Wei Feng

Forest fires pose a serious threat to forest ecosystems, biodiversity, carbon sequestration, and sustainable socioeconomic development. Timely and accurate detection is essential for effective early warning and emergency response. However, existing deep learning-based methods often face challenges in balancing detection accuracy and computational efficiency under UAV-assisted forestry monitoring scenarios. This study proposes YOLO-MR, a compact and efficient forest fire detection framework based on YOLOv8n for UAV-assisted intelligent forestry monitoring. The proposed method incorporates a multi-scale deep dilated spatial pyramid fast pooling module, GhostConv, an improved ReC2f module, and an additional P2 detection layer, as well as PIoUv2 loss, enabling more effective multi-scale feature representation and improved target localization. Evaluation using the public M4SFWD dataset together with the self-constructed UFFD dataset verifies the effectiveness and generalization capability of YOLO-MR. Compared against YOLOv8n, the proposed YOLO-MR delivers increases of 1.3%, 2.1%, 1.3%, and 2.1% for Precision, Recall, mAP50, and mAP50-90, respectively, on the M4SFWD dataset, with corresponding gains of 0.9%, 1.8%, 1.8%, and 2.5% on the UFFD dataset. Meanwhile, the proposed model requires only 2.85 M parameters, which is 5.3% fewer than the baseline, demonstrating improved parameter efficiency. The experimental results demonstrate that YOLO-MR provides an effective trade-off between detection performance and computational efficiency, offering a promising solution for Unmanned Aerial Vehicle (UAV)-assisted forest fire monitoring and early warning. By enabling timely wildfire detection with a compact and efficient architecture, the proposed framework supports sustainable intelligent forestry monitoring, contributes to ecosystem protection and disaster prevention, and promotes sustainable forest management.

The development of spatial transcriptomics(ST) technologies enables the exploration of tissue structure and cellular function within a spatial context. However, mainstream analytical methods primarily focus on the gene expression modality itself and struggle to fully leverage auxiliary information from histological images, which limits precise dissection of complex biological structures. To address this, we propose SpatialModal, a multimodal graph learning framework that effectively integrates gene expression data and histological images from spatial transcriptomics to construct unified spatial representations. SpatialModal effectively captures synergistic interactions between modalities by jointly modeling intra- and inter-modal complementary features while incorporating spatial adjacency information. Furthermore, it employs a dual contrastive learning strategy to enhance the discriminative power of representations, thereby enabling efficient and robust analysis of tissue structures. We validate the effectiveness of this approach on multiple public datasets covering diverse tissue types, species, and resolutions. Experiments demonstrate that SpatialModal exhibits significant advantages in downstream tasks, including spatial domain identification, gene expression reconstruction, and pseudotime inference, and accurately elucidates the hierarchical structure of the mouse cortex as well as the metabolic-immune dynamic boundaries in the breast cancer microenvironment. Additionally, SpatialModal shows exceptional robustness in single-cell resolution and multi-slice integration tasks, providing an efficient and broadly applicable analytical tool for spatial transcriptomics research.

Recent advances in spatially resolved transcriptomics have enabled high-throughput gene expression profiling while preserving spatial context, thereby facilitating the investigation of spatial heterogeneity within tissues. Here, we present SpatialGEO, a spatial geometry-aware deep learning framework designed to decipher tissue organizational structures through dual-encoder feature extraction and geometric graph learning. Experimental results demonstrate the superior accuracy of SpatialGEO in identifying spatial regions across datasets encompassing diverse biological contexts and spatial resolutions. Moreover, SpatialGEO uncovers key mechanisms of immune evasion and regulation within the tumor microenvironment and provides novel biological insights into mouse embryonic development.

