Search

01 februari 2023, Item does not contain fulltext, In this work, we report the set-up and results of the Liver Tumor Segmentation Benchmark (LiTS), which was organized in conjunction with the IEEE International Symposium on Biomedical Imaging (ISBI) 2017 and the International Conferences on Medical Image Computing and Computer-Assisted Intervention (MICCAI) 2017 and 2018. The image dataset is diverse and contains primary and secondary tumors with varied sizes and appearances with various lesion-to-background levels (hyper-/hypo-dense), created in collaboration with seven hospitals and research institutions. Seventy-five submitted liver and liver tumor segmentation algorithms were trained on a set of 131 computed tomography (CT) volumes and were tested on 70 unseen test images acquired from different patients. We found that not a single algorithm performed best for both liver and liver tumors in the three events. The best liver segmentation algorithm achieved a Dice score of 0.963, whereas, for tumor segmentation, the best algorithms achieved Dices scores of 0.674 (ISBI 2017), 0.702 (MICCAI 2017), and 0.739 (MICCAI 2018). Retrospectively, we performed additional analysis on liver tumor detection and revealed that not all top-performing segmentation algorithms worked well for tumor detection. The best liver tumor detection method achieved a lesion-wise recall of 0.458 (ISBI 2017), 0.515 (MICCAI 2017), and 0.554 (MICCAI 2018), indicating the need for further research. LiTS remains an active benchmark and resource for research, e.g., contributing the liver-related segmentation tasks in http://medicaldecathlon.com/. In addition, both data and online evaluation are accessible via https://competitions.codalab.org/competitions/17094.

Artificial intelligence (AI) is proliferating and developing faster than any domain scientist can adapt. To support the scientific enterprise in the Helmholtz association, a network of AI specialists has been set up to disseminate AI expertise as an infrastructure among domain scientists. As this effort exposes an evolutionary step in science organization in Germany, this article aspires to describe our setup, goals, and motivations. We comment on past experiences, current developments, and future ideas as we bring our expertise as an infrastructure closer to scientists across our organization. We hope that this offers a brief yet insightful view of our activities as well as inspiration for other science organizations.

We present our contribution to the HECKTOR 2021 challenge. We created a Survival Random Forest model based on clinical features, and a few radiomics features that have been extracted with and without using the given tumor masks, for Task 3 and Task 2 of the challenge, respectively. To decide on which radiomics features to include into the model, we proceeded both to automatic feature selection, using several established methods, and to literature review of radiomics approaches for similar tasks. Our best performing model includes one feature selected from the literature (Metabolic Tumor Volume derived from the FDG-PET image), one via stability selection (Inverse Variance of the Gray Level Co-occurrence Matrix of the CT image), and one selected via permutation-based feature importance (Tumor Sphericity). This hybrid approach turns-out to be more robust to overfitting than models based on automatic feature selection. We also show that simple ROI definition for the radiomics features, derived by thresholding the Standard Uptake Value in the FDG-PET images, outperforms the given expert tumor delineation in our case. Team name on the AIcrowd platform: ia-h-ai.

We present our contribution to the HECKTOR 2021 challenge. We created a Survival Random Forest model based on clinical features, and a few radiomics features that have been extracted with and without using the given tumor masks, for Task 3 and Task 2 of the challenge, respectively. To decide on which radiomics features to include into the model, we proceeded both to automatic feature selection, using several established methods, and to literature review of radiomics approaches for similar tasks. Our best performing model includes one feature selected from the literature (Metabolic Tumor Volume derived from the FDG-PET image), one via stability selection (Inverse Variance of the Gray Level Co-occurrence Matrix of the CT image), and one selected via permutation-based feature importance (Tumor Sphericity). This hybrid approach turns-out to be more robust to overfitting than models based on automatic feature selection. We also show that simple ROI definition for the radiomics features, derived by thresholding the Standard Uptake Value in the FDG-PET images, outperforms the given expert tumor delineation in our case. Team name on the AIcrowd platform: ia-h-ai.

