Browse Articles

Reduced cloud cover errors in a hybrid AI-climate model through equation discovery and automatic tuning

Abstract Cloud-related parameterizations remain a leading source of uncertainty in climate projections. Although machine learning holds promise for Earth system models (ESMs), many data-driven parameterizations lack interpretability, physical consistency, and smooth integration into ESMs. Here, a two-step method is presented to improve a climate model with data-driven parameterizations. First, we incorporate a physically consistent cloud cover parameterization—derived from storm-resolving simulations via symbolic regression, preserving interpretability while enhancing accuracy—into the ICON global atmospheric model. Second, we apply the gradient-free Nelder–Mead optimizer to automatically recalibrate the hybrid model against Earth observations, tuning in nested stages (2-, 7-, 30- and 365-day runs) to ensure stability and tractability. The tuned hybrid model substantially reduces long-standing biases in cloud cover—particularly over the Southern Ocean (by 75%) and subtropical stratocumulus regions (by 44%)—and remains robust under +4K surface warming. These results demonstrate that interpretable machine-learned parameterizations, paired with practical tuning, can efficiently and transparently strengthen ESM fidelity.

Enhanced production of active species and NH3 using non-equilibrium ferroelectric barrier discharge

Interaction-induced magnetotransport in a 2D Dirac–Heavy hole hybrid band system

Targeting chaperone-mediated autophagy inhibits properties of glioblastoma stem cells and restores anti-tumor immunity

The application of ABC-VED with multi-criteria analysis for drug inventory management

Optical analysis of 3D-printed terahertz waveplates from common thermoplastics

CuFe2O4 nanoparticles via thermal decomposition as recyclable magnetic catalysts for perimidine synthesis

Characterization of broad host range bacteriophages vKpIN31 and vKpIN32 against hospital-acquired Klebsiella pneumoniae in Dakar, Senegal

An explainable hybrid framework for early detection of cardiovascular diseases using Categorical Boosting and Bees algorithm

Abstract Cardiovascular disease (CVD) remains one of the leading causes of death worldwide, claiming millions of lives each year. The early detection of CVD enables healthcare professionals to make informed decisions about the patient’s health. Machine learning (ML)- based frameworks have been extremely popular in predicting diseases. However, results generated from traditional ML models are “black-box,” lacking transparency and interpretability. The objective of the present study is to develop an ML framework that detects CVD with promising accuracy and, further, provide interpretability to the generated outcomes to ensure targeted therapies. The Framingham, Massachusetts CVD dataset, which is publicly available from the Kaggle Repository, is used in this study. As part of the data pre-processing, the Random Oversampling (RO) technique is applied to overcome the data imbalance problem, followed by Pearson Correlation analysis to understand the correlation between attributes. Then, the Min–Max scaling technique is used for data normalization. The pre-processed data is fed into a hybrid ML framework incorporating the Categorical Boosting (CatBoost) and BEEs algorithms to achieve optimized CVD prediction results. The proposed Hybrid model yielded 98.04% accuracy, a Precision of 97.09%, a Recall of 98.96%, an F1-score of 98.02%, and a Specificity of 97.16%, with a total execution time of 26.6580 s. The proposed model outperformed contemporary state-of-the-art algorithms, considering most evaluation metrics. Additionally, Explainable Artificial Intelligence (XAI) techniques, such as LIME and SHAP, are implemented to identify the contribution of the most significant attributes towards the occurrence of CVD, offering valuable insights into the detection of the disease and enabling healthcare providers to make accurate and timely treatment decisions.

Predicting critical crack propagation length in sustainable additive-enhanced concrete using explainable machine learning

Automatic 3D railroad alignment detection using modified Hough transform

Effortless facial expression recognition without motor simulation

Beyond colonoscopy, faecal DNA mutation screening provides a potential and viable path to early colorectal cancer detection

Origin and evolution of fluids and heatflow in geothermal systems of Indus River Basin (IRB), India

A digital twin approach for sustainable construction: predictive optimization of concrete strength using industry 4.0 principles

Bridging classical and neural methods for improved segmentation in mathematical text based images

Impact of cilia length and variable fluid properties on electroosmotic nanofluid flow in an inclined converging microchannel

Abstract This study presents a mathematical model for the transport of non-Newtonian nanofluids in an inclined ciliated converging microchannel. The analysis focuses on the combined effects of cilia length variation, electroosmotic effect, and temperature-dependent viscosity and thermal conductivity. The governing equations were derived using the Debye–Hückel approximation along with lubrication theory. These equations were then solved semi-analytically using the Homotopy Perturbation Method (HPM) in MATHEMATICA. The resulting solutions were visualized by plotting graphs in MATLAB. The results indicate that increased cilia length leads to a reduction in axial velocity but lowers the external pressure required to maintain flow, allowing for precise adjustments to transport dynamics. Variable viscosity and thermal conductivity improve flow and heat transfer under mild obstruction. The applied electric field accelerates the fluid by offsetting the drag caused by cilia, thereby enhancing overall transport efficiency. These findings illustrate the capability of cilia to serve as moderators of flow and transport in applications like targeted drug delivery, lab-on-a-chip diagnostics, microscale heat exchangers, and bio-inspired pumping systems.

Patient satisfaction and disease-related knowledge in otorhinolaryngology outpatient care in Saudi Arabia

A human iPSC-based neural spheroid platform for modelling glioblastoma infiltration using high-content imaging

Abstract Glioblastoma is the most aggressive adult brain tumour, characterised by resistance to therapy and high recurrence due to diffuse infiltration. We developed a physiologically relevant co-culture model, combining patient-derived glioblastoma cell lines with cortical-like neural spheroids differentiated from human induced pluripotent stem cells. Using high-content imaging, we demonstrate that GBM20 and GBM1 cell lines migrate directