3D Segmentation in the Clinic: A Grand Challenge II: MS lesion segmentation

Styner, Martin; Lee, Joohwi; Chin, Brian; Chin, Matthew; Commowick, Olivier; Tran, Hoai-Huong; Marković-Pleše, Silva; Jewells, Valerie; Warfield, Simon K.

doi:10.54294/lmkqvm

Cited by 97 publications

(82 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In this section, the most important available datasets used to diagnose MS using MRI neuroimaging modalities are discussed. There are several datasets available for researchers to diagnose MS, including MICCAI 2008 [77], MICCAI 2016 [78], ISBI 2015 [79][80], and eHealth Lab [81]. In the following, the details for available MS datasets based on MRI neuroimaging modalities are presented.…”

Section: Datasetsmentioning

confidence: 99%

“…This dataset contains MR images of research subjects from the University of North Carolina (UNC) and Boston Children's Hospital (CHB) [77]. All dataset images were segmented by one CHB expert and 2 UNC experts.…”

Section: The Miccai 2008 Ms Lesion Segmentation Challenge Datasetmentioning

confidence: 99%

“…All dataset images were segmented by one CHB expert and 2 UNC experts. The training database has 20 samples, of which 10 CHB samples and 10 UNC samples were manually segmented from the CHB expert [77]. The test dataset includes 25 samples (15 CHB samples and 10 UNC samples) without any segmentation [77].…”

Section: The Miccai 2008 Ms Lesion Segmentation Challenge Datasetmentioning

confidence: 99%

“…The training database has 20 samples, of which 10 CHB samples and 10 UNC samples were manually segmented from the CHB expert [77]. The test dataset includes 25 samples (15 CHB samples and 10 UNC samples) without any segmentation [77]. In the dataset, all the modalities of T1WI, T2WI, FLAIR, DTI-derived FA, and MD images are presented.…”

Section: The Miccai 2008 Ms Lesion Segmentation Challenge Datasetmentioning

confidence: 99%

See 3 more Smart Citations

Applications of deep learning techniques for automated multiple sclerosis detection using magnetic resonance imaging: A review

Shoeibi

Khodatars

Jafari

et al. 2021

Computers in Biology and Medicine

118

View full text Add to dashboard Cite

Section: Datasetsmentioning

confidence: 99%

Section: The Miccai 2008 Ms Lesion Segmentation Challenge Datasetmentioning

confidence: 99%

Section: The Miccai 2008 Ms Lesion Segmentation Challenge Datasetmentioning

confidence: 99%

Section: The Miccai 2008 Ms Lesion Segmentation Challenge Datasetmentioning

confidence: 99%

See 2 more Smart Citations

Applications of deep learning techniques for automated multiple sclerosis detection using magnetic resonance imaging: A review

Shoeibi

Khodatars

Jafari

et al. 2021

Computers in Biology and Medicine

118

View full text Add to dashboard Cite

“…Já a base de dados Multi-Modal MRI Reproducibility Resource (KIRBY) [6] contém 21 pacientes saudáveis, dispondo 42 exames de cada ponderamento FLAIR, T1 e T2. O repositório de imagens MICCAI08 [7], apresenta 51 pacientes diagnosticados com esclerose múltipla, divididos em 20 casos para treino e 31 casos para teste. O conjunto de dados conta com 51 exemplares de cada ponderamento de imagem FLAIR, T1 e T2.…”

Section: Trabalhos Relacionadosunclassified

UniMRI: Unified Repository of Magnetic Resonance Images for Multiple Sclerosis Diagnosis

Silveira¹,

Korb²,

Dimuro³

et al. 2021

Anais Do XVII Workshop De Visão Computacional (WVC 2021)

View full text Add to dashboard Cite

Multiple sclerosis is an autoimmune disease that affects the central nervous system, destroying myelin. To detect multiple sclerosis, you need to have MRI scans so you can see the areas where myelin has been damaged. This analysis is complex and costly due to the time required to assess injuries. The use of machine learning is desirable as these exams are taken periodically. However, the number of public databases present in the literature containing patients with multiple sclerosis is small when compared to the amount of data needed to train deep neural networks. Thus, the objective of this work is to join public databases of magnetic resonance images existing in the literature, proposing a software library to manipulate and pre-process these data.

show abstract

Efficient brain lesion segmentation using multi‐modality tissue‐based feature selection and support vector machines

Fiot

Cohen

Raniga

et al. 2013

Numer Methods Biomed Eng

View full text Add to dashboard Cite

Support vector machines (SVM) are machine learning techniques that have been used for segmentation and classification of medical images, including segmentation of white matter hyper-intensities (WMH). Current approaches using SVM for WMH segmentation extract features from the brain and classify these followed by complex post-processing steps to remove false positives. The method presented in this paper combines advanced pre-processing, tissue-based feature selection and SVM classification to obtain efficient and accurate WMH segmentation. Features from 125 patients, generated from up to four MR modalities [T1-w, T2-w, proton-density and fluid attenuated inversion recovery(FLAIR)], differing neighbourhood sizes and the use of multi-scale features were compared. We found that although using all four modalities gave the best overall classification (average Dice scores of 0.54 ± 0.12, 0.72 ± 0.06 and 0.82 ± 0.06 respectively for small, moderate and severe lesion loads); this was not significantly different (p = 0.50) from using just T1-w and FLAIR sequences (Dice scores of 0.52 ± 0.13, 0.71 ± 0.08 and 0.81 ± 0.07). Furthermore, there was a negligible difference between using 5 × 5 × 5 and 3 × 3 × 3 features (p = 0.93). Finally, we show that careful consideration of features and pre-processing techniques not only saves storage space and computation time but also leads to more efficient classification, which outperforms the one based on all features with post-processing.

show abstract

3D Segmentation in the Clinic: A Grand Challenge II: MS lesion segmentation

Cited by 97 publications

References 3 publications

Applications of deep learning techniques for automated multiple sclerosis detection using magnetic resonance imaging: A review

Applications of deep learning techniques for automated multiple sclerosis detection using magnetic resonance imaging: A review

UniMRI: Unified Repository of Magnetic Resonance Images for Multiple Sclerosis Diagnosis

Efficient brain lesion segmentation using multi‐modality tissue‐based feature selection and support vector machines

Contact Info

Product

Resources

About