Region-based U-nets for fast, accurate, and scalable deep brain segmentation: Application to Parkinson Plus Syndromes

The early diagnosis of age-related neurodegenerative diseases, which often progress to dementia, poses significant clinical challenges due to subtle and overlapping symptoms of these diseases at early stage. Automated MRI segmentation is important for early detection, as it offers consistent measure...

Full description

Saved in:
Bibliographic Details
Main Authors: Mengyu Li, Magnús Magnússon, Ingibjörg Kristjánsdóttir, Sigrún Helga Lund, Thilo van Eimeren, Lotta M. Ellingsen
Format: Article
Language:English
Published: Elsevier 2025-01-01
Series:NeuroImage: Clinical
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S2213158225000774
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849416056237457408
author Mengyu Li
Magnús Magnússon
Ingibjörg Kristjánsdóttir
Sigrún Helga Lund
Thilo van Eimeren
Lotta M. Ellingsen
author_facet Mengyu Li
Magnús Magnússon
Ingibjörg Kristjánsdóttir
Sigrún Helga Lund
Thilo van Eimeren
Lotta M. Ellingsen
author_sort Mengyu Li
collection DOAJ
description The early diagnosis of age-related neurodegenerative diseases, which often progress to dementia, poses significant clinical challenges due to subtle and overlapping symptoms of these diseases at early stage. Automated MRI segmentation is important for early detection, as it offers consistent measurements and the ability to detect subtle structural changes in the brain. Manual segmentation is impractical for large datasets or clinical use. Deep learning approaches provide fast processing, however, they often encounter graphics processing unit (GPU) memory constraints when handling large datasets. Here we introduce a deep learning-based approach using region-based U-nets specifically designed to segment 12 deep-brain structures relevant to Parkinson Plus Syndromes. By dividing the brain image into targeted regions around the brainstem, ventricular system, and striatum, our method optimizes GPU usage and significantly reduces training times, while maintaining high accuracy. Validating the proposed method on three datasets, including a 660-subject clinical dataset comprising both healthy controls and patients with various movement disorders, we demonstrate robustness and practical applicability in separating different diseases. The method achieves superior segmentation performance compared to state-of-the-art methods, with a mean Dice Similarity Coefficient (DSC) of 0.90, a 95% Hausdorff Distance (HD95) of 1.35 mm, and an Average Symmetric Surface Distance (ASSD) of 0.45 mm, showcasing its segmentation accuracy and robustness. Furthermore, our method outperforms these methods by reducing training time from several days to a few hours while providing a processing time of less than a second per subject. The source code and trained model will be made publicly available on GitHub.
format Article
id doaj-art-616436ebf3214d5492bb9baaa900dfa1
institution Kabale University
issn 2213-1582
language English
publishDate 2025-01-01
publisher Elsevier
record_format Article
series NeuroImage: Clinical
spelling doaj-art-616436ebf3214d5492bb9baaa900dfa12025-08-20T03:33:18ZengElsevierNeuroImage: Clinical2213-15822025-01-014710380710.1016/j.nicl.2025.103807Region-based U-nets for fast, accurate, and scalable deep brain segmentation: Application to Parkinson Plus SyndromesMengyu Li0Magnús Magnússon1Ingibjörg Kristjánsdóttir2Sigrún Helga Lund3Thilo van Eimeren4Lotta M. Ellingsen5University of Iceland, Faculty of Electrical and Computer Engineering, Reykjavik, IcelandUniversity of Iceland, Faculty of Electrical and Computer Engineering, Reykjavik, IcelandUniversity of Iceland, Faculty of Medicine, Reykjavik, IcelandUniversity of Iceland, Faculty of Physical Sciences, Reykjavik, IcelandUniversity of Cologne, Faculty of Medicine and University Hospital, Department of Nuclear Medicine and Neurology, Cologne, GermanyUniversity of Iceland, Faculty of Electrical and Computer Engineering, Reykjavik, Iceland; Correspondence to: