| 1. |
- Ravikumar, Sadhana, et al.
(författare)
-
Improved Segmentation of Deep Sulci in Cortical Gray Matter Using a Deep Learning Framework Incorporating Laplace’s Equation
- 2023
-
Ingår i: Information Processing in Medical Imaging - 28th International Conference, IPMI 2023, Proceedings. - 0302-9743 .- 1611-3349. - 9783031340475 ; 13939 LNCS, s. 692-704
-
Konferensbidrag (refereegranskat)abstract
- When developing tools for automated cortical segmentation, the ability to produce topologically correct segmentations is important in order to compute geometrically valid morphometry measures. In practice, accurate cortical segmentation is challenged by image artifacts and the highly convoluted anatomy of the cortex itself. To address this, we propose a novel deep learning-based cortical segmentation method in which prior knowledge about the geometry of the cortex is incorporated into the network during the training process. We design a loss function which uses the theory of Laplace’s equation applied to the cortex to locally penalize unresolved boundaries between tightly folded sulci. Using an ex vivo MRI dataset of human medial temporal lobe specimens, we demonstrate that our approach outperforms baseline segmentation networks, both quantitatively and qualitatively.
|
|
| 2. |
- de Flores, Robin, et al.
(författare)
-
Characterization of hippocampal subfields using ex vivo MRI and histology data : Lessons for in vivo segmentation
- 2020
-
Ingår i: Hippocampus. - : Wiley. - 1050-9631 .- 1098-1063. ; 30:6, s. 545-564
-
Tidskriftsartikel (refereegranskat)abstract
- Hippocampal subfield segmentation on in vivo MRI is of great interest for cognition, aging, and disease research. Extant subfield segmentation protocols have been based on neuroanatomical references, but these references often give limited information on anatomical variability. Moreover, there is generally a mismatch between the orientation of the histological sections and the often anisotropic coronal sections on in vivo MRI. To address these issues, we provide a detailed description of hippocampal anatomy using a postmortem dataset containing nine specimens of subjects with and without dementia, which underwent a 9.4 T MRI and histological processing. Postmortem MRI matched the typical orientation of in vivo images and segmentations were generated in MRI space, based on the registered annotated histological sections. We focus on the following topics: the order of appearance of subfields, the location of subfields relative to macroanatomical features, the location of subfields in the uncus and tail and the composition of the dark band, a hypointense layer visible in T2-weighted MRI. Our main findings are that: (a) there is a consistent order of appearance of subfields in the hippocampal head, (b) the composition of subfields is not consistent in the anterior uncus, but more consistent in the posterior uncus, (c) the dark band consists only of the CA-stratum lacunosum moleculare, not the strata moleculare of the dentate gyrus, (d) the subiculum/CA1 border is located at the middle of the width of the hippocampus in the body in coronal plane, but moves in a medial direction from anterior to posterior, and (e) the variable location and composition of subfields in the hippocampal tail can be brought back to a body-like appearance when reslicing the MRI scan following the curvature of the tail. Our findings and this publicly available dataset will hopefully improve anatomical accuracy of future hippocampal subfield segmentation protocols.
|
|
| 3. |
- Hallström, Elinor, et al.
(författare)
-
Dietary environmental impacts relative to planetary boundaries for six environmental indicators – A population-based study
- 2022
-
Ingår i: Journal of Cleaner Production. - : Elsevier Ltd. - 0959-6526 .- 1879-1786. ; 373
-
Tidskriftsartikel (refereegranskat)abstract
- The environmental impact of Swedish diets was assessed for six indicators (greenhouse gas [GHG] emissions, cropland use, nitrogen application, phosphorus application, consumptive water use and extinction rate), using self-reported food intake within two population-based cohorts of men and women, 56–96 years of age. The dietary environmental impact was assessed in relation to per capita planetary boundaries, overall and by population subgroups, addressing the relative importance of specific foods and food groups. The total average dietary impact exceeded the planetary boundaries by 1.6 to 4-fold for five of the six environmental indicators; consumptive water use did not exceed the boundaries. Comparing the highest with lowest quintiles of the population impact showed >2.5-fold differences across all environmental indicators. Of the diet's total average environmental impact, animal-based, plant-based and discretionary foods accounted for 28–83%, 8–40% and 9–37%, respectively, across the six indicators. Animal-based foods dominated the impact on GHG emissions, cropland use and nitrogen and phosphorus application, while plant-based and discretionary foods contributed more to consumptive water use and extinction rate. Environmental impact was driven predominantly by consumption of red meat, dairy, fresh fruit and coffee. The findings show major challenges in affluent countries that have to be addressed to achieving sustainable food production systems and diets. They provide guidance on critical food groups, environmental indicators and population subgroups to prioritize in future efforts to reduce the environmental impact. © 2022 The Authors
|
|
| 4. |
- Roh, Hyung S., et al.
(författare)
-
Integrating Color Deconvolution Thresholding and Weakly Supervised Learning for Automated Segmentation of Neurofibrillary Tangle and Neuropil Threads
- 2023
-
Ingår i: Medical Imaging 2023 : Digital and Computational Pathology - Digital and Computational Pathology. - 1605-7422. - 9781510660472 ; 12471
-
Konferensbidrag (refereegranskat)abstract
- Abnormally phosphorylated tau proteins are known to be a major indicator of Alzheimer's Disease (AD) with strong association with memory loss and cognitive decline. Automated generation of pixel-wise accurate neurofibrillary tangles (NFTs) and neuropil threads (NTs) segmentation is a challenging task, due to lack of ground truth segmentation data of these abnormal tau pathology. This problem is most prominent in the case of segmenting NTs, where the small threadlike morphology makes pixel-wise labeling a laborious task and unrealistic for large-scale studies. Lack of ground truth data poses a significant limitation for many learning-based methods to generate accurate segmentations of NFTs and NTs. This work presents an automated pipeline for pixel level segmentation of NFTs and NTs that does not rely on ground truth segmentation data. The pipeline is composed of four main steps: (1) color deconvolution is used to separate histopathology images into staining channels (DAB, Hematoxylin, and Eosin), (2) Otsu's thresholding is used on the DAB stain channel to generate pixel level segmentation of abnormal tau proteins staining, (3) a weakly-supervised learning paradigm (WildCat), using only global descriptors of images, is used to generate density maps of potential regions of NFTs and NTs, and (4) density maps and segmentations are then integrated using connected component analysis to localize NFTs and NTs in the detected tau segmentations. Our results show high global classification accuracy for NFTs (Acc:0.96) and NTs (Acc:0.91), and statistically significant distinctions when evaluating the percent area occupied of the detected NTs relative to expert ratings of NTs severity. Qualitative assessment of the NFTs and NTs results showed accurate pixel-level segmentations of the NFTs, while modest performance for NTs.
|
|
| 5. |
- Xie, Long, et al.
(författare)
-
Deep Label Fusion : A 3D End-To-End Hybrid Multi-atlas Segmentation and Deep Learning Pipeline
- 2021
-
Ingår i: Information Processing in Medical Imaging - 27th International Conference, IPMI 2021, Proceedings. - Cham : Springer International Publishing. - 1611-3349 .- 0302-9743. - 9783030781903 ; 12729 LNCS, s. 428-439
-
Konferensbidrag (refereegranskat)abstract
- Deep learning (DL) is the state-of-the-art methodology in various medical image segmentation tasks. However, it requires relatively large amounts of manually labeled training data, which may be infeasible to generate in some applications. In addition, DL methods have relatively poor generalizability to out-of-sample data. Multi-atlas segmentation (MAS), on the other hand, has promising performance using limited amounts of training data and good generalizability. A hybrid method that integrates the high accuracy of DL and good generalizability of MAS is highly desired and could play an important role in segmentation problems where manually labeled data is hard to generate. Most of the prior work focuses on improving single components of MAS using DL rather than directly optimizing the final segmentation accuracy via an end-to-end pipeline. Only one study explored this idea in binary segmentation of 2D images, but it remains unknown whether it generalizes well to multi-class 3D segmentation problems. In this study, we propose a 3D end-to-end hybrid pipeline, named deep label fusion (DLF), that takes advantage of the strengths of MAS and DL. Experimental results demonstrate that DLF yields significant improvements over conventional label fusion methods and U-Net, a direct DL approach, in the context of segmenting medial temporal lobe subregions using 3T T1-weighted and T2-weighted MRI. Further, when applied to an unseen similar dataset acquired in 7T, DLF maintains its superior performance, which demonstrates its good generalizability.
|
|
| 6. |
- Xie, Long, et al.
(författare)
-
Deep label fusion : A generalizable hybrid multi-atlas and deep convolutional neural network for medical image segmentation
- 2023
-
Ingår i: Medical Image Analysis. - : Elsevier BV. - 1361-8415. ; 83
-
Tidskriftsartikel (refereegranskat)abstract
- Deep convolutional neural networks (DCNN) achieve very high accuracy in segmenting various anatomical structures in medical images but often suffer from relatively poor generalizability. Multi-atlas segmentation (MAS), while less accurate than DCNN in many applications, tends to generalize well to unseen datasets with different characteristics from the training dataset. Several groups have attempted to integrate the power of DCNN to learn complex data representations and the robustness of MAS to changes in image characteristics. However, these studies primarily focused on replacing individual components of MAS with DCNN models and reported marginal improvements in accuracy. In this study we describe and evaluate a 3D end-to-end hybrid MAS and DCNN segmentation pipeline, called Deep Label Fusion (DLF). The DLF pipeline consists of two main components with learnable weights, including a weighted voting subnet that mimics the MAS algorithm and a fine-tuning subnet that corrects residual segmentation errors to improve final segmentation accuracy. We evaluate DLF on five datasets that represent a diversity of anatomical structures (medial temporal lobe subregions and lumbar vertebrae) and imaging modalities (multi-modality, multi-field-strength MRI and Computational Tomography). These experiments show that DLF achieves comparable segmentation accuracy to nnU-Net (Isensee et al., 2020), the state-of-the-art DCNN pipeline, when evaluated on a dataset with similar characteristics to the training datasets, while outperforming nnU-Net on tasks that involve generalization to datasets with different characteristics (different MRI field strength or different patient population). DLF is also shown to consistently improve upon conventional MAS methods. In addition, a modality augmentation strategy tailored for multimodal imaging is proposed and demonstrated to be beneficial in improving the segmentation accuracy of learning-based methods, including DLF and DCNN, in missing data scenarios in test time as well as increasing the interpretability of the contribution of each individual modality.
|
|
