Biology subjects

Pushparajah, K.

Publications and source records attributed to Pushparajah, K..

6 recordsLinked to original sources

bioRxiv2023

BOUNTI: Brain vOlumetry and aUtomated parcellatioN for 3D feTal MRI

Fetal MRI is widely used for quantitative brain volumetry studies. However, currently, there is a lack of universally accepted protocols for fetal brain parcellation and segmentation. Published clinical studies tend to use different segmentation approaches that also reportedly require significant amounts of time-consuming manual refinement. In this work, we propose to address this challenge by developing a new robust deep learning-based fetal brain segmentation pipeline for 3D T2w motion corrected brain images. At first, we defined a new refined brain tissue parcellation protocol with 19 regions-of-interest using the new fetal brain MRI atlas from the Developing Human Connectome Project. This protocol design was based on evidence from histological brain atlases, clear visibility of the structures in individual subject 3D T2w images and the clinical relevance to quantitative studies. It was then used as a basis for developing an automated deep learning brain tissue parcellation pipeline trained on 360 fetal MRI datasets with different acquisition parameters using semi-supervised approach with manually refined labels propagated from the atlas. The pipeline demonstrated robust performance for different acquisition protocols and GA ranges. Analysis of tissue volumetry for 390 normal participants (21-38 weeks gestational age range), scanned with three different acquisition protocols, did not reveal significant differences for major structures in the growth charts. Only minor errors were present in < 15% of cases thus significantly reducing the need for manual refinement. In addition, quantitative comparison between 65 fetuses with ventriculomegaly and 60 normal control cases were in agreement with the findings reported in our earlier work based on manual segmentations. These preliminary results support the feasibility of the proposed atlas-based deep learning approach for large-scale volumetric analysis. The created fetal brain volumetry centiles and a docker with the proposed pipeline are publicly available online at https://hub.docker.com/r/fetalsvrtk/segmentation (tag brain bounti tissue).

neuroscience↗

bioRxiv2022

3D MRI atlases of congenital aortic arch anomalies and normal fetal heart: application to automated multi-label segmentation

Background3D image-domain reconstruction of black blood contrast T2w SSTSE fetal MRI datasets using slice-to-volume registration methods showed to provide high-resolution 3D images of the heart with superior visualisation of fetal aortic arch anomalies [1]. However, there is a lack of formalisation of the MRI appearance of fetal cardiovascular anatomy and standardisation of vessel segmentation protocols. MethodsIn this work, we present the first set of 3D fetal MRI atlases defining normal and abnormal fetal aortic arch anatomy created from 3D reconstructed images from 87 subjects scanned between 29-34 weeks of gestation with postnatally confirmed outcomes. We also implement and evaluate atlas-guided registration and deep learning (UNETR) methods for automated 3D multi-label fetal heart vessel segmentation. ResultsWe created four atlases representing the average anatomy of the normal fetal heart, coarctation of the aorta, right aortic arch and suspected double aortic arch. Inspection of atlases confirmed the expected pronounced differences in the anatomy of the aortic arch. The results of the multi-label heart vessel UNETR segmentation showed 100% per-vessel detection rate for both normal and abnormal aortic arch anatomy. ConclusionsThis work introduces the first set of 3D black blood T2w MRI atlases of the normal and abnormal fetal cardiovascular anatomy along with detailed segmentation of the major cardiovascular structures. We also demonstrated the feasibility of using deep learning for multi-label vessel segmentation.

bioinformatics↗

bioRxiv2022

Automated atlas-based multi-label fetal cardiac vesselsegmentation in Congenital Heart Disease

Congenital heart disease (CHD) is the most commonly diagnosed birth defect. T2w black blood MRI provides optimal vessel visualisation, aiding prenatal CHD diagnosis. Common clinical practice involves manual segmentation of fetal heart and vessels for visualisation and reporting purposes. We propose an automated multi-label fetal cardiac vessels deep learning segmentation approach for T2w black blood MRI. Our network is trained using single-label manual segmentations obtained through current clinical practice, combined with a multi-label anatomical atlas with desired multi-label segmentation protocol. Our framework combines deep learning label propagation with 3D residual U-Net segmentation to produce high-quality multi-label output well adapted to the individual subject anatomy. We train and evaluate the network using forty fetal subjects with suspected coarctation of the aorta, achieving a dice score of 0.79 {+/-} 0.02 for the fetal cardiac vessels region. The proposed network outperforms the label propagation and achieves a statistically equivalent performance to a 3D residual U-Net trained exclusively on manual single-label data (p-value>0.05). This multi-label framework therefore represents an advancement over the single-label approach, providing label-specific anatomical information, particularly useful for assessing specific anomaly areas in CHD.

