Reproducibility of automated habenula segmentation via deep learning in major depressive disorder and normal controls with 7 Tesla MRI.
Ontology highlight
ABSTRACT:
SUBMITTER: Lim SH
PROVIDER: S-EPMC8241874 | biostudies-literature |
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:In vivo morphological study of the human habenula, a pair of small epithalamic nuclei adjacent to the dorsomedial thalamus, has recently gained significant interest for its role in reward and aversion processing. However, segmenting the habenula from in vivo magnetic resonance imaging (MRI) is challenging due to the habenula's small size and low anatomical contrast. Although manual and semi-automated habenula segmentation methods have been reported, the test-retest reproducibility of the segmented habenula volume and the consistency of the boundaries of habenula segmentation have not been investigated. In this study, we evaluated the intra- and inter-site reproducibility of in vivo human habenula segmentation from 3T MRI (0.7-0.8 mm isotropic resolution) using our previously proposed semi-automated myelin contrast-based method and its fully-automated version, as well as a previously published manual geometry-based method. The habenula segmentation using our semi-automated method showed consistent boundary definition (high Dice coefficient, low mean distance, and moderate Hausdorff distance) and reproducible volume measurement (low coefficient of variation). Furthermore, the habenula boundary in our semi-automated segmentation from 3T MRI agreed well with that in the manual segmentation from 7T MRI (0.5 mm isotropic resolution) of the same subjects. Overall, our proposed semi-automated habenula segmentation showed reliable and reproducible habenula localization, while its fully-automated version offers an efficient way for large sample analysis.
| S-EPMC6033622 | biostudies-literature
Project description:In the last two decades, neuroscience has produced intriguing evidence for a central role of the claustrum in mammalian forebrain structure and function. However, relatively few in vivo studies of the claustrum exist in humans. A reason for this may be the delicate and sheet-like structure of the claustrum lying between the insular cortex and the putamen, which makes it not amenable to conventional segmentation methods. Recently, Deep Learning (DL) based approaches have been successfully introduced for automated segmentation of complex, subcortical brain structures. In the following, we present a multi-view DL-based approach to segment the claustrum in T1-weighted MRI scans. We trained and evaluated the proposed method in 181 individuals, using bilateral manual claustrum annotations by an expert neuroradiologist as reference standard. Cross-validation experiments yielded median volumetric similarity, robust Hausdorff distance, and Dice score of 93.3%, 1.41 mm, and 71.8%, respectively, representing equal or superior segmentation performance compared to human intra-rater reliability. The leave-one-scanner-out evaluation showed good transferability of the algorithm to images from unseen scanners at slightly inferior performance. Furthermore, we found that DL-based claustrum segmentation benefits from multi-view information and requires a sample size of around 75 MRI scans in the training set. We conclude that the developed algorithm allows for robust automated claustrum segmentation and thus yields considerable potential for facilitating MRI-based research of the human claustrum. The software and models of our method are made publicly available.
| S-EPMC8596988 | biostudies-literature
Project description:IntroductionThe increasing use of large sample sizes for population and personalized medicine requires high-throughput tools for imaging processing that can handle large amounts of data with diverse image modalities, perform a biologically meaningful information reduction, and result in comprehensive quantification. Exploring the reproducibility of these tools reveals the specific strengths and weaknesses that heavily influence the interpretation of results, contributing to transparence in science.MethodsWe tested-retested the reproducibility of MRICloud, a free automated method for whole-brain, multimodal MRI segmentation and quantification, on two public, independent datasets of healthy adults.ResultsThe reproducibility was extremely high for T1-volumetric analysis, high for diffusion tensor images (DTI) (however, regionally variable), and low for resting-state fMRI.ConclusionIn general, the reproducibility of the different modalities was slightly superior to that of widely used software. This analysis serves as a normative reference for planning samples and for the interpretation of structure-based MRI studies.
| S-EPMC6790328 | biostudies-literature
Project description:PurposeTo study the interscan reproducibility of manual versus automated segmentation of carotid artery plaque components, and the agreement between both methods, in high and lower quality MRI scans.Methods24 patients with 30-70% carotid artery stenosis were planned for 3T carotid MRI, followed by a rescan within 1 month. A multicontrast protocol (T1w,T2w, PDw and TOF sequences) was used. After co-registration and delineation of the lumen and outer wall, segmentation of plaque components (lipid-rich necrotic cores (LRNC) and calcifications) was performed both manually and automated. Scan quality was assessed using a visual quality scale.ResultsAgreement for the detection of LRNC (Cohen's kappa (k) is 0.04) and calcification (k = 0.41) between both manual and automated segmentation methods was poor. In the high-quality scans (visual quality score ? 3), the agreement between manual and automated segmentation increased to k = 0.55 and k = 0.58 for, respectively, the detection of LRNC and calcification larger than 1 mm2. Both manual and automated analysis showed good interscan reproducibility for the quantification of LRNC (intraclass correlation coefficient (ICC) of 0.94 and 0.80 respectively) and calcified plaque area (ICC of 0.95 and 0.77, respectively).ConclusionAgreement between manual and automated segmentation of LRNC and calcifications was poor, despite a good interscan reproducibility of both methods. The agreement between both methods increased to moderate in high quality scans. These findings indicate that image quality is a critical determinant of the performance of both manual and automated segmentation of carotid artery plaque components.
