Background and objective The volume of the intracerebral hemorrhage (ICH) obtained from CT scans is essential for quantification and treatment planning. However,a fast and accurate volume acquisition brings great challenges. On the one hand, it is both time consuming and operator dependent for manual segmentation, which is the gold standard for volume estimation. On the other hand, low contrast to normal tissues, irregular shapes and distributions of the hemorrhage make the existing automatic segmentation methods hard to achieve satisfactory performance. Method To solve above problems, a CNN-based architecture is proposed in this work, consisting of a novel model, which is named as Ψ-Net and a multi-level training strategy. In the structure of Ψ-Net, a self-attention block and a contextual-attention block is designed to suppresses the irrelevant information and segment border areas of the hemorrhage more finely. Further, an multi-level training strategy is put forward to facilitate the training process. By adtime for training and achives superior performance than previous ICH segmentaion methods.Background and objective The manual measurement of arterial diameter and wall thickness using imaging modalities demand expertise, and the state-of-art automated or semi-automated measurement features are seldom available in the entry-level systems. The advanced ultrasound modalities are expensive, non-scalable, and less favorable for field and resource-constrained settings. In this work, we present a novel method to measure arterial diameter (D), surrogate intima-media thickness (sIMT), and with them their intra-cardiac cycle changes by employing an affordable image-free ultrasound technology. Methods The functionality of the method was systematically validated on a simulation testbed, phantoms and, 40 human subjects. The accuracy, agreement, inter-beat, and inter-operator variabilities were quantified. The in-vivo measurement performance of the method was compared against two reference B-mode tools – Carotid Studio and CAROLAB. Results Simulations revealed that for the A-mode frames with SNR > 10 dB, the proposed method identifies the desired arterial wall interfaces with an RMSE less then 20 μm. The RMSE for the diameter and wall thickness measurements from the static phantom were 111 μm and 14 μm, and for the dynamic phantom were 117 μm and 18 μm, respectively. Strong agreement was seen between the in-vivo measurements of the proposed method and the two reference tools. The mean absolute errors against the two references and the inter-beat variability were smaller than 0.18 mm for D and smaller than 36 μm for sIMT measurements. Likewise, the respective inter-observer variabilities were 0.16 ± 0.23 mm and 43 ± 25 μm. Conclusion Acceptable accuracy and repeatability were observed during the validation, that were on a par with the recently reported B-mode techniques in the literature. The technology being real-time, automated, and relatively inexpensive, is promising for field and low-resource settings.During organogenesis groups of differentiating cells self-organize into a series of structural intermediates with defined architectural forms. Evidence is emerging that such architectural forms are important in guiding cell fate, yet in vitro methods to guide cell fate have focused primarily on un-patterned exposure of stems cells to developmentally relevant chemical cues. We set out to ask if organizing differentiating lung progenitors into developmentally relevant structures could be used to influence differentiation status. Specifically, we use elastomeric substrates to guide self-assembly of human pluripotent stem cell-derived lung progenitors into developmentally-relevant sized tubes and assess the impact on differentiation. Culture in 100 μm tubes reduced the percentage of SOX2+SOX9+ cells and reduced proximal fate potential compared to culture in 400 μm tubes or on flat surfaces. Cells in 100 μm tubes curved to conform to the tube surface and experienced increased cellular tension and reduced elongation. Pharmacologic disruption of tension through inhibition of ROCK, myosin II activity and actin polymerization in tubes resulted in maintenance of SOX2+SOX9+ populations. Furthermore, this effect required canonical WNT signaling. This data suggests that structural forms, when developmentally relevant, can drive fate choice during directed differentiation via a tension-based canonical WNT dependent mechanism.Environmental DNA (eDNA) can exist in water with various sizes and states. Among them, relative to extra-cellular DNA, intra-cellular DNA such as cell and tissue fragments can mainly be detected at larger size fractions, and may be protected from enzymatic DNA degradation processes. Here, we verified the hypothesis that the selective collection of such large-sized eDNA enhances the efficiency of capturing less-degraded eDNA, based on a tank experiment using Japanese Jack Mackerel (Trachurus japonicus) as a model species. 1400W We concentrated different volumes of rearing water using the filters with different pore sizes (0.7 μm and 2.7 μm), and quantified the copy number of short and long mitochondrial and short nuclear DNA fragments of target species in water samples. As a result, the ratio of long to short eDNA concentrations was higher in the larger pore size filter, which would support our stated hypothesis. In addition, the ratio of nuclear to mitochondrial eDNA was lower in the larger pore size filter. These results imply a difference in the persistence of nuclear and mitochondrial DNA between intra- and extra-cellular environments. Moreover, larger filter pore size did not necessarily decrease the yields of eDNA, and there was little difference in yields in smaller filtration volumes. The findings of this study indicate the potential to select information from eDNA signals by focusing on eDNA of specific size and state, which may contribute to efficient utilization of the information on species taxonomy and physiology in water samples.Global food systems contribute to climate change, the transgression of planetary boundaries and deforestation. An improved understanding of the environmental impacts of different food system futures is crucial for forging strategies to sustainably nourish a growing world population. We here quantify the greenhouse gas (GHG) emissions of global food system scenarios within a biophysically feasible “option space” in 2050 comprising all scenarios in which biomass supply – calculated as function of agricultural area and yields – is sufficient to cover biomass demand – derived from human diets and the feed demand of livestock. We assessed the biophysical feasibility of 520 scenarios in a hypothetical no-deforestation world. For all feasible scenarios, we calculate (in) direct GHG emissions related to agriculture. We also include (possibly negative) GHG emissions from land-use change, including changes in soil organic carbon (SOC) and carbon sinks from vegetation regrowth on land spared from food production. We identify 313 of 520 scenarios as feasible.