Hippocampus Subfield Segmentation Using a Patch-Based Boosted Ensemble of Autocontext Neural Networks

Manjón, José V.; Coupé, Pierrick

doi:10.1007/978-3-319-67434-6_4

Cited by 5 publications

(2 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Beyond their value for fundamental neuroscientific inquiry, as carried out in the current study, this approach may also be beneficial when translating findings between animal models and patient groups in a preclinical/clinical context. Methods to automatize hippocampal subfield segmentation 30,56,57 and to enhance cross-modal and inter-individual registrations 30 are already well established in humans, and current efforts to adapt them to non-human primate brains will facilitate that process. In this context, open data sharing projects such as the BigMac dataset 32 used in this study, but also initiatives such as PRIME-DE 58 may make an invaluable contribution, as they will allow for the aggregation of a diverse set of data in non-human primates, and their dissemination to a wide range of researchers.…”

Section: Discussionmentioning

confidence: 99%

Hippocampal connectivity patterns echo macroscale cortical evolution in the primate brain

Eichert,

DeKraker,

Howard

et al. 2023

Preprint

View full text Add to dashboard Cite

The hippocampus is involved in numerous cognitive functions, some of which have uniquely human aspects, such as autobiographical memory. Hippocampal anatomy, however, is typically considered conserved across primates and its evolutionary diversification is rarely studied. Comparing hippocampal structure and function is, therefore, critical for understanding human brain architecture. Here, we developed a novel comparative framework to study the hippocampus across species characterising its geometry, microstructure, and functional network embedding. In humans and macaques, we generated a new comparative space that represents the hippocampus as an unfolded surface, which respects its sheet-like anatomy. We mapped histological and MRI-derived markers of microstructure to the hippocampal surface and integrated it with low-dimensional embedding of resting-state MRI connectivity data. Our results demonstrate that the micro– and macro-structural organisation of the hippocampus are overall conserved in both species, showing consistent anterior-posterior and subfield-to-subfield differentiation. Furthermore, while hippocampal functional organisation also follows anterior-posterior trends in both species, hippocampal functional connectivity markedly reflected evolutionary reconfiguration of transmodal networks, in particular the default-mode network. Specifically, the inferior parietal lobe in the macaque mirrors an incomplete integration of the default mode network in non-human primates. By combining fine-grained anatomical investigation with large-scale functional imaging, we showed that microstructurally preserved regions like the hippocampus may still undergo functional reconfiguration, due to their embedding in higher-order association networks.SummaryWhile the hippocampus is key for uniquely human cognitive abilities, it is also a phylogenetically old cortex and paradoxically considered evolutionarily preserved. Here, we introduce a comparative framework to quantify preservation and reconfiguration of hippocampal organisation in primate evolution, by analysing the hippocampus as an unfolded cortical surface that is geometrically matched across species. Our findings revealed an overall conservation of hippocampal macro– and micro-structure, showing anterior-posterior and, perpendicularly, subfield-related organisational axes in both humans and macaques. However, while functional organisation in both species also followed an anterior-posterior axis, the latter showed a marked evolutionary reconfiguration, which mirrors a rudimentary integration of the default-mode-network in non-human primates. Our findings suggest that microstructurally preserved regions like the hippocampus may still undergo functional reconfiguration in primate evolution, due to their embedding in heteromodal association networks.

show abstract

Section: Discussionmentioning

confidence: 99%

Hippocampal connectivity patterns echo macroscale cortical evolution in the primate brain

Eichert,

DeKraker,

Howard

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…There are three categories of the conventional denoising algorithms: image filtering and post-processing (Chan et al 2010, Peltonen et al 2011, Dutta et al 2013, Arabi and Zaidi 2021, image translation and denoising (Lin et al 2001, Green, 2005, Le Pogam et al 2013, and iterative reconstruction (Riddell et al 2001, Somayajula et al 2010, Cheng et al 2021. Deep-learning approach can learn both the noise distribution and the image prior information which makes it outperform many conventional methods (Wang et al 2021a) in medical image denoising fileds, including CT (Chen et al 2017, Gholizadeh-Ansari et al 2020, MRI (Manjón andCoupe 2018, Tian et al 2022), Ultrasound (Karaoğlu et al 2022, Khor et al 2022, and PET (Zhou et al 2021). Xu et al applied a residual convolutional neural network (CNN) (Zhang et al 2017) to denoise the ultra-low-dose (0.5% counts level) brain PET data (Xu et al 2017).…”

Section: Introductionmentioning

confidence: 99%

A personalized deep learning denoising strategy for low-count PET images

Liu

Mirian

et al. 2022

Phys. Med. Biol.

View full text Add to dashboard Cite

Objective. Deep learning denoising networks are typically trained with images that are representative of the testing data. Due to the large variability of the noise levels in PET images, it is challenging to develop a proper training set for general clinical use. Our work aims to develop a personalized denoising strategy for the low-count PET images at various noise levels. Approach. We first investigated the impact of the noise level in the training images on the model performance. Five 3D U-Net models were trained on five groups of images at different noise levels, and a one-size-fits-all model was trained on images covering a wider range of noise levels. We then developed a personalized weighting method by linearly blending the results from two models trained on 20%-count level images and 60%-count level images to balance the trade-off between noise reduction and spatial blurring. By adjusting the weighting factor, denoising can be conducted in a personalized and task-dependent way. Main results. The evaluation results of the six models showed that models trained on noisier images had better performance in denoising but introduced more spatial blurriness, and the one-size-fits-all model did not generalize well when deployed for testing images with a wide range of noise levels. The personalized denoising results showed that noisier images require higher weights on noise reduction to maximize the structural similarity (SSIM) and mean squared error (MSE). And model trained on 20%-count level images can produce the best liver lesion detectability. Significance. Our study demonstrated that in deep learning-based low dose PET denoising, noise levels in the training input images have a substantial impact on the model performance. The proposed personalized denoising strategy utilized two training sets to overcome the drawbacks introduced by each individual network and provided a series of denoised results for clinical reading.

show abstract

Hippocampus Segmentation Using U-Net Convolutional Network from Brain Magnetic Resonance Imaging (MRI)

et al. 2022

View full text Add to dashboard Cite

Hippocampus Subfield Segmentation Using a Patch-Based Boosted Ensemble of Autocontext Neural Networks

Cited by 5 publications

References 19 publications

Hippocampal connectivity patterns echo macroscale cortical evolution in the primate brain

Hippocampal connectivity patterns echo macroscale cortical evolution in the primate brain

A personalized deep learning denoising strategy for low-count PET images

Hippocampus Segmentation Using U-Net Convolutional Network from Brain Magnetic Resonance Imaging (MRI)

Contact Info

Product

Resources

About