Hardness, as a typical mechanical property of dispersion-strengthened tungsten alloy, is influenced by various coupled factors. This paper aims to identify the key factors affecting the hardness of the dispersion-strengthened tungsten alloys with different carbides and oxides as the reinforcement phase in order to enable the high-throughput prediction of hardness. A dataset was established with alloy hardness as the target variable, and the features included the content of reinforcement phase, the Vickers hardness of reinforcement phase, the melting point of the reinforcement phase, the valence electron number of the reinforcement phase, the sintering temperature, the sintering time, pressure, relative density, and grain size. Seven regression models were trained, and we selected random forest, support vector regression, and XGBoost regression machine learning models with better performance to construct a hardness prediction model of the dispersion-strengthened tungsten alloy. SHAP analysis, based on random forests, shows that the content of reinforcement phase, grain size, and relative density have the most significant impact on the hardness. A random forest model is the most suitable machine learning method for predicting the hardness of dispersion-strengthened tungsten alloys in this work. The R² values of the training and test sets are 0.93 and 0.80, and the MAE values of the training and test sets are 22.72 and 38.37. The influence of the most important features on the hardness was also discussed based on the random forest model. This study provides a data-driven approach for the accurate and efficient prediction of the hardness of dispersion-strengthened tungsten alloys, offering an important reference for the design and development of high-performance tungsten alloy materials.

Spatial transcriptomics (ST) has emerged as an essential technology for interpreting the molecular profiles underlying pathological tissue morphology. Most existing ST analyses are limited to 2D sections, which ignore the complex structural and molecular heterogeneity of biological tissues in 3D space and may cause diagnostic oversights. Since acquiring complete 3D ST volumes is resource-intensive, recent 3D imputation paradigms provide a cost-effective alternative by integrating 3D whole-slide images (WSIs) with sparse 2D ST references (e.g., a single slide). Despite this methodological advancement, effectively modeling complex cross-layer spatial dependencies remains challenging. Current mainstream solutions predominantly adopt standard Transformers for cross-scale feature aggregation, which may bring computational overhead and higher overfitting risk while having limited explicit mechanisms for hierarchical anatomical guidance. To address these limitations, we propose a Hierarchical MambaOut-based Spatial Imputation Graph Network (HM-ASIGN) for anatomy-aware 3D spatial transcriptomics imputation. Our architecture leverages MambaOut’s dynamic gated 1D convolutions as a parameter-efficient alternative to dense global self-attention. This design captures the depth-wise evolution of pathological features while reducing over-parameterization. Inspired by the macro-to-micro diagnostic reasoning of clinical pathologists, HM-ASIGN introduces a multi-scale recursive guidance mechanism. It constructs a top-down information flow by extracting global anatomical priors at macroscopic scales and injecting them as contextual anchors into regional and spot-level features in a cascaded manner. This helps ensure that fine-grained molecular predictions are properly constrained by global morphological structures. Evaluation experiments on multiple public breast cancer datasets demonstrate that HM-ASIGN achieves competitive reference-level performance against existing baselines, reaching a Pearson Correlation Coefficient (PCC) of 0.772. Specifically, when evaluated against the foundational ASIGN framework, it improves predictive accuracy while reducing the total parameter count by approximately 33.3% and improving inference throughput. Our results suggest that HM-ASIGN provides a computationally efficient approach for 3D spatial molecular mapping.