Dataset Information

VP-Nets : Efficient automatic localization of key brain structures in 3D fetal neurosonography.

ABSTRACT: Three-dimensional (3D) fetal neurosonography is used clinically to detect cerebral abnormalities and to assess growth in the developing brain. However, manual identification of key brain structures in 3D ultrasound images requires expertise to perform and even then is tedious. Inspired by how sonographers view and interact with volumes during real-time clinical scanning, we propose an efficient automatic method to simultaneously localize multiple brain structures in 3D fetal neurosonography. The proposed View-based Projection Networks (VP-Nets), uses three view-based Convolutional Neural Networks (CNNs), to simplify 3D localizations by directly predicting 2D projections of the key structures onto three anatomical views. While designed for efficient use of data and GPU memory, the proposed VP-Nets allows for full-resolution 3D prediction. We investigated parameters that influence the performance of VP-Nets, e.g. depth and number of feature channels. Moreover, we demonstrate that the model can pinpoint the structure in 3D space by visualizing the trained VP-Nets, despite only 2D supervision being provided for a single stream during training. For comparison, we implemented two other baseline solutions based on Random Forest and 3D U-Nets. In the reported experiments, VP-Nets consistently outperformed other methods on localization. To test the importance of loss function, two identical models are trained with binary corss-entropy and dice coefficient loss respectively. Our best VP-Net model achieved prediction center deviation: 1.8?±?1.4 mm, size difference: 1.9?±?1.5 mm, and 3D Intersection Over Union (IOU): 63.2?±?14.7% when compared to the ground truth. To make the whole pipeline intervention free, we also implement a skull-stripping tool using 3D CNN, which achieves high segmentation accuracy. As a result, the proposed processing pipeline takes a raw ultrasound brain image as input, and output a skull-stripped image with five detected key brain structures.

SUBMITTER: Huang R

PROVIDER: S-EPMC5988265 | biostudies-literature | 2018 Jul

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

VP-Nets : Efficient automatic localization of key brain structures in 3D fetal neurosonography.

Huang Ruobing R Xie Weidi W Alison Noble J J

Medical image analysis 20180423

Three-dimensional (3D) fetal neurosonography is used clinically to detect cerebral abnormalities and to assess growth in the developing brain. However, manual identification of key brain structures in 3D ultrasound images requires expertise to perform and even then is tedious. Inspired by how sonographers view and interact with volumes during real-time clinical scanning, we propose an efficient automatic method to simultaneously localize multiple brain structures in 3D fetal neurosonography. The ...[more]

PMID: 29715691

Similar Datasets

Project description:PurposeFluoroscopy is the standard imaging modality used to guide hip surgery and is therefore a natural sensor for computer-assisted navigation. In order to efficiently solve the complex registration problems presented during navigation, human-assisted annotations of the intraoperative image are typically required. This manual initialization interferes with the surgical workflow and diminishes any advantages gained from navigation. In this paper, we propose a method for fully automatic registration using anatomical annotations produced by a neural network.MethodsNeural networks are trained to simultaneously segment anatomy and identify landmarks in fluoroscopy. Training data are obtained using a computationally intensive, intraoperatively incompatible, 2D/3D registration of the pelvis and each femur. Ground truth 2D segmentation labels and anatomical landmark locations are established using projected 3D annotations. Intraoperative registration couples a traditional intensity-based strategy with annotations inferred by the network and requires no human assistance.ResultsGround truth segmentation labels and anatomical landmarks were obtained in 366 fluoroscopic images across 6 cadaveric specimens. In a leave-one-subject-out experiment, networks trained on these data obtained mean dice coefficients for left and right hemipelves, left and right femurs of 0.86, 0.87, 0.90, and 0.84, respectively. The mean 2D landmark localization error was 5.0 mm. The pelvis was registered within [Formula: see text] for 86% of the images when using the proposed intraoperative approach with an average runtime of 7 s. In comparison, an intensity-only approach without manual initialization registered the pelvis to [Formula: see text] in 18% of images.ConclusionsWe have created the first accurately annotated, non-synthetic, dataset of hip fluoroscopy. By using these annotations as training data for neural networks, state-of-the-art performance in fluoroscopic segmentation and landmark localization was achieved. Integrating these annotations allows for a robust, fully automatic, and efficient intraoperative registration during fluoroscopic navigation of the hip.

