Counterfactual Bidirectional Co-Attention Transformer for Integrative Histology-Genomic Cancer Risk Stratification.

Applying deep learning to predict patient prognostic survival outcomes using histological whole-slide images (WSIs) and genomic data is challenging due to the morphological and transcriptomic heterogeneity present in the tumor microenvironment. Existing deep learning-enabled methods often exhibit learning biases, primarily because the genomic knowledge used to guide directional feature extraction from WSIs may be irrelevant or incomplete. This results in a suboptimal and sometimes myopic understanding of the overall pathological landscape, potentially overlooking crucial histological insights.

Rethinking Copy-Paste for Consistency Learning in Medical Image Segmentation.

Semi-supervised learning based on consistency learning offers significant promise for enhancing medical image segmentation. Current approaches use copy-paste as an effective data perturbation technique to facilitate weak-to-strong consistency learning. However, these techniques often lead to a decrease in the accuracy of synthetic labels corresponding to the synthetic data and introduce excessive perturbations to the distribution of the training data.

Deep Learning-Enabled Integration of Histology and Transcriptomics for Tissue Spatial Profile Analysis.

Spatially resolved transcriptomics enable comprehensive measurement of gene expression at subcellular resolution while preserving the spatial context of the tissue microenvironment. While deep learning has shown promise in analyzing SCST datasets, most efforts have focused on sequence data and spatial localization, with limited emphasis on leveraging rich histopathological insights from staining images. We introduce GIST, a deep learning-enabled gene expression and histology integration for spatial cellular profiling.

Efficient Light-Driven Ion Pumping for Deep Desalination via the Vertical Gradient Protonation of Covalent Organic Framework Membranes.

Traditional desalination methods face criticism due to high energy requirements and inadequate trace ion removal, whereas natural light-driven ion pumps offer superior efficiency. Current synthetic systems are constrained by short exciton lifetimes, which limit their ability to generate sufficient electric fields for effective ion pumping. We introduce an innovative approach utilizing covalent-organic framework membranes that enhance light absorption and reduce charge recombination through vertical gradient protonation of imine linkages during acid-catalyzed liquid-liquid interfacial polymerization.

Survival Prediction After Transarterial Chemoembolization for Hepatocellular Carcinoma: a Deep Multitask Survival Analysis Approach.

The accurate prediction of postoperative survival time of patients with Barcelona Clinic Liver Cancer (BCLC) stage B hepatocellular carcinoma (HCC) is important for postoperative health care. Survival analysis is a common method used to predict the occurrence time of events of interest in the medical field. At present, the mainstream survival analysis models, such as the Cox proportional risk model, should make strict assumptions about the potential random process to solve the censored data, thus potentially limiting their application in clinical practice.

Achieving High Selectivity in Photocatalytic Oxidation of Toluene on Amorphous BiOCl Nanosheets Coupled with TiO.

The inert C()-H bond and easy overoxidation of toluene make the selective oxidation of toluene to benzaldehyde a great challenge. Herein, we present that a photocatalyst, constructed with a small amount (1 mol %) of amorphous BiOCl nanosheets assembled on TiO (denoted as 0.01BOC/TiO), shows excellent performance in toluene oxidation to benzaldehyde, with 85% selectivity at 10% conversion, and the benzaldehyde formation rate is up to 1.

Heat-Treatment-Assisted Molten-Salt Strategy to Enhance Electrochemical Performances of Li-Rich Assembled Microspheres by Tailoring Their Surface Features.

Constructing Li-rich Mn-based layered oxide (LMRO) assembled microspheres with fast kinetics and a stable surface will significantly improve discharge capacity and cyclic stability. In this work, a heat-treatment-assisted (HA) molten-salt (MS) strategy has been designed to prepare LMRO assembled microspheres HA-MS-LMRO (LMRO with heat-treatment-assisted molten-salt process). Electrochemical measurements demonstrate that HA-MS-LMRO possesses superior performance as a cathode for lithium-ion batteries.

Elemental contents in serum of pregnant women with gestational diabetes mellitus.

Diabetes mellitus is characterized by hyperglycemia and is closely related to trace elements. Quite a few pregnant women suffer from impaired glucose tolerance (IGT) or gestational diabetes mellitus (GDM). Investigation of the changes of elemental contents in serum of the pregnant women with IGT and GDM is significant in the etiological research and cure of the diseases.