Guideline-driven timing of treatment initiation plays a central role in clinical decision-making and health policy, yet conventional analyses of randomized clinical trials typically estimate only an average guideline effect and may fail to recover the timing-specific impacts that are most relevant for practice. In this work, we address two key challenges in evaluating guideline effects: (1) treatment initiation times vary substantially across individuals under the reference guideline, and (2) counterfactual outcomes for patients who would initiate treatment at a given time under the guideline are unobserved under alternative initiation strategies. We introduce a causal estimand that captures the average guideline effect for individuals who would initiate treatment at a given time under the reference guideline, and we establish its identification conditions. We then propose a weighting estimator that combines nonparametric outcome regression with estimated treatment-initiation densities to recover these timing-specific effects. We investigate the estimator’s theoretical properties and evaluate its finite-sample performance through simulations. Finally, we apply the proposed method to a randomized HIV treatment trial to estimate the heterogeneous effects of early versus guideline-based antiretroviral therapy initiation on tuberculosis risk.
Accurate identification of patients at high risk of in-hospital mortality in intensive care units (ICUs) is vital for enhancing clinical decision-making and improving patient care strategies. As traditional statistical models often fall short in modeling nonlinear and multifactorial clinical variables, this study explores a machine learning (ML) approach to overcome these limitations.We conducted a retrospective study using the MIMIC-IV database, focusing on 1132 adult ICU patients with cardiac arrest identified from index ICU admissions after applying predefined inclusion and exclusion criteria. Numerical features were summarized using mean aggregation over the first 48 h of ICU admission to provide a compact and clinically interpretable early-risk representation, while categorical attributes underwent structured encoding. The dataset was split into 70% for training and 30% for testing. We applied a combination of regularization techniques (LASSO, Ridge, ElasticNet) and Random Forest-based importance ranking for feature selection.Multiple supervised ML algorithms, including Logistic Regression, LightGBM, CatBoost, XGBoost, and a feed-forward Neural Network, were benchmarked using metrics such as AUC-ROC, calibration plots, and decision curve analysis.The XGBoost algorithm achieved the most favorable results with a test AUC of 0.804 (95% CI: 0.760–0.849) and accuracy of 0.762. Key predictors identified by the final model included neurological status measures, lactate, sodium, anion gap, and coagulation-related indices. These findings suggest that the proposed model offers a reliable, interpretable, and potentially deployable framework for ICU mortality risk prediction.Unlike prior MIMIC-based mortality studies that primarily focused on discrimination, our framework additionally emphasizes calibration, decision-analytic utility, and model interpretability for early ICU risk stratification. Rather than replacing traditional ICU severity scores, the proposed framework is intended to augment them through integrated risk modeling and interpretable prediction.
Quantum spin liquids can arise from Kitaev magnetic interactions, and exhibit fractionalized excitations with the potential for a topological form of quantum computation. This review surveys recent experimental and theoretical progress on the pursuit of phenomena related to Kitaev magnetism in layered and exfoliatable materials, which offer numerous opportunities to apply powerful techniques from the field of atomically thin materials. We primarily focus on the antiferromagnetic Mott insulatorα-RuCl3, which exhibits Kitaev couplings and is readily exfoliated to single- or few-layer sheets, and thus serves as a test bed for developing probes of Kitaev phenomena in atomically thin materials and devices. We introduce the Kitaev model and how it is realized inα-RuCl3and other material candidates; and coverα-RuCl3synthesis and fabrication into van der Waals heterostructure devices. A key discovery is a work-function-mediated charge transfer that heavily dopes both theα-RuCl3and proximate materials, and can enhance Kitaev interactions by up to 50%. We further discuss a wide range of recent results in electronic transport and optical and tunneling spectroscopies ofα-RuCl3devices. The experimental techniques and theoretical insights developed forα-RuCl3establish a framework for discovering and engineering superior two-dimensional Kitaev materials that may ultimately realize elusive quantum spin liquid phases.
Aliphatic ligands are often sidelined in the design of framework materials because their conformational flexibility can contribute to problems such as difficult crystallization, low porosity, and stability. Attempts to boost porosity by ligand elongation usually worsen these problems. Here we propose an expanded bioisosteric replacement (eBIS) concept capable of both scaling up and rigidifying aliphatic ligands. We demonstrate one example realized via linking two cyclohexyl rings in series, which restricts ligand flexibility through intramolecular non-covalent interactions providing an alternative to the π-conjugation-based rigidity. The resulting ligand displays consistent rigidity across multiple MOF platforms. On the pacs platform, it can realize extreme pore geometry with the highest hexagonal c/a ratio and new metal-cluster chemistry such as the first synthesis of nickel-titanium oxocluster. It can boost the BET surface area to as high as 2810 m2 g-1, likely the highest among aliphatic-dicarboxylate MOFs. Furthermore, it leads to possibly largest C2H6/C2H4 uptake differences (88 cm3 g-1, uptake ratio of 1.83, 273 K) among rigid MOFs, a desired property for C2H6-selective separation, which is confirmed by breakthrough experiments. The remarkably low adsorption enthalpies for C2H6 (14.7 kJ mol-1) and C2H4 (15.1 kJ mol-1) enables low-energy adsorbent regeneration benefitting practical separation.
This study presents an experimental investigation of the bearing performance and deformation behavior of tire cell and tire cell-geogrid reinforced subgrades under static loading. Three physical models were tested: an unreinforced subgrade, a tire cell reinforced subgrade, and a tire cell-geogrid reinforced subgrade. Instrumentation, including pressure cells and strain gauges, was carefully calibrated to ensure accurate measurement of vertical stresses and reinforcement strains. The experiments quantified the effects of reinforcement configuration on surface settlement, vertical stress distribution within the reinforced zone. Results show that tire cells substantially reduced surface settlement and redistributed vertical stresses beneath the loading plate, while the addition of a geogrid layer further enhanced stress spreading and structural stiffness. Geogrid tensile strains peaked near the load center and gradually decreased outward, indicating the tensile force of reinforcement mobilization. Overall, the tire cell-geogrid system provided a more uniform stress field and higher load-bearing stiffness compared with the unreinforced or tire cell-only models. The observed synergistic interaction between tire cells and the geogrid highlights the potential of tire cell-geogrid reinforcement approach for sustainable subgrade stabilization in transportation engineering.