Project description:OBJECTIVE:The aim of this study was to evaluate the validity of a semiautomated volumetric approach (5DCNS+) for the detailed assessment of the fetal brain in a clinical setting. METHODS:Stored 3D volumes of >?1100 consecutive 2nd and 3rd trimester pregnancies (range 15-36 gestational weeks) were analyzed using a workflow-based volumetric approach 5DCNS+, enabling semiautomated reconstruction of diagnostic planes of the fetal central nervous system (CNS). All 3D data sets were examined for plane accuracy, the need for manual adjustment, and fetal-maternal characteristics affecting successful plane reconstruction. We also examined the potential of these standardized views to give additional information on proper gyration and sulci formation with advancing gestation. RESULTS:Based on our data, we were able to show that gestational age with an OR of 1.085 (95% CI 1.041-1.132) and maternal BMI with an OR of 1.022 (95% CI 1.041-1.054) only had a slight impact on the number of manual adjustments needed to reconstruct the complete volume, while maternal age and fetal position during acquisition (p?=?0.260) did not have a significant effect. For the vast majority (958/1019; 94%) of volumes, using 5DCNS+ resulted in proper reconstruction of all nine diagnostic planes. In less than 1% (89/9171 planes) of volumes, the program failed to give sufficient information. 5DCNS+ was able to show the onset and changing appearance of CNS folding in a detailed and timely manner (lateral/parietooccipital sulcus formation seen in <?65% at 16-17 gestational weeks vs. 94.6% at 19 weeks). CONCLUSIONS:The 5DCNS+ method provides a reliable algorithm to produce detailed, 3D volume-based assessments of fetal CNS integrity through a standardized reconstruction of the orthogonal diagnostic planes. The method further gives valid and reproducible information regarding ongoing cortical development retrieved from these volume sets that might aid in earlier in utero recognition of subtle structural CNS anomalies.

Project description:BackgroundSegmentation of medical image volumes is a time-consuming manual task. Automatic tools are often tailored toward specific patient cohorts, and it is unclear how they behave in other clinical settings.PurposeTo evaluate the performance of the open-source Multi-Planar U-Net (MPUnet), the validated Knee Imaging Quantification (KIQ) framework, and a state-of-the-art two-dimensional (2D) U-Net architecture on three clinical cohorts without extensive adaptation of the algorithms.Study typeRetrospective cohort study.SubjectsA total of 253 subjects (146 females, 107 males, ages 57 ± 12 years) from three knee osteoarthritis (OA) studies (Center for Clinical and Basic Research [CCBR], Osteoarthritis Initiative [OAI], and Prevention of OA in Overweight Females [PROOF]) with varying demographics and OA severity (64/37/24/53/2 scans of Kellgren and Lawrence [KL] grades 0-4).Field strength/sequence0.18 T, 1.0 T/1.5 T, and 3 T sagittal three-dimensional fast-spin echo T1w and dual-echo steady-state sequences.AssessmentAll models were fit without tuning to knee magnetic resonance imaging (MRI) scans with manual segmentations from three clinical cohorts. All models were evaluated across KL grades.Statistical testsSegmentation performance differences as measured by Dice coefficients were tested with paired, two-sided Wilcoxon signed-rank statistics with significance threshold α = 0.05.ResultsThe MPUnet performed superior or equal to KIQ and 2D U-Net on all compartments across three cohorts. Mean Dice overlap was significantly higher for MPUnet compared to KIQ and U-Net on CCBR ( 0.83±0.04 vs. 0.81±0.06 and 0.82±0.05 ), significantly higher than KIQ and U-Net OAI ( 0.86±0.03 vs. 0.84±0.04 and 0.85±0.03) , and not significantly different from KIQ while significantly higher than 2D U-Net on PROOF ( 0.78±0.07 vs. 0.77±0.07 , P=0.10 , and 0.73±0.07) . The MPUnet performed significantly better on N=22 KL grade 3 CCBR scans with 0.78±0.06 vs. 0.75±0.08 for KIQ and 0.76±0.06 for 2D U-Net.Data conclusionThe MPUnet matched or exceeded the performance of state-of-the-art knee MRI segmentation models across cohorts of variable sequences and patient demographics. The MPUnet required no manual tuning making it both accurate and easy-to-use.Level of evidence3 TECHNICAL EFFICACY: Stage 2.

Dataset Information

VP-Nets : Efficient automatic localization of key brain structures in 3D fetal neurosonography.

Publications

VP-Nets : Efficient automatic localization of key brain structures in 3D fetal neurosonography.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets