Existing predictive models of opioid use disorder (OUD) may change as the rate of opioid prescribing decreases. Using Veterans Administration’s EHR data, we developed machine-learning predictive models of new OUD diagnoses and ranked the importance of patient features based on their ability to predict a new OUD diagnosis in 2000–2012 and 2013–2021. Using patient characteristics, the three separate machine learning techniques were comparable in predicting OUD, achieving an accuracy of >80%. Using the random forest classifier, opioid prescription features such as early refills and length of prescription consistently ranked among the top five factors that predict new OUD. Younger age was positively associated with new OUD, and older age inversely associated with new OUD. Age stratification revealed prior substance abuse and alcohol dependency as more impactful in predicting OUD for younger patients. There was no significant difference in the set of factors associated with new OUD in 2000–2012 compared to 2013–2021. Characteristics of opioid prescriptions are the most impactful variables that predict new OUD both before and after the peak in opioid prescribing rates. Predictive models should be tailored to age groups. Further research is warranted to determine if machine learning models perform better when tailored to other patient subgroups.

Two-dimensional (2D) Dirac states with linear dispersion have been observed in graphene and on the surface of topological insulators. 2D Dirac states discovered so far are exclusively pinned at high-symmetry points of the Brillouin zone, for example, surface Dirac states at [Formula: see text] in topological insulators Bi2Se(Te)3 and Dirac cones at K and [Formula: see text] points in graphene. The low-energy dispersion of those Dirac states are isotropic due to the constraints of crystal symmetries. In this work, we report the observation of novel 2D Dirac states in antimony atomic layers with phosphorene structure. The Dirac states in the antimony films are located at generic momentum points. This unpinned nature enables versatile ways such as lattice strains to control the locations of the Dirac points in momentum space. In addition, dispersions around the unpinned Dirac points are highly anisotropic due to the reduced symmetry of generic momentum points. The exotic properties of unpinned Dirac states make antimony atomic layers a new type of 2D Dirac semimetals that are distinct from graphene.

Porous carbon (PC) has been widely regarded as one of the most promising absorbents for methane storage. Studies show that its uptake capacity and selectivity highly depend on textural structures. Although much effort has been made, unveiling their detailed structure-performance relationship remains a challenge. Here, we propose an innovative study where, with the assistance of machine learning, the hidden relationship of the textural structures of PC with the methane uptake and separation can be derived from existing data in material literature. Machine learning models were trained by the data, including specific surface area, micropore volume, mesopore volume, temperature, and pressure as the input variables and methane uptake as the output variable for prediction. Among the tested models, the multilayer perceptron (MLP) shows the highest accuracy in predicting the methane uptake. In addition, the model enables to automatically construct a uptake performance map in terms of micropore volume and mesopore volume. The obtained MLP model was also extended to explore the CO2/CH4 selectivity by retraining it with the data collected from literature of PC for the CO2 uptake. The constructed 2D selectivity map shows that the high selectivity can be achieved in the low CH4 uptake region.

The discovery of graphene has stimulated enormous interest in two-dimensional (2D) electron gas with linear band dispersion. However, to date, 2D Dirac semimetals are still very rare due to the fact that 2D Dirac states are generally fragile against perturbations such as spin-orbit coupling. Nonsymmorphic crystal symmetries can enforce the formation of Dirac nodes, providing another route to establishing symmetry-protected Dirac states in 2D materials. Here, we report the symmetry-protected Dirac states in nonsymmorphic alpha-antimonene (Sb monolayer). The antimonene was synthesized by the method of molecular beam epitaxy. 2D Dirac states with large anisotropy were observed by angle-resolved photoemission spectroscopy. The Dirac states in alpha-antimonene are spin-orbit coupled in contrast to the spinless Dirac states in graphene. The result extends the graphene physics into a family of 2D materials where spin-orbit coupling is present.

The discovery of graphene has stimulated enormous interest in two-dimensional (2D) electron gas with linear band structure. 2D Dirac materials possess many intriguing physical properties such as high carrier mobility and zero-energy Landau level thanks to the relativistic dispersion and chiral spin/pseudospin texture. 2D Dirac states discovered so far are exclusively pinned at high-symmetry points of the Brillouin zone, for example, surface Dirac states at Γ in topological insulators Bi_2Se(Te)_3 and Dirac cones at K and K' in graphene. In this work, we report the realization of 2D Dirac states at generic k-points in antimony atomic layers with phosphorene structure (i.e. α-antimonene). The unpinned nature enables versatile ways to control the locations of the Dirac points in momentum space. In addition, dispersions around the unpinned Dirac points exhibit intrinsically anisotropic behaviors due to the reduced symmetry of generic momentum points. These properties make the α-antimonene films a promising platform for exploring interesting physics in unpinned 2D Dirac fermions that are distinct from the conventional Dirac states in graphene.

A 1992 algorithm by Rothblum & Zenios [1], which we will refer to as the RZ algorithm, computes a positive diagonal scaling of an arbitrary matrix so that the products of the nonzero normed elements of each row and column (i.e., zero elements are ignored) are equal to user-specified constants. It has recently been recognized that in the case in which the constants are all unity, a unique scaling always exists and can be used as a canonical scaling for performing scale-consistent or scale-invariant operations on matrices [5]. For example, the algorithm can be applied to define a generalized matrix inverse that is consistent with respect to diagonal scalings (e.g., due to changes of units on variables), as opposed to the Moore-Penrose pseudoinverse which provides consistency with respect to unitary transformations (e.g., rotations) [5]. Scale consistency is critical to applications in which performance is expected to be invariant with respect to the choice of state variables, e.g., whether lengths are represented in centimeters, meters, or kilometers in robotics applications [8, 9, 10, 6] or unknown units specific to each individual’s ratings of products in the context of recommender systems. In this paper we generalize the canonical positive scaling of rows and columns of a matrix to the scaling of selected-rank subtensors of an arbitrary tensor. We expect our results and framework will prove useful for sparse-tensor completion required for generalizations of the recommender system problem beyond a matrix of user-product ratings to multidimensional arrays involving coordinates based both on user attributes (e.g., age, gender, geographical location, etc.) and product/item attributes (e.g., price, size, weight, etc.).

In this paper we generalize the canonical positive scaling of rows and columns of a matrix to the scaling of selected-rank subtensors of an arbitrary tensor. We expect our results and framework will prove useful for sparse-tensor completion required for generalizations of the recommender system problem beyond a matrix of user-product ratings to multidimensional arrays involving coordinates based both on user attributes (e.g., age, gender, geographical location, etc.) and product/item attributes (e.g., price, size, weight, etc.).