| S-EPMC5145140 | biostudies-literature
Project description:Using semi-automated software simplifies quantitative analysis of the visible burden of disease on whole-body MRI diffusion-weighted images. To establish the intra- and inter-observer reproducibility of apparent diffusion coefficient (ADC) measures, we retrospectively analyzed data from 20 patients with bone metastases from breast (BCa; n = 10; aged 62.3 ± 14.8) or prostate cancer (PCa; n = 10; aged 67.4 ± 9.0) who had undergone examinations at two timepoints, before and after hormone-therapy. Four independent observers processed all images twice, first segmenting the entire skeleton on diffusion-weighted images, and then isolating bone metastases via ADC histogram thresholding (ADC: 650-1400 µm2/s). Dice Similarity, Bland-Altman method, and Intraclass Correlation Coefficient were used to assess reproducibility. Inter-observer Dice similarity was moderate (0.71) for women with BCa and poor (0.40) for men with PCa. Nonetheless, the limits of agreement of the mean ADC were just ±6% for women with BCa and ±10% for men with PCa (mean ADCs: 941 and 999 µm2/s, respectively). Inter-observer Intraclass Correlation Coefficients of the ADC histogram parameters were consistently greater in women with BCa than in men with PCa. While scope remains for improving consistency of the volume segmented, the observer-dependent variability measured in this study was appropriate to distinguish the clinically meaningful changes of ADC observed in patients responding to therapy, as changes of at least 25% are of interest.
| S-EPMC7998160 | biostudies-literature
Project description:Segmentation of anatomical structures is valuable in a variety of tasks, including 3D visualization, surgical planning, and quantitative image analysis. Manual segmentation is time-consuming and deals with intra and inter-observer variability. To develop a deep-learning approach for the fully automated segmentation of the inner ear in MRI, a 3D U-net was trained on 944 MRI scans with manually segmented inner ears as reference standard. The model was validated on an independent, multicentric dataset consisting of 177 MRI scans from three different centers. The model was also evaluated on a clinical validation set containing eight MRI scans with severe changes in the morphology of the labyrinth. The 3D U-net model showed precise Dice Similarity Coefficient scores (mean DSC-0.8790) with a high True Positive Rate (91.5%) and low False Discovery Rate and False Negative Rates (14.8% and 8.49% respectively) across images from three different centers. The model proved to perform well with a DSC of 0.8768 on the clinical validation dataset. The proposed auto-segmentation model is equivalent to human readers and is a reliable, consistent, and efficient method for inner ear segmentation, which can be used in a variety of clinical applications such as surgical planning and quantitative image analysis.
| S-EPMC7858625 | biostudies-literature
Project description:OBJECTIVES:Magnetic resonance imaging (MRI) is the method of choice for imaging meningiomas. Volumetric assessment of meningiomas is highly relevant for therapy planning and monitoring. We used a multiparametric deep-learning model (DLM) on routine MRI data including images from diverse referring institutions to investigate DLM performance in automated detection and segmentation of meningiomas in comparison to manual segmentations. METHODS:We included 56 of 136 consecutive preoperative MRI datasets [T1/T2-weighted, T1-weighted contrast-enhanced (T1CE), FLAIR] of meningiomas that were treated surgically at the University Hospital Cologne and graded histologically as tumour grade I (n = 38) or grade II (n = 18). The DLM was trained on an independent dataset of 249 glioma cases and segmented different tumour classes as defined in the brain tumour image segmentation benchmark (BRATS benchmark). The DLM was based on the DeepMedic architecture. Results were compared to manual segmentations by two radiologists in a consensus reading in FLAIR and T1CE. RESULTS:The DLM detected meningiomas in 55 of 56 cases. Further, automated segmentations correlated strongly with manual segmentations: average Dice coefficients were 0.81 ± 0.10 (range, 0.46-0.93) for the total tumour volume (union of tumour volume in FLAIR and T1CE) and 0.78 ± 0.19 (range, 0.27-0.95) for contrast-enhancing tumour volume in T1CE. CONCLUSIONS:The DLM yielded accurate automated detection and segmentation of meningioma tissue despite diverse scanner data and thereby may improve and facilitate therapy planning as well as monitoring of this highly frequent tumour entity. KEY POINTS:• Deep learning allows for accurate meningioma detection and segmentation • Deep learning helps clinicians to assess patients with meningiomas • Meningioma monitoring and treatment planning can be improved.
