Automatic thalamus and hippocampus segmentation from MP2RAGE: comparison of publicly available methods and implications for DTI quantification

Næss‐Schmidt, Erhard Trillingsgaard; Tietze, Anna; Blicher, Jakob Udby; Petersen, Mikkel V.; Mikkelsen, Irene Klærke; Coupé, Pierrick; Manjón, José V.; Eskildsen, Simon Fristed

doi:10.1007/s11548-016-1433-0

Cited by 51 publications

(46 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Manual delineations from HHP (Boccardi et al, 2015), representing state-of-the-art in manual hippocampus segmentation, were used as expert priors (Anker, 2014, Anker et al, 2014). The method has previously demonstrated a better atrophy-based AD diagnostic performance than static FreeSurfer (Anker et al, 2014), and have in recent comparison studies performed by other research groups demonstrated better correspondence with manual segmentations than FreeSurfer (Manjón and Coupé, 2016, Næss-Schmidt et al, 2016). …”

Section: Methodsmentioning

confidence: 96%

Differential diagnosis of mild cognitive impairment and Alzheimer's disease using structural MRI cortical thickness, hippocampal shape, hippocampal texture, and volumetry

Sørensen

Igel

Pai

et al. 2017

NeuroImage: Clinical

159

View full text Add to dashboard Cite

This paper presents a brain T1-weighted structural magnetic resonance imaging (MRI) biomarker that combines several individual MRI biomarkers (cortical thickness measurements, volumetric measurements, hippocampal shape, and hippocampal texture). The method was developed, trained, and evaluated using two publicly available reference datasets: a standardized dataset from the Alzheimer's Disease Neuroimaging Initiative (ADNI) and the imaging arm of the Australian Imaging Biomarkers and Lifestyle flagship study of ageing (AIBL). In addition, the method was evaluated by participation in the Computer-Aided Diagnosis of Dementia (CADDementia) challenge. Cross-validation using ADNI and AIBL data resulted in a multi-class classification accuracy of 62.7% for the discrimination of healthy normal controls (NC), subjects with mild cognitive impairment (MCI), and patients with Alzheimer's disease (AD). This performance generalized to the CADDementia challenge where the method, trained using the ADNI and AIBL data, achieved a classification accuracy 63.0%. The obtained classification accuracy resulted in a first place in the challenge, and the method was significantly better (McNemar's test) than the bottom 24 methods out of the total of 29 methods contributed by 15 different teams in the challenge. The method was further investigated with learning curve and feature selection experiments using ADNI and AIBL data. The learning curve experiments suggested that neither more training data nor a more complex classifier would have improved the obtained results. The feature selection experiment showed that both common and uncommon individual MRI biomarkers contributed to the performance; hippocampal volume, ventricular volume, hippocampal texture, and parietal lobe thickness were the most important. This study highlights the need for both subtle, localized measurements and global measurements in order to discriminate NC, MCI, and AD simultaneously based on a single structural MRI scan. It is likely that additional non-structural MRI features are needed to further improve the obtained performance, especially to improve the discrimination between NC and MCI.

show abstract

Section: Methodsmentioning

confidence: 96%

Differential diagnosis of mild cognitive impairment and Alzheimer's disease using structural MRI cortical thickness, hippocampal shape, hippocampal texture, and volumetry

Sørensen

Igel

Pai

et al. 2017

NeuroImage: Clinical

159

View full text Add to dashboard Cite

show abstract

“…Previous studies have explored segmentation of MP2RAGE using traditional software packages as well as customized algorithms . Although some promising results were shown, there are still some unresolved issues.…”

Section: Discussionmentioning

confidence: 99%

“…A complex ratio of the two volumes was used to generate uniform T 1 ‐weighted images and quantitative T 1 maps insensitive to B 1 field inhomogeneity . A few studies have explored brain segmentation of the uniform T 1 ‐weighted images using both conventional segmentation tools and customized methods for MP2RAGE . Although most existing segmentation methods are mainly optimized for conventional T 1 ‐weighted images, some studies showed promising segmentation performance using MP2RAGE compared to using MPRAGE with commonly used softwares .…”

mentioning

confidence: 99%

Segmentation of gray matter, white matter, and CSF with fluid and white matter suppression using MP2RAGE

Wang

Zhang

et al. 2018

Magnetic Resonance Imaging

View full text Add to dashboard Cite

show abstract

“…compared the performances of publicly available segmentation tools (volBrain, FSL, FreeSurfer and SPM) and their impact on diffusion quantification, emphasizing the importance of using recently developed segmentation algorithms and imaging techniques. They found that volBrain is superior in thalamus and hippocampus segmentation compared to FSL, FreeSurfer and SPM 3 . Shape is important information for brain structure segmentation.…”

Section: Introductionmentioning

confidence: 99%

Automatic Thalamus Segmentation from Magnetic Resonance Images Using Multiple Atlases Level Set Framework (MALSF)

Zhang

Feng

2017

Sci Rep

View full text Add to dashboard Cite

In this paper, we present an original multiple atlases level set framework (MALSF) for automatic, accurate and robust thalamus segmentation in magnetic resonance images (MRI). The contributions of the MALSF method are twofold. First, the main technical contribution is a novel label fusion strategy in the level set framework. Label fusion is achieved by seeking an optimal level set function that minimizes energy functional with three terms: label fusion term, image based term, and regularization term. This strategy integrates shape prior, image information and the regularity of the thalamus. Second, we use propagated labels from multiple registration methods with different parameters to take full advantage of the complementary information of different registration methods. Since different registration methods and different atlases can yield complementary information, multiple registration and multiple atlases can be incorporated into the level set framework to improve the segmentation performance. Experiments have shown that the MALSF method can improve the segmentation accuracy for the thalamus. Compared to ground truth segmentation, the mean Dice metrics of our method are 0.9239 and 0.9200 for left and right thalamus.

show abstract

Automatic thalamus and hippocampus segmentation from MP2RAGE: comparison of publicly available methods and implications for DTI quantification

Abstract: Overall we find that volBrain is superior in thalamus and hippocampus segmentation compared to FSL, FreeSurfer and SPM. Furthermore, the choice of segmentation technique and training library affects quantitative results from diffusivity measures in thalamus and hippocampus.

Cited by 51 publications

References 25 publications

Differential diagnosis of mild cognitive impairment and Alzheimer's disease using structural MRI cortical thickness, hippocampal shape, hippocampal texture, and volumetry

Differential diagnosis of mild cognitive impairment and Alzheimer's disease using structural MRI cortical thickness, hippocampal shape, hippocampal texture, and volumetry

Segmentation of gray matter, white matter, and CSF with fluid and white matter suppression using MP2RAGE

Automatic Thalamus Segmentation from Magnetic Resonance Images Using Multiple Atlases Level Set Framework (MALSF)

Contact Info

Product

Resources

About