Faculty of Electrical and Computer Engineering, University of Iceland, Tæknigarður Dunhagi 5, Reykjavik, Iceland.The early diagnosis of age-related neurodegenerative diseases, which often progress to dementia, poses significant clinical challenges due to subtle and overlapping symptoms of these diseases at early stage. Automated MRI segmentation is important for early detection, as it offers consistent measurements and the ability to detect subtle structural changes in the brain. Manual segmentation is impractical for large datasets or clinical use. Deep learning approaches provide fast processing, however, they often encounter graphics processing unit (GPU) memory constraints when handling large datasets. Here we introduce a deep learning-based approach using region-based U-nets specifically designed to segment 12 deep-brain structures relevant to Parkinson Plus Syndromes. By dividing the brain image into targeted regions around the brainstem, ventricular system, and striatum, our method optimizes GPU usage and significantly reduces training times, while maintaining high accuracy. Validating the proposed method on three datasets, including a 660-subject clinical dataset comprising both healthy controls and patients with various movement disorders, we demonstrate robustness and practical applicability in separating different diseases. The method achieves superior segmentation performance compared to state-of-the-art methods, with a mean Dice Similarity Coefficient (DSC) of 0.90, a 95% Hausdorff Distance (HD95) of 1.35 mm, and an Average Symmetric Surface Distance (ASSD) of 0.45 mm, showcasing its segmentation accuracy and robustness. Furthermore, our method outperforms these methods by reducing training time from several days to a few hours while providing a processing time of less than a second per subject. The source code and trained model will be made publicly available on GitHub.http://www.sciencedirect.com/science/article/pii/S2213158225000774MRISegmentationDeep neural networksParkinson-plus syndromesBrainstemVentricles
spellingShingle Mengyu Li
Magnús Magnússon
Ingibjörg Kristjánsdóttir
Sigrún Helga Lund
Thilo van Eimeren
Lotta M. Ellingsen
Region-based U-nets for fast, accurate, and scalable deep brain segmentation: Application to Parkinson Plus Syndromes
NeuroImage: Clinical
MRI
Segmentation
Deep neural networks
Parkinson-plus syndromes
Brainstem
Ventricles
title Region-based U-nets for fast, accurate, and scalable deep brain segmentation: Application to Parkinson Plus Syndromes
title_full Region-based U-nets for fast, accurate, and scalable deep brain segmentation: Application to Parkinson Plus Syndromes
title_fullStr Region-based U-nets for fast, accurate, and scalable deep brain segmentation: Application to Parkinson Plus Syndromes
title_full_unstemmed Region-based U-nets for fast, accurate, and scalable deep brain segmentation: Application to Parkinson Plus Syndromes
title_short Region-based U-nets for fast, accurate, and scalable deep brain segmentation: Application to Parkinson Plus Syndromes
title_sort region based u nets for fast accurate and scalable deep brain segmentation application to parkinson plus syndromes
topic MRI
Segmentation
Deep neural networks
Parkinson-plus syndromes
Brainstem
Ventricles
url http://www.sciencedirect.com/science/article/pii/S2213158225000774
work_keys_str_mv AT mengyuli regionbasedunetsforfastaccurateandscalabledeepbrainsegmentationapplicationtoparkinsonplussyndromes
AT magnusmagnusson regionbasedunetsforfastaccurateandscalabledeepbrainsegmentationapplicationtoparkinsonplussyndromes
AT ingibjorgkristjansdottir regionbasedunetsforfastaccurateandscalabledeepbrainsegmentationapplicationtoparkinsonplussyndromes
AT sigrunhelgalund regionbasedunetsforfastaccurateandscalabledeepbrainsegmentationapplicationtoparkinsonplussyndromes
AT thilovaneimeren regionbasedunetsforfastaccurateandscalabledeepbrainsegmentationapplicationtoparkinsonplussyndromes
AT lottamellingsen regionbasedunetsforfastaccurateandscalabledeepbrainsegmentationapplicationtoparkinsonplussyndromes