A novel transfer learning approach for the classification of histological images of colorectal cancer

Colorectal cancer (CRC) is the second most diagnosed cancer in the United States. It is identified by histopathological evaluations of microscopic images of the cancerous region, relying on a subjective interpretation. The Colorectal Histology dataset used in this study contains 5000 images, made available by the University Medical Center Mannheim. This approach proposes the automatic identification of eight types of tissues found in CRC histopathological evaluation. We apply Transfer Learning from architectures of Convolutional Neural Networks (CNNs). We modify the structures of CNNs to extract features from the images and input them to well-known machine learning methods: Naive Bayes, Multilayer Perceptron, k-Nearest Neighbors, Random Forest, and Support Vector Machine (SVM). We evaluated 108 extractor–classifier combinations. The one that achieved the best results is DenseNet169 with SVM (RBF), reaching an Accuracy of 92.083% and F1-Score of 92.117%. Therefore, our approach is capable of distinguishing tissues found in CRC histopathological evaluation.

     

Heat stress affects the expression of key genes in the placenta,placental characteristics,and efficiency of Saanen goats and the survival and growth of their kids

Heat stress is detrimental during gestation; however, the effects of heat stress on goat placental characteristics and kid survival remain unclear. The objective of this study was to evaluate the effects of heat stress at final gestation on cortisol concentration, placenta characteristics, and the expression of genes related to placenta. Forty-six primiparous and multiparous Saanen goats were subjected to control (CT; under a thermoneutral environment: air temperature between 12°C and 25°C and the relative humidity from 45 to 73%, n = 23) or heat stress (HS; under a climatic chamber: air temperature at 37°C and the relative humidity at 60 to 70% from 0800 to 1600 h, n = 23) from the last 60 d of pregnancy until the first colostrum suckling. The heat challenge imposed on HS goats during the prepartum period increased their rectal temperature, respiratory frequency, and cortisol levels in plasma and amniotic fluid versus CT goats. In the placenta, HS treatment also increased the expression of the HSPA1A gene. Heat-stressed goats also showed significantly lower expression of HSD11B2 and greater expression of MC2R and NR3C1 than CT goats, suggesting that heat stress decreased the effectiveness by which the HSD11B2 enzyme converts cortisol to cortisone and increased placental responsiveness to cortisol. The HS goats took longer to release the placenta with lighter placental cotyledons, and HS goats had a lower ratio between the kid's weight at birth and placenta weight than CT goats. There was no treatment effect on the kids' survival or weights at birth, but the kids from goats subjected to HS presented lesser cortisol concentration and greater mortality rates at weaning than kids from CT goats. Finally, the overexpression of HSPA1A by HS goats suggests a protective response of placenta. However, the heat stress negatively affected the placenta's expulsion length, placental cotyledons number, weight and area, the ratio between kid's weight and placenta weight, and cortisol signaling. Indeed, the upregulation of MC2R and NR3C1 and downregulation of HSD11B2 on placenta caused by heat stress were associated with greater cortisol concentrations in the amniotic fluid of HS goats. Although HS and CT kids had adequate weights and survival rate during the first weeks of life, the heat stress increased the mortality at weaning of HS kids versus CT kids, suggesting that the heat stress effect persists and can change the ability of kids to respond to weaning challenge.     

Simulation of membrane chemical degradation in a proton exchange membrane fuel cell by computational fluid dynamics

Membrane chemical degradation is a major contributor to the still limited lifetime of proton exchange membrane (PEM) fuel cells. In the present work, this phenomenon is simulated by computational fluid dynamics (CFD). The main advantage of the CFD model is that it can provide the degradation profile across the cell active area. Results reveal that degradation accelerates when voltage, temperature and pressure are increased and when reactants humidity and membrane thickness are decreased. Moreover, membrane deterioration is found to be more severe where oxygen pressure is higher, and more heterogeneous when oxygen distribution is less uniform. Generally, conditions that increase current production and thus oxygen depletion along the cell increase degradation heterogeneity. The flow field design is also found to influence the membrane degradation spatial profile. The modeling strategy here applied, the incorporation of a degradation sub-model into a general-purpose CFD code, can be used to include other degradation mechanisms.     

Room temperature flash of single crystal titania: Electronic and optical properties

A single-crystal specimen of rutile (titania) was flashed repetitively, while increasing the electric field after each cycle. As expected, the flash onset temperature continued to drop modestly at higher fields. However, when the field was increased from 400 to 450 V cm^–1, the flashed onset fell dramatically down to room temperature. We have investigated the electrical and optical properties of this room temperature flashed specimen (called SZ). The specimen was electronically conducting. Optical absorption spectroscopy revealed a narrow band of new energy levels that were generated just below the conduction band. The gap between the conduction band and this flash-induced energy level agreed with the peak in the electroluminescence spectrum. Optical second harmonic generation (SHG) is reported. The flash-on condition significantly lowered the SHG, which rebounded when the flash was turned off. This result suggests that the structure becomes more centrosymmetric in the state of flash, which may represent a disordered state of defects. The possibility of studying flash behavior at room temperature, without a furnace (as in SZ type specimens), opens a considerable simplification for in-situ characterization of flash behavior. For example, a possible relationship between memristor physics and the flash phenomenon can be studied.     