Tetracycline (TCs) is widely used in the treatment of human and animal infectious disease. TCs gives rise to a growing threat to the human health and environment protection due to its overuse. Therefore, it is important to remove TCs contaminants from waste effluents. In this work, MoS₂@Fe₃O₄ catalytic material was fabricated by the simple hydrothermal method, which was applied in the photo-Fenton system to degrade TCs. The crystal structure, surface morphology, elemental composition, chemical state, electrochemical properties, and separability of MoS₂@Fe₃O₄ catalytic materials were analyzed by X-ray diffraction (XRD), scanning electron microscope (SEM), conventional and high-resolution transmission electron microscopy (TEM/HRTEM), X-ray photoelectron spectroscopy (XPS), electrochemical impedance spectroscopy (EIS), and vibrating sample magnetometry (VSM). Furthermore, MoS₂@Fe₃O₄ could degrade 98.6% of TCs within 60 min under the optimum reaction conditions (the catalyst dosage of 3 g/L, H₂O₂ concentration of 5 mmol/L, the initial TCs concentration of 50 mg/L, and the initial pH of 5), which was a significant increase compared with pure Fe₃O₄. MoS₂ can accelerate the Fe³⁺/Fe²⁺ cycle through electron transfer from Mo⁴⁺ to Fe³⁺, resulting in the improvement in the degradation efficiency of TCs. The quenching and electron paramagnetic resonance (EPR) results showed that OH^• and photogenic hole h+ was the main active species in the photo-Fenton system. What is more, MoS₂@Fe₃O₄ catalytic materials had remarkable stability and reusability, and can be handily regained via magnetic separation technology in a real scenario.

Heavy metal pollution in agricultural soils poses a serious threat to food security. Therefore, it is crucial to conduct risk assessments and issue early warnings about high levels of metal contamination for the sustained prosperity of agriculture. To assess the risks, identify the sources, quantify the amounts, and determine the extent of pollution from seven heavy metals, as well as to provide early warnings, 78 soil samples were collected from farmed lands in the Songnen Plain of Jilin Province. The average concentrations of Zn, Cu, Mn, Pb, Cd, Ni, and As were found to be 2.05, 1.5, 0.2, 1.09, 2.68, 1.53, and 1.17 times higher than the background values of Chinese soils, respectively. Source analysis indicated that toxic Pb originates from vehicle exhaust emission, while Cd, Cu, and Ni are attributed to industrial activities. Zn and As are likely associated with agricultural practices, and Mn predominantly stems from natural environmental sources. The geo-accumulation index suggests relatively high, accumulation levels for Zn, Cu, Mn, and Pb. Meanwhile, the single-factor pollution index indicates elevated pollution levels of Zn, Cu, and Cd. Potential ecological risk assessment reveals that certain areas within Changchun and Baicheng cities exhibit higher ecological risks. Notably, Cd has the highest potential ecological risk index (RI) of the seven heavy metals and warrants special attention. By analyzing annual pollutant accumulations, predictions can be made about the heavy metal content in four cities within the Songnen Plain, enabling the issuance of early warnings regarding soil heavy metal risks. The findings suggest that without proactive measures to mitigate heavy metal accumulation in soils, Changchun and Songyuan will reach severe pollution levels by 2031 and 2029, respectively.

A novel hierarchical porous biomorphic ZnO/SnO was facilely synthesized in one step using bagasse as a bio-template. The structural features of the ZnO/SnO₂ n–n heterostructures were characterized by X-ray diffraction (XRD), scanning electron microscopy (SEM), transmission electron microscopy (TEM), and X-ray photoelectron spectroscopy (XPS). The results revealed that the as-prepared ZnO/SnO₂ retained the original pore morphology of the bagasse material, and the ZnO/SnO₂ was demonstrated with higher sensing performance as compared with the pure SnO₂. Particularly, when the molar ratio of SnO₂:ZnO = 1:1, the sensor displayed the highest response, showing an excellent response value of 37 under 100 ppm methanol at 340 °C. Meanwhile, the ZnO/SnO₂ composite exhibited good gas selectivity and stability to methanol, which could mainly be attributed to the formation of n-n junctions between SnO₂ and ZnO and the high capability of absorbed oxygen species of the ZnO/SnO₂ composite.

