A generalizable deep voxel-guided morphometry algorithm for the detection of subtle lesion dynamics in multiple sclerosis
IntroductionMultiple sclerosis (MS) is a chronic neurological disorder characterized by the progressive loss of myelin and axonal structures in the central nervous system. Accurate detection and monitoring of MS-related changes in brain structures are crucial for disease management and treatment eva...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2024-01-01
|
| Series: | Frontiers in Neuroscience |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fnins.2024.1326108/full |
| _version_ | 1827375022076854272 |
|---|---|
| author | Anish Raj Anish Raj Achim Gass Achim Gass Philipp Eisele Philipp Eisele Andreas Dabringhaus Matthias Kraemer Matthias Kraemer Frank G. Zöllner Frank G. Zöllner |
| author_facet | Anish Raj Anish Raj Achim Gass Achim Gass Philipp Eisele Philipp Eisele Andreas Dabringhaus Matthias Kraemer Matthias Kraemer Frank G. Zöllner Frank G. Zöllner |
| author_sort | Anish Raj |
| collection | DOAJ |
| description | IntroductionMultiple sclerosis (MS) is a chronic neurological disorder characterized by the progressive loss of myelin and axonal structures in the central nervous system. Accurate detection and monitoring of MS-related changes in brain structures are crucial for disease management and treatment evaluation. We propose a deep learning algorithm for creating Voxel-Guided Morphometry (VGM) maps from longitudinal MRI brain volumes for analyzing MS disease activity. Our approach focuses on developing a generalizable model that can effectively be applied to unseen datasets.MethodsLongitudinal MS patient high-resolution 3D T1-weighted follow-up imaging from three different MRI systems were analyzed. We employed a 3D residual U-Net architecture with attention mechanisms. The U-Net serves as the backbone, enabling spatial feature extraction from MRI volumes. Attention mechanisms are integrated to enhance the model's ability to capture relevant information and highlight salient regions. Furthermore, we incorporate image normalization by histogram matching and resampling techniques to improve the networks' ability to generalize to unseen datasets from different MRI systems across imaging centers. This ensures robust performance across diverse data sources.ResultsNumerous experiments were conducted using a dataset of 71 longitudinal MRI brain volumes of MS patients. Our approach demonstrated a significant improvement of 4.3% in mean absolute error (MAE) against the state-of-the-art (SOTA) method. Furthermore, the algorithm's generalizability was evaluated on two unseen datasets (n = 116) with an average improvement of 4.2% in MAE over the SOTA approach.DiscussionResults confirm that the proposed approach is fast and robust and has the potential for broader clinical applicability. |
| first_indexed | 2024-03-08T11:43:36Z |
| format | Article |
| id | doaj.art-973da63c3d0143b79e8f85afaf8d67ad |
| institution | Directory Open Access Journal |
| issn | 1662-453X |
| language | English |
| last_indexed | 2024-03-08T11:43:36Z |
| publishDate | 2024-01-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Neuroscience |