bioengineering↗

bioRxiv2021

Automated 3D reconstruction of the fetal thorax in the standard atlas space from motion-corrupted MRI stacks for 21-36 weeks GA range

Slice-to-volume registration (SVR) methods allow reconstruction of high-resolution 3D images from multiple motion-corrupted stacks. SVR-based pipelines have been increasingly used for motion correction for T2-weighted fetal MRI since they allow more informed and detailed diagnosis of brain and body anomalies including congenital heart defects (Lloyd et al., 2019). Recently, fully automated rigid SVR reconstruction of the fetal brain in the atlas space was achieved in (Salehi et al., 2019) that used convolutional neural networks (CNNs) for segmentation and pose estimation. However, these CNN-based methods have not yet been applied to the fetal trunk region. Meanwhile, the existing rigid and deformable SVR (DSVR) solutions (Uus et al., 2020) for the fetal trunk region are limited by the requirement of manual input as well the narrow capture range of the classical gradient descent based registration methods that cannot resolve severe fetal motion frequently occurring at the early gestational age (GA). Furthermore, in our experience, the conventional 2D slice-wise CNN-based brain masking solutions are reportedly prone to errors that require manual corrections when applied on a wide range of acquisition protocols or abnormal cases in clinical setting. In this work, we propose a fully automated pipeline for reconstruction of the fetal thorax region for 21-36 weeks GA range T2-weighted MRI datasets. It includes 3D CNN-based intra-uterine localisation of the fetal trunk and landmark-guided pose estimation steps that allow automated DSVR reconstruction in the standard radiological space irrespective of the fetal trunk position or the regional stack coverage. The additional step for generation of the common template space and rejection of outliers provides the means for automated exclusion of stacks affected by low image quality or extreme motion. The pipeline was evaluated on a series of experiments including fetal MRI datasets and simulated rotation motion. Furthermore, we performed a qualitative assessment of the image reconstruction quality in terms of the definition of vascular structures on 100 early (median 23.14 weeks) and late (median 31.79 weeks) GA group MRI datasets covering 21 to 36 weeks GA range.

bioinformatics↗

bioRxiv2021

3D UNet with GAN discriminator for robust localisation of the fetal brain and trunk in MRI with partial coverage of the fetal body

In fetal MRI, automated localisation of the fetal brain or trunk is a prerequisite for motion correction methods. However, the existing CNN-based solutions are prone to errors and may require manual editing. In this work, we propose to combine a multi-label 3D UNet with a GAN discriminator for localisation of both fetal brain and trunk in fetal MRI stacks. The proposed method is robust for datasets with both full and partial coverage of the fetal body.

bioengineering↗

bioRxiv2019

Fetal whole-heart 4D flow cine MRI using multiple non-coplanar balanced SSFP stacks

PurposeTo develop an MRI framework for reconstruction of 4D velocity vector blood flow volumes for visualisation and quantification of circulation in the fetal heart and major vessels.\n\nMethodsA novel method of velocity-encoding using multiple non-coplanar stacks of bSSFP phase images was combined with a previous framework for reconstruction of motion-corrected 4D magnitude cine volumes to generate spatiotemporally-paired 4D flow cine volumes of the fetal circulatory system. The multiple stack velocity-encoding scheme was validated in a simulated flow phantom and compared with a gold-standard method for velocity-encoding in a physical flow phantom. The 4D flow cine framework was evaluated in seven fetal subjects. Reconstructed 4D flow volumes were evaluated by an expert fetal cardiologist and preliminary flow measurements were taken in various major vessels of the heart.\n\nResultsPhantom experiments showed that the multiple non-coplanar stack velocity-encoding scheme was accurate. The 4D flow cine reconstruction framework was robust in fetal subjects and generated multi-dimensional velocity vector maps of blood flow through the cardiac cycle. Directionality of blood flow was consistent with expected fetal circulatory hemodynamics. Relative blood flow rates in the major vessels were in line with previous observations, although absolute values were underestimated by a factor of approximately two due to limitations of spatial and temporal resolution.\n\nConclusion4D flow cine volumes can be reconstructed from multiple non-coplanar stacks of slices. The proposed framework was used to visualise and quantify flow through the whole fetal heart and great vessels, but is applicable to any imaging scenario where motion is a major challenge.

bioengineering↗