Automated identification of brain tumours from single MR images based on segmentation with refined patient-specific priors |
| title_fullStr | Automated identification of brain tumours from single MR images based on segmentation with refined patient-specific priors |
| title_full_unstemmed | Automated identification of brain tumours from single MR images based on segmentation with refined patient-specific priors |
| title_short | Automated identification of brain tumours from single MR images based on segmentation with refined patient-specific priors |
| title_sort | automated identification of brain tumours from single mr images based on segmentation with refined patient specific priors |
| topic | MRI segmentation spatial normalization fuzzy clustering Automated lesion identification |
| url | http://journal.frontiersin.org/Journal/10.3389/fnins.2013.00241/full |
| work_keys_str_mv | AT anaesanjuan automatedidentificationofbraintumoursfromsinglemrimagesbasedonsegmentationwithrefinedpatientspecificpriors AT anaesanjuan automatedidentificationofbraintumoursfromsinglemrimagesbasedonsegmentationwithrefinedpatientspecificpriors AT cathyjprice automatedidentificationofbraintumoursfromsinglemrimagesbasedonsegmentationwithrefinedpatientspecificpriors AT lauraemancini automatedidentificationofbraintumoursfromsinglemrimagesbasedonsegmentationwithrefinedpatientspecificpriors AT goulvenejosse automatedidentificationofbraintumoursfromsinglemrimagesbasedonsegmentationwithrefinedpatientspecificpriors AT aliceegrogan automatedidentificationofbraintumoursfromsinglemrimagesbasedonsegmentationwithrefinedpatientspecificpriors AT adamkyamamoto automatedidentificationofbraintumoursfromsinglemrimagesbasedonsegmentationwithrefinedpatientspecificpriors AT sharonegeva automatedidentificationofbraintumoursfromsinglemrimagesbasedonsegmentationwithrefinedpatientspecificpriors AT alexpleff automatedidentificationofbraintumoursfromsinglemrimagesbasedonsegmentationwithrefinedpatientspecificpriors AT alexpleff automatedidentificationofbraintumoursfromsinglemrimagesbasedonsegmentationwithrefinedpatientspecificpriors AT tarekayousry automatedidentificationofbraintumoursfromsinglemrimagesbasedonsegmentationwithrefinedpatientspecificpriors AT mohamedlseghier automatedidentificationofbraintumoursfromsinglemrimagesbasedonsegmentationwithrefinedpatientspecificpriors |