| spelling | doaj.art-973da63c3d0143b79e8f85afaf8d67ad2024-01-25T04:22:04ZengFrontiers Media S.A.Frontiers in Neuroscience1662-453X2024-01-011810.3389/fnins.2024.13261081326108A generalizable deep voxel-guided morphometry algorithm for the detection of subtle lesion dynamics in multiple sclerosisAnish Raj0Anish Raj1Achim Gass2Achim Gass3Philipp Eisele4Philipp Eisele5Andreas Dabringhaus6Matthias Kraemer7Matthias Kraemer8Frank G. Zöllner9Frank G. Zöllner10Computer Assisted Clinical Medicine, Medical Faculty Mannheim, Heidelberg University, Mannheim, Baden Württemberg, GermanyMannheim Institute for Intelligent Systems in Medicine, Medical Faculty Mannheim, Heidelberg University, Mannheim, Baden Württemberg, GermanyDepartment of Neurology, University Medical Centre Mannheim, Medical Faculty Mannheim, Heidelberg University, Mannheim, Baden Württemberg, GermanyMannheim Center for Translational Neurosciences, Heidelberg University, Mannheim, Baden Württemberg, GermanyDepartment of Neurology, University Medical Centre Mannheim, Medical Faculty Mannheim, Heidelberg University, Mannheim, Baden Württemberg, GermanyMannheim Center for Translational Neurosciences, Heidelberg University, Mannheim, Baden Württemberg, GermanyVGMorph GmbH, Mülheim an der Ruhr, Nordrhein-Westfalen, GermanyVGMorph GmbH, Mülheim an der Ruhr, Nordrhein-Westfalen, GermanyNeuroCentrum, Grevenbroich, Nordrhein-Westfalen, GermanyComputer Assisted Clinical Medicine, Medical Faculty Mannheim, Heidelberg University, Mannheim, Baden Württemberg, GermanyMannheim Institute for Intelligent Systems in Medicine, Medical Faculty Mannheim, Heidelberg University, Mannheim, Baden Württemberg, GermanyIntroductionMultiple sclerosis (MS) is a chronic neurological disorder characterized by the progressive loss of myelin and axonal structures in the central nervous system. Accurate detection and monitoring of MS-related changes in brain structures are crucial for disease management and treatment evaluation. We propose a deep learning algorithm for creating Voxel-Guided Morphometry (VGM) maps from longitudinal MRI brain volumes for analyzing MS disease activity. Our approach focuses on developing a generalizable model that can effectively be applied to unseen datasets.MethodsLongitudinal MS patient high-resolution 3D T1-weighted follow-up imaging from three different MRI systems were analyzed. We employed a 3D residual U-Net architecture with attention mechanisms. The U-Net serves as the backbone, enabling spatial feature extraction from MRI volumes. Attention mechanisms are integrated to enhance the model's ability to capture relevant information and highlight salient regions. Furthermore, we incorporate image normalization by histogram matching and resampling techniques to improve the networks' ability to generalize to unseen datasets from different MRI systems across imaging centers. This ensures robust performance across diverse data sources.ResultsNumerous experiments were conducted using a dataset of 71 longitudinal MRI brain volumes of MS patients. Our approach demonstrated a significant improvement of 4.3% in mean absolute error (MAE) against the state-of-the-art (SOTA) method. Furthermore, the algorithm's generalizability was evaluated on two unseen datasets (n = 116) with an average improvement of 4.2% in MAE over the SOTA approach.DiscussionResults confirm that the proposed approach is fast and robust and has the potential for broader clinical applicability.https://www.frontiersin.org/articles/10.3389/fnins.2024.1326108/fullmultiple sclerosisdeep learningbrain MRIvoxel-guided morphometrylongitudinal change detection mapattention mechanism |
| spellingShingle | Anish Raj Anish Raj Achim Gass Achim Gass Philipp Eisele Philipp Eisele Andreas Dabringhaus Matthias Kraemer Matthias Kraemer Frank G. Zöllner Frank G. Zöllner A generalizable deep voxel-guided morphometry algorithm for the detection of subtle lesion dynamics in multiple sclerosis Frontiers in Neuroscience multiple sclerosis deep learning brain MRI voxel-guided morphometry longitudinal change detection map attention mechanism |