We present our contribution to the HECKTOR 2021 challenge. We created a Survival Random Forest model based on clinical features, and a few radiomics features that have been extracted with and without using the given tumor masks, for Task 3 and Task 2 of the challenge, respectively. To decide on which radiomics features to include into the model, we proceeded both to automatic feature selection, using several established methods, and to literature review of radiomics approaches for similar tasks. Our best performing model includes one feature selected from the literature (Metabolic Tumor Volume derived from the FDG-PET image), one via stability selection (Inverse Variance of the Gray Level Co-occurrence Matrix of the CT image), and one selected via permutation-based feature importance (Tumor Sphericity). This hybrid approach turns-out to be more robust to overfitting than models based on automatic feature selection. We also show that simple ROI definition for the radiomics features, derived by thresholding the Standard Uptake Value in the FDG-PET images, outperforms the given expert tumor delineation in our case. Team name on the AIcrowd platform: ia-h-ai.

Auch wenn dem Denken von Marx heute ein erneutes philosophisches Interesse gilt, bleibt oft unklar, worin dieses genau besteht. Unter dem Eindruck der Feuerbachthesen, in denen Marx dazu auffordert, die «Welt» zu «verändern», wurde die genuin philosophische Dimension seines Denkens oft zugunsten Lesart vernachlässigt, die an der Umwälzung der ökonomischen und politischen Verhältnisse orientiert ist. Erst im 20. Jahrhundert wurde die ursprüngliche Verbindung zwischen Marx und der Philosophie wiederhergestellt. Die Zweideutigkeit der Beziehung von Marx zur Philosophie und der Philosophie zu Marx zeigt sich in der Vielfalt der Beiträge in diesem Band. Diese befassen sich sowohl mit grundsätzlichen Fragen der Marx’schen Philosophie als auch mit den philosophischen Diskussionen über den Marxismus. Si Marx fait aujourd’hui l’objet d’un regain d’intérêt pour la philosophie «officielle», cet intérêt n’en est pas moins pétri d’ambiguïtés. Affirmant lui-même dans ses thèses sur Feuerbach qu’il fallait dépasser la philosophie, la dimension proprement philosophique de ses recherches a longtemps été négligé au profit d’une lecture très économiciste. Il a fallu attendre le 20ème siècle pour que se renoue un lien original entre Marx et la philosophie L’ambiguïté des relations qu’entretiennent Marx à la philosophie et la philosophie à Marx est tangible dans la diversité des neuf contributions à ce numéro. Celles-ci abordent en effet, les enjeux propres à la philosophie de Marx, les objets traités philosophiquement par Marx, autant que les discussions philosophiques du marxisme.

We study the quantum phases of bosons with repulsive contact interactions on a two-leg ladder in the presence of a uniform Abelian gauge field. The model realizes many interesting states, including Meissner phases, vortex fluids, vortex lattices, charge density waves, and the biased-ladder phase. Our work focuses on the subset of these states that breaks a discrete symmetry. We use density matrix renormalization group simulations to demonstrate the existence of three vortex-lattice states at different vortex densities and we characterize the phase transitions from these phases into neighboring states. Furthermore, we provide an intuitive explanation of the chiral-current reversal effect that is tied to some of these vortex lattices. We also study a charge-density-wave state that exists at 1/4 particle filling at large interaction strengths and flux values close to half a flux quantum. By changing the system parameters, this state can transition into a completely gapped vortex-lattice Mott-insulating state. We elucidate the stability of these phases against nearest-neighbor interactions on the rungs of the ladder relevant for experimental realizations with a synthetic lattice dimension. A charge-density-wave state at 1/3 particle filling can be stabilized for flux values close to half a flux quantum and for very strong on-site interactions in the presence of strong repulsion on the rungs. Finally, we analytically describe the emergence of these phases in the low-density regime, and, in particular, we obtain the boundaries of the biased-ladder phase, i.e., the phase that features a density imbalance between the legs. We make contact with recent quantum-gas experiments that realized related models and discuss signatures of these quantum states in experimentally accessible observables. © 2016 American Physical Society.

We establish the phase diagram of the strongly interacting Bose-Hubbard model defined on a two-leg ladder geometry in the presence of a homogeneous flux. Our work is motivated by a recent experiment [M. Atala et al., Nat. Phys. 10, 588 (2014) 1745-247310.1038/nphys2998], which studied the same system, in the complementary regime of weak interactions. Based on extensive density matrix renormalization group simulations and a bosonization analysis, we fully explore the parameter space spanned by filling, interleg tunneling, and flux. As a main result, we demonstrate the existence of gapless and gapped Meissner and vortex phases, with the gapped states emerging in Mott-insulating regimes. We calculate experimentally accessible observables such as chiral currents and vortex patterns. © 2015 American Physical Society.