| S-EPMC6291436 | biostudies-literature
Project description:Deep brain stimulation (DBS) surgery has been shown to dramatically improve the quality of life for patients with various motor dysfunctions, such as those afflicted with Parkinson's disease (PD), dystonia, and essential tremor (ET), by relieving motor symptoms associated with such pathologies. The success of DBS procedures is directly related to the proper placement of the electrodes, which requires the ability to accurately detect and identify relevant target structures within the subcortical basal ganglia region. In particular, accurate and reliable segmentation of the globus pallidus (GP) interna is of great interest for DBS surgery for PD and dystonia. In this study, we present a deep-learning based neural network, which we term GP-net, for the automatic segmentation of both the external and internal segments of the globus pallidus. High resolution 7 Tesla images from 101 subjects were used in this study; GP-net is trained on a cohort of 58 subjects, containing patients with movement disorders as well as healthy control subjects. GP-net performs 3D inference in a patient-specific manner, alleviating the need for atlas-based segmentation. GP-net was extensively validated, both quantitatively and qualitatively over 43 test subjects including patients with movement disorders and healthy control and is shown to consistently produce improved segmentation results compared with state-of-the-art atlas-based segmentations. We also demonstrate a postoperative lead location assessment with respect to a segmented globus pallidus obtained by GP-net.
| S-EPMC8127160 | biostudies-literature
Project description:PurposeIntroduce and validate a novel, fast, and fully automated deep learning pipeline (FatSegNet) to accurately identify, segment, and quantify visceral and subcutaneous adipose tissue (VAT and SAT) within a consistent, anatomically defined abdominal region on Dixon MRI scans.MethodsFatSegNet is composed of three stages: (a) Consistent localization of the abdominal region using two 2D-Competitive Dense Fully Convolutional Networks (CDFNet), (b) Segmentation of adipose tissue on three views by independent CDFNets, and (c) View aggregation. FatSegNet is validated by: (1) comparison of segmentation accuracy (sixfold cross-validation), (2) test-retest reliability, (3) generalizability to randomly selected manually re-edited cases, and (4) replication of age and sex effects in the Rhineland Study-a large prospective population cohort.ResultsThe CDFNet demonstrates increased accuracy and robustness compared to traditional deep learning networks. FatSegNet Dice score outperforms manual raters on VAT (0.850 vs. 0.788) and produces comparable results on SAT (0.975 vs. 0.982). The pipeline has excellent agreement for both test-retest (ICC VAT 0.998 and SAT 0.996) and manual re-editing (ICC VAT 0.999 and SAT 0.999).ConclusionsFatSegNet generalizes well to different body shapes, sensitively replicates known VAT and SAT volume effects in a large cohort study and permits localized analysis of fat compartments. Furthermore, it can reliably analyze a 3D Dixon MRI in ∼1 minute, providing an efficient and validated pipeline for abdominal adipose tissue analysis in the Rhineland Study.
| S-EPMC6949410 | biostudies-literature
Project description:PurposeTo generate fully automated and fast 4D-flow MRI-based 3D segmentations of the aorta using deep learning for reproducible quantification of aortic flow, peak velocity, and dimensions.MethodsA total of 1018 subjects with aortic 4D-flow MRI (528 with bicuspid aortic valve, 376 with tricuspid aortic valve and aortic dilation, 114 healthy controls) comprised the data set. A convolutional neural network was trained to generate 3D aortic segmentations from 4D-flow data. Manual segmentations served as the ground truth (N = 499 training, N = 101 validation, N = 418 testing). Dice scores, Hausdorff distance, and average symmetrical surface distance were calculated to assess performance. Aortic flow, peak velocity, and lumen dimensions were quantified at the ascending, arch, and descending aorta and compared using Bland-Altman analysis. Interobserver variability of manual analysis was assessed on a subset of 40.ResultsConvolutional neural network segmentation required 0.438 ± 0.355 seconds versus 630 ± 254 seconds for manual analysis and demonstrated excellent performance with a median Dice score of 0.951 (0.930-0.966), Hausdorff distance of 2.80 (2.13-4.35), and average symmetrical surface distance of 0.176 (0.119-0.290). Excellent agreement was found for flow, peak velocity, and dimensions with low bias and limits of agreement less than 10% difference versus manual analysis. For aortic volume, limits of agreement were moderate within 16.3%. Interobserver variability (median Dice score: 0.950; Hausdorff distance: 2.45; and average symmetrical surface distance: 0.145) and convolutional neural network-based analysis (median Dice score: 0.953-0.959; Hausdorff distance: 2.24-2.91; and average symmetrical surface distance: 0.145-1.98 to observers) demonstrated similar reproducibility.ConclusionsDeep learning enabled fast and automated 3D aortic segmentation from 4D-flow MRI, demonstrating its potential for efficient clinical workflows. Future studies should investigate its utility for other vasculature and multivendor applications.
| S-EPMC7724647 | biostudies-literature