Deep neural networks with powerful auto-optimization tools have been widely applied in various research fields, e.g., NLP and computer vision. However, existing neural network architectures typically are constructed using different inductive biases, e.g., preconceptions, expecting to decrease parameter search space during training, reduce computational cost or introduce expert knowledge in the neural network design. As an alternative, Multilayer Perceptron (MLP) provides much better freedom for exploration, has a lower inductive bias than convolutional neural networks (CNNs), and offers good flexibility in learning complex patterns. Even though, such neural architectures are commonly built in a flat Euclidean space, which is not necessarily the optimal space for any data, and is especially not good for modeling hierarchical correlations. Hyperbolic neural networks (HNNs) have gained attention for their ability to capture hierarchical structures present in complex data types like graphs. Recently, there has been an increasing interest to extend HNNs to computer vision tasks, motivated by the observations that images possess rich hierarchical relations. However, this is generally applied by employing a Euclidean backbone for learning higher-level semantic representations and only incorporating a hyperbolic classifier for classification, which, we argue, does not make full use of the advantage of hyperbolic space. Considering the recovery of the attention-free Multilayer Perceptron(MLP), in this paper, we extend it to non-Euclidean space and propose a novel architecture, named Hyperbolic Res-MLP (HR-MLP), that leverages fully hyperbolic layers to learn feature embeddings and perform image classification in an end-to-end fashion. With the help of the proposed Lorentz cross-patch and cross-channel layers, we can directly perform operations in the hyperbolic domain with fewer parameters, making it faster to train and providing comparatively better performance than its Euclidean counterpart. Experiments on CIFAR10, CIFAR100, and MiniImageNet demonstrate a comparable and superior performance when compared to Euclidean baselines. Our code is available at (https://github.com/Ahmad-Omar-Ahsan/HR-MLP)

Metal plasmonic nanostructures have promising applications in biosensing due to their ability to facilitate light-matter interaction. However, the damping of the metal leads to a wide full width at half maximum (FWHM) spectrum and results which restricts its sensing capabilities. In this research, we present a novel non-full-metal nanostructure, namely indium-tin oxide (ITO)-Au nanodisk arrays consisting of top ITO nanodisk arrays and a bottom gold layer, and compare the coupling mode with the full-metal nanodisk arrays. The FWHM of proposed ITO-Au nanodisk array reduces to one fifth of that all metal nanodisk arrays, which is generated mode-coupling by surface plasmon modes at metal interfaces with magnetic resonance mode. Furthermore, the thickness variation of nanodisks has no impact on the sensing performance of this ITO-based nanostructure, ensuring excellent tolerance during preparation. We fabricated both structures by using template transfer and vacuum deposition techniques and calibrated their sensing performances by detect Immunoglobulin G (IgG) protein molecules. Both theoretical and experimental results demonstrate that ITO-Au nanodisk arrays provide an effective solution for biosensing applications.

In the last decade, the tremendous improvement in the sensitivity and also affordability of Liquid Chromatography-tandem Mass Spectrometry (LC-MS/MS) have revolutionized its application in pharmaceutical analysis, resulting in wide-spread of employing LC-MS/MS for determining pharmaceutical compounds including anticancer drugs in pharmaceutical research and also industries. Currently, LC-MS/MS has been widely used to quantify small molecule oncology drugs in various biological matrices to support preclinical and clinical Pharmacokinetic studies in R & D of oncology drugs. This mini-review article will describe the state-of-the art LC-MS/MS and its application in rapid quantification of small molecule anticancer drugs. In addition, efforts have also been made in this review to address several key aspects in the development of rapid LC-MS/MS methods, including sample preparation, chromatographic separation and matrix effect evaluation.

A sensitive and robust liquid chromatography-tandem mass spectrometric (LC-MS/MS) method was developed and validated for the determination of nimbolide in mouse serum. Exemestane was used as the internal standard (IS). Here, we employed acetonitrile-based protein precipitation (PPT) for serum sample preparation, and performed chromatographic separation using an ODS Hypersil C18 column (100×2.1 mm, 5µm) with gradient elution (0.1% formic acid in water vs 100% acetonitrile). The run time was 6 min. Instrumental analysis was performed by electrospray ionization tandem mass spectrometry (ESI-MS/MS) in the multiple-reaction monitoring (MRM) under positive mode. A good linear calibration was achieved in the 5–1000 ng/ml range. The intra- and inter-day precisions for nimbolide were ≤ 12.6% and ≤ 13.9 % respectively. Intra-day accuracy ranged from 96.9% – 109.3% while inter-day accuracy ranged from 94.3% – 110.2%. The matrix effect of nimbolide, detected but consistent at low and high concentrations, do not affect linearity of standard curve. In conclusion, we have developed and validated a sensitive analytical method for determination of a novel natural compound nimbolide in mouse serum and it has been successfully applied to our preclinical study in investigating the pharmacokinetic properties of nimbolide, which could greatly facilitate the preclinical development of the promising lead compound for anticancer therapy.

Retiring coal power plants can reduce air pollution and health damages. However, the spatial distribution of those impacts remains unclear due to complex power system adjustments and pollution chemistry and transport. Focusing on coal retirements in Pennsylvania (PA), we analyze six counterfactual scenarios for 2019 that differ in retirement targets (e.g., reducing 50% of coal-based installed capacity vs. generation) and priorities (e.g., closing plants with higher cost, closer to Environmental Justice Areas, or with higher CO₂ emissions). Using a power system model of the PJM Interconnection, we find that coal retirements in PA shift power generation across PA and the Rest of PJM region leading to scenario-varying changes in the plant-level release of several air pollutants. Due to air pollution chemistry and transport and socio-demographics, these changes in turn give rise to a reduction of 10 to 182 PM_2.5-attributable deaths in PJM across the six scenarios, with most reductions occurring in PA. Scenarios that reduce more coal power generation yield greater aggregate health benefits due to air quality improvements in PA and adjacent downwind regions. In addition, vulnerable populations—in both PA and Rest of PJM—benefit most in scenarios that prioritize plant closures in PA near Environmental Justice Areas. These results demonstrate the importance of considering cross-regional linkages and socio-demographics in designing equitable retirement strategies.

A novel, rapid and sensitive liquid chromatography-tandem mass spectrometric (LC-MS/MS) method was developed and validated for the evaluation of OTS167 pharmacokinetics in mouse serum. SN-38 was optimized to be selected as the internal standard. Chromatographic separation was performed on a BDS Hypersil C18 column (100 x 2.1 mm, 5 µm) using gradient elution with mobile phase solvent A as water containing 0.1% formic acid and solvent B as acetonitrile containing 0.1% formic acid. The analysis was carried out using multiple reaction monitoring (MRM) with a triple-quadrupole mass spectrometer operated in the positive electrospray ionization mode. Mass transitions of 487.2 > 348.0 and 393.2 > 349.2 were monitored for OTS167 and SN-38 respectively. The standard calibration curve demonstrated high linearity at a range of 5-1000 ng/mL, with a coefficient of determination (r2) ≥ 0.996. The accuracy for OTS167 ranged from 92.3% to 102.2% and the precision was ≤ 12.7%. Recovery was consistent at about 83% to 89%. No significant matrix effect was observed. This method was successfully applied to monitor the pharmacokinetic profiles in mice over 24 h after ingestion of 5 mg/kg of OTS167. Maximum serum concentrations (3197 ng/mL) of OTS167 was observed at 1.67 hour after OTS167 ingestion via garage at 5 mg/kg and area under the curve (AUC) was 28579 h*ng/mL, suggesting that OTS167 can be quickly absorbed after oral administration.