The interplay between spontaneous symmetry breaking in many-body systems, the wavelike nature of quantum particles and lattice effects produces an extraordinary behavior of the chiral current of bosonic particles in the presence of a uniform magnetic flux defined on a two-leg ladder. While noninteracting as well as strongly interacting particles, stirred by the magnetic field, circulate along the system's boundary in the counterclockwise direction in the ground state, interactions stabilize vortex lattices. These states break translational symmetry, which can lead to a reversal of the circulation direction. Our predictions could readily be accessed in quantum gas experiments with existing setups or in arrays of Josephson junctions. © 2015 American Physical Society.

Contains fulltext : 137499.pdf (publisher's version ) (Open Access), BACKGROUND: Congenital disorders of glycosylation are genetic syndromes that result in impaired glycoprotein production. We evaluated patients who had a novel recessive disorder of glycosylation, with a range of clinical manifestations that included hepatopathy, bifid uvula, malignant hyperthermia, hypogonadotropic hypogonadism, growth retardation, hypoglycemia, myopathy, dilated cardiomyopathy, and cardiac arrest. METHODS: Homozygosity mapping followed by whole-exome sequencing was used to identify a mutation in the gene for phosphoglucomutase 1 (PGM1) in two siblings. Sequencing identified additional mutations in 15 other families. Phosphoglucomutase 1 enzyme activity was assayed on cell extracts. Analyses of glycosylation efficiency and quantitative studies of sugar metabolites were performed. Galactose supplementation in fibroblast cultures and dietary supplementation in the patients were studied to determine the effect on glycosylation. RESULTS: Phosphoglucomutase 1 enzyme activity was markedly diminished in all patients. Mass spectrometry of transferrin showed a loss of complete N-glycans and the presence of truncated glycans lacking galactose. Fibroblasts supplemented with galactose showed restoration of protein glycosylation and no evidence of glycogen accumulation. Dietary supplementation with galactose in six patients resulted in changes suggestive of clinical improvement. A new screening test showed good discrimination between patients and controls. CONCLUSIONS: Phosphoglucomutase 1 deficiency, previously identified as a glycogenosis, is also a congenital disorder of glycosylation. Supplementation with galactose leads to biochemical improvement in indexes of glycosylation in cells and patients, and supplementation with complex carbohydrates stabilizes blood glucose. A new screening test has been developed but has not yet been validated. (Funded by the Netherlands Organization for Scientific Research and others.).

Sialic acid storage disease (SASD) is an inborn error resulting from defects in the lysosomal membrane protein sialin. The SASD phenotypical spectrum ranges from a severe presentation, infantile sialic acid storage disease (ISSD) which may present as hydrops fetalis, to a relatively mild form, Salla disease. Screening for SASD is performed by determination of free sialic acid (FSA) in urine or amniotic fluid supernatant (AFS). Subsequent diagnosis of SASD is performed by quantification of FSA in cultured fibroblasts and by mutation analysis of the sialin gene, SLC17A5. We describe simple quantitative procedures to determine FSA as well as conjugated sialic acid in AFS, and FSA in cultured fibroblasts, using isotope dilution (13C3-sialic acid) and multiple reaction monitoring LC-ESI-MS/MS. The whole procedure can be performed in 2-4 h. Reference values in AFS were 0-8.2 μmol/L for 15-25 weeks of gestation and 3.2-12.0 μmol/L for 26- 38 weeks of gestation. In AFS samples from five fetuses affected with ISSD FSA was 23.9-58.9 μmol/L demonstrating that this method is able to discriminate ISSD pregnancies from normal ones. The method was also validated for determination of FSA in fibroblast homogenates. FSA in SASD fibroblasts (ISSD; 20-154 nmol/mg protein, intermediate SASD; 12.9-15.1 nmol/mg, Salla disease; 5.9-7.4 nmol/mg) was clearly elevated compared to normal controls (0.3-2.2 nmol/mg). In conclusion, we report simple quantitative procedures to determine FSA in AFS and cultured fibroblasts improving both prenatal diagnostic efficacy for ISSD as well as confirmatory testing in cultured fibroblasts following initial screening in urine or AFS.