| title | A generalizable deep voxel-guided morphometry algorithm for the detection of subtle lesion dynamics in multiple sclerosis |
| title_full | A generalizable deep voxel-guided morphometry algorithm for the detection of subtle lesion dynamics in multiple sclerosis |
| title_fullStr | A generalizable deep voxel-guided morphometry algorithm for the detection of subtle lesion dynamics in multiple sclerosis |
| title_full_unstemmed | A generalizable deep voxel-guided morphometry algorithm for the detection of subtle lesion dynamics in multiple sclerosis |
| title_short | A generalizable deep voxel-guided morphometry algorithm for the detection of subtle lesion dynamics in multiple sclerosis |
| title_sort | generalizable deep voxel guided morphometry algorithm for the detection of subtle lesion dynamics in multiple sclerosis |
| topic | multiple sclerosis deep learning brain MRI voxel-guided morphometry longitudinal change detection map attention mechanism |
| url | https://www.frontiersin.org/articles/10.3389/fnins.2024.1326108/full |
| work_keys_str_mv | AT anishraj ageneralizabledeepvoxelguidedmorphometryalgorithmforthedetectionofsubtlelesiondynamicsinmultiplesclerosis AT anishraj ageneralizabledeepvoxelguidedmorphometryalgorithmforthedetectionofsubtlelesiondynamicsinmultiplesclerosis AT achimgass ageneralizabledeepvoxelguidedmorphometryalgorithmforthedetectionofsubtlelesiondynamicsinmultiplesclerosis AT achimgass ageneralizabledeepvoxelguidedmorphometryalgorithmforthedetectionofsubtlelesiondynamicsinmultiplesclerosis AT philippeisele ageneralizabledeepvoxelguidedmorphometryalgorithmforthedetectionofsubtlelesiondynamicsinmultiplesclerosis AT philippeisele ageneralizabledeepvoxelguidedmorphometryalgorithmforthedetectionofsubtlelesiondynamicsinmultiplesclerosis AT andreasdabringhaus ageneralizabledeepvoxelguidedmorphometryalgorithmforthedetectionofsubtlelesiondynamicsinmultiplesclerosis AT matthiaskraemer ageneralizabledeepvoxelguidedmorphometryalgorithmforthedetectionofsubtlelesiondynamicsinmultiplesclerosis AT matthiaskraemer ageneralizabledeepvoxelguidedmorphometryalgorithmforthedetectionofsubtlelesiondynamicsinmultiplesclerosis AT frankgzollner ageneralizabledeepvoxelguidedmorphometryalgorithmforthedetectionofsubtlelesiondynamicsinmultiplesclerosis AT frankgzollner ageneralizabledeepvoxelguidedmorphometryalgorithmforthedetectionofsubtlelesiondynamicsinmultiplesclerosis AT anishraj generalizabledeepvoxelguidedmorphometryalgorithmforthedetectionofsubtlelesiondynamicsinmultiplesclerosis AT anishraj generalizabledeepvoxelguidedmorphometryalgorithmforthedetectionofsubtlelesiondynamicsinmultiplesclerosis AT achimgass generalizabledeepvoxelguidedmorphometryalgorithmforthedetectionofsubtlelesiondynamicsinmultiplesclerosis AT achimgass generalizabledeepvoxelguidedmorphometryalgorithmforthedetectionofsubtlelesiondynamicsinmultiplesclerosis AT philippeisele generalizabledeepvoxelguidedmorphometryalgorithmforthedetectionofsubtlelesiondynamicsinmultiplesclerosis AT philippeisele generalizabledeepvoxelguidedmorphometryalgorithmforthedetectionofsubtlelesiondynamicsinmultiplesclerosis AT andreasdabringhaus generalizabledeepvoxelguidedmorphometryalgorithmforthedetectionofsubtlelesiondynamicsinmultiplesclerosis AT matthiaskraemer generalizabledeepvoxelguidedmorphometryalgorithmforthedetectionofsubtlelesiondynamicsinmultiplesclerosis AT matthiaskraemer generalizabledeepvoxelguidedmorphometryalgorithmforthedetectionofsubtlelesiondynamicsinmultiplesclerosis AT frankgzollner generalizabledeepvoxelguidedmorphometryalgorithmforthedetectionofsubtlelesiondynamicsinmultiplesclerosis AT frankgzollner generalizabledeepvoxelguidedmorphometryalgorithmforthedetectionofsubtlelesiondynamicsinmultiplesclerosis |