Photosynthetic Cyanobacteria can Clearly Induce Efficient Muscle Tissue Regeneration of Bioprinted Cell-Constructs

Tissue engineering strategies using cell-laden constructs have shown promising results in the treatment of various types of damaged tissues. However, inadequate oxygen delivery to the macroscale 3D cell-constructs for regenerating skeletal muscle tissue has remained a multiplex issue owing to the pivotal factors including cell metabolism and several regulatory intercellular pathways that eventually influence various cellular activities and determines cell phenotype. To overcome this issue, a photosynthetic cyanobacterium (Synechococcus elongatus) is employed in a methacrylated gelatin bioink. Furthermore, to effectively induce cell alignment in the bioink, in situ electric field stimulation is used in a bioprinting system to fabricate cell-laden scaffolds for regenerating skeletal muscle tissue. Owing to the synergistic effects of the bioactive microenvironment that rescues cells from hypoxic conditions and activations of voltage-gated ion channels, highly aligned, multi-nucleated myofibers are obtained as well as significant upregulation (7–10-fold) of myogenic-related genes compared with conventionally prepared cell-constructs. In addition, in vivo studies using a mouse volumetric muscle loss model demonstrate considerable restoration of muscle functionality and regeneration.     

Degradation Pathways of Cobalt-Free LiNiO2 Cathode in Lithium Batteries

Electrode-electrolyte reactivity (EER) and particle cracking (PC) are considered two main causes of capacity fade in high-nickel layered oxide cathodes in lithium-based batteries. However, whether EER or PC is more critical remains debatable. Herein, the fundamental correlation between EER and PC is systematically investigated with LiNiO₂ (LNO), the ultimate cobalt-free lithium layered oxide cathode. Specifically, EER is found more critical than secondary particle cracking (SPC) in determining the cycling stability of LNO; EER leads to primary particle cracking, but mitigates SPC due to the inhibition of H2-H3 phase transformation. Two surface degradation pathways are identified for cycled LNO under low and high EERs. A common blocking surface reconstruction layer (SRL) containing electrochemically-inactive Ni₃O₄ spinel and NiO rock-salt phases is formed on LNO in an electrolyte with a high EER; in contrast, an electrochemically-active SRL featuring regions of electron- and lithium-ion-conductive LiNi₂O₄ spinel phase is formed on LNO in an electrolyte with a low EER. These findings unveil the intrinsic degradation pathways of LNO cathode and are foreseen to provide new insights into the development of lithium-based batteries with a minimized EER and a maximized service life.     

Application of Seeded Membrane Distillation Crystallization for Zero Liquid Discharge

Zero liquid discharge (ZLD) is a crucial requirement in industrial operations concerned with water scarcity and environmental protection. This study investigates the feasibility of using membrane distillation combined with crystallization (MDC) as a potential solution for ZLD. High-saline NaCl solutions were used to evaluate the performance of a seeded MDC system, which demonstrated excellent recovery rates for both water (over 95 %) and salt (over 95.5 %). The presence of seeding crystals in the MDC system was found to inhibit membrane wetting. The MDC system achieved an average thermal efficiency of 61.5 %, showcasing its potential for long-term processes. These results indicate that introducing seed crystals to the membrane's feed side enhances the MDC system's viability as a promising solution for ZLD. Additionally, the MDC system offers the prospect of high water and salt recovery, making it a sustainable approach to wastewater treatment and water reclamation.     

Physicochemical and sensory characteristics of commercially available rice-based Doenjang and their correlation to consumer acceptability for Korean young adults

Food Science and Biotechnology - There is an increased demand for rice doenjang by consumers who are allergic to beans as well as for those seeking diverse flavors. The objective of this study was...     

Video frame interpolation for high dynamic range sequences captured with dual-exposure sensors

Video frame interpolation (VFI) enables many important applications such as slow motion playback and frame rate conversion. However, one major challenge in using VFI is accurately handling high dynamic range (HDR) scenes with complex motion. To this end, we explore the possible advantages of dual-exposure sensors that readily provide sharp short and blurry long exposures that are spatially registered and whose ends are temporally aligned. This way, motion blur registers temporally continuous information on the scene motion that, combined with the sharp reference, enables more precise motion sampling within a single camera shot. We demonstrate that this facilitates a more complex motion reconstruction in the VFI task, as well as HDR frame reconstruction that so far has been considered only for the originally captured frames, not in-between interpolated frames. We design a neural network trained in these tasks that clearly outperforms existing solutions. We also propose a metric for scene motion complexity that provides important insights into the performance of VFI methods at test time.