Hydrotalcite-like materials (OSA-LDH) were prepared used oil shale ash (OSA), which came from a thermal power plant area, as the main raw material. The characterization results of X-ray diffraction (XRD), scanning electron microscope (SEM), transmission electron microscope (TEM), and thermogravimetric-differential scanning calorimetry (TG-DSC) showed that the prepared samples were mesoporous materials in a crystal state and were layered and contained lattice oxygen and a large number of surface hydroxyl groups. The adsorption property of the prepared samples was confirmed and evaluated by adsorption experiments with Pb²⁺ as the target pollutant. The adsorption process was in accord with the Langmuir isothermal adsorption equation, and the adsorption data fitted perfectly with the pseudo-second kinetic equation. The saturated adsorption capacity for Pb²⁺ was 120.92 mg·g⁻¹ at a temperature of 298 K and initial concentration of 300 mg·L⁻¹. The main adsorption mechanisms of OSA-LDH for Pb²⁺ were chemical bond cooperation and electrostatic bond cooperation. This paper aimed to not only prepare an economical and effective adsorbent to remove heavy metal ions from the solution but also provide a new path for the treatment and utilization of OSA so as to realize efficient waste resource utilization.

A novel hierarchical porous biomorphic ZnO/SnO has been facilely synthesized in one step using bagasse as bio-template. The structural features of the ZnO/SnO2 n–n heterostructures were characterized by X-ray diffraction (XRD), scanning electron microscope (SEM), transmission electron microscope (TEM), and X-ray photoelectron spectroscopy (XPS). The results revealed that the as-prepared ZnO/SnO2 retained the original pore morphology of the bagasse material, and the ZnO/SnO2 was demonstrated with higher sensing performance as compared with the pure SnO2. Particularly, when molar ratio SnO2:ZnO=1:1, the sensor displayed the highest response, showing an excellent response value of 37 under 100 ppm methanol at 340℃. Meanwhile, ZnO/SnO2 composite exhibit good gas selectivity and stability to methanol, which can mainly be attributed to the formation of n-n junctions between SnO2 and ZnO, high capability of absorbed oxygen species of the ZnO/SnO2 composite.

The recyclable paper based on photochromic materials not only reduces the pollution in the paper manufacture process, but also reduces the pollution caused by the use of ink, which receives wide attention. In this paper, a series of phosphomolybdic acid–phosphotungstic acid/ZnO/polyvinylpyrrolidone (PMoA-PWA/ZnO/PVP) hybrid films, which had different ratio of PMoA/PWA, was prepared by the ultrasonic composite method. The results indicated that the hybrid film prepared when the ratio of PMoA to PWA was 3 had the best photochromic performance. In this system, ZnO was the photosensitizer, while PMoA/PWA was the chromophore. The photochromic mechanism of the PMoA-PWA/ZnO/PVP hybrid film was based on the photogenerated electron transfer mechanism. ZnO generated photoelectron under the excitation of visible light, then PMoA and PWA obtained the photoelectron and produced photoreduction reaction to generate heteropolyblue. The visible light photochromic paper was prepared by loaded PMoA-PWA/ZnO/PVP hybrid film (A₃) on A4 paper. Application tests showed that the prepared paper had extremely stable, excellent and reversible visible light photochromic properties, whether it was printing patterns or words, and could replace ordinary paper to realize the reuse of paper.

A visible-light photochromic hybrid film was synthesized based on combining phosphomolybdic acid (PMoA) with the polythiophene (PTh) matrix. The microstructure and photochromic properties of the materials were analyzed through atomic force microscopy (AFM), Fourier transform infrared spectroscopy (FTIR), X-ray photoelectron spectroscopy (XPS) and ultraviolet-visible spectra (UV-vis). According to FTIR spectra, the geometries of PMoA and PTh were well preserved in hybrid film and there exists a strong interaction at the interface of PMoA and PTh. The XPS spectra revealed the change in the chemical microenvironment and the reduction of Mo⁶⁺ atoms in the photoreduction reaction. Under visible light irradiation, the composite film changed from transparent to blue and deepened gradually, generating heteropoly blue. The hybrid film also shows reversibility in the presence of oxygen. The results indicated that the photochromic reaction was inconsistent with photoinduced electron transfer mechanism.