Biology subjects

Namburete, A. I.

Publications and source records attributed to Namburete, A. I..

4 recordsLinked to original sources

bioRxiv2024

UniFed: A unified deep learning framework for segmentation of partially labelled, distributed neuroimaging data

It is essential to be able to combine datasets across imaging centres to represent the breadth of biological variability present in clinical populations. This, however, leads to two challenges: an increase in non-biological variance due to scanner differences, known as the harmonisation problem, and, data privacy concerns due to the inherently personal nature of medical images. Federated learning has been proposed to train deep learning models on distributed data; however, the majority of approaches assume fully labelled data at each participating site, which is unlikely to exist due to the time and skill required to produce manual segmentation labels. Further, they assume all of the sites are available for training. Thus, we introduce UniFed, a unified federated harmonisation framework that enables three key processes to be completed: 1) the training of a federated harmonisation network, 2) the selection of the most appropriate pretrained model for a new unseen site, and 3) the incorporation of a new site into the harmonised federation. We show that when working with partially labelled distributed datasets, UniFedproduces high-quality segmentations and enable all sites to benefit from the knowledge of the federation. The code is available at https://github.com/nkdinsdale/UniFed.

neuroscience↗

bioRxiv2021

STAMP: Simultaneous Training and Model Pruning for Low Data Regimes in Medical Image Segmentation

Acquisition of high quality manual annotations is vital for the development of segmentation algorithms. However, to create them we require a substantial amount of expert time and knowledge. Large numbers of labels are required to train convolutional neural networks due to the vast number of parameters that must be learned in the optimisation process. Here, we develop the STAMP algorithm to allow the simultaneous training and pruning of a UNet architecture for medical image segmentation with targeted channelwise dropout to make the network robust to the pruning. We demonstrate the technique across segmentation tasks and imaging modalities. It is then shown that, through online pruning, we are able to train networks to have much higher performance than the equivalent standard UNet models while reducing their size by more than 85% in terms of parameters. This has the potential to allow networks to be directly trained on datasets where very low numbers of labels are available.

neuroscience↗

bioRxiv2021

Subcortical Segmentation of the Fetal Brain in 3D Ultrasound using Deep Learning

The quantification of subcortical volume development from 3D fetal ultrasound can provide important diagnostic information during pregnancy monitoring. However, manual segmentation of subcortical structures in ultra-sound volumes is time-consuming and challenging due to low soft tissue contrast, speckle and shadowing artifacts. For this reason, we developed a convolutional neural network (CNN) for the automated segmentation of the choroid plexus (CP), lateral posterior ventricle horns (LPVH), cavum septum pellucidum et vergae (CSPV), and cerebellum (CB) from 3D ultrasound. As ground-truth labels are scarce and expensive to obtain, we applied few-shot learning, in which only a small number of manual annotations (n = 9) are used to train a CNN. We compared training a CNN with only a few individually annotated volumes versus many weakly labelled volumes obtained from atlas-based segmentations. This showed that segmentation performance close to intra-observer variability can be obtained with only a handful of manual annotations. Finally, the trained models were applied to a large number (n = 278) of ultrasound image volumes of a diverse, healthy population, obtaining novel US-specific growth curves of the respective structures during the second trimester of gestation.

neuroscience↗

bioRxiv2020

Deep Learning-Based Unlearning of Dataset Bias for MRI Harmonisation and Confound Removal

Increasingly large MRI neuroimaging datasets are becoming available, including many highly multi-site multi-scanner datasets. Combining the data from the different scanners is vital for increased statistical power; however, this leads to an increase in variance due to nonbiological factors such as the differences in acquisition protocols and hardware, which can mask signals of interest. We propose a deep learning based training scheme, inspired by domain adaptation techniques, which uses an iterative update approach to aim to create scanner-invariant features while simultaneously maintaining performance on the main task of interest, thus reducing the influence of scanner on network predictions. We demonstrate the framework for regression, classification and segmentation tasks with two different network architectures. We show that not only can the framework harmonise many-site datasets but it can also adapt to many data scenarios, including biased datasets and limited training labels. Finally, we show that the framework can be extended for the removal of other known confounds in addition to scanner. The overall framework is therefore flexible and should be applicable to a wide range of neuroimaging studies. 1. HighlightsO_LIWe demonstrate a flexible deep-learning-based harmonisation framework C_LIO_LIApplied to age prediction and segmentation tasks in a range of datasets C_LIO_LIScanner information is removed, maintaining performance and improving generalisability C_LIO_LIThe framework can be used with any feedforward network architecture C_LIO_LIIt successfully removes additional confounds and works with varied distributions C_LI

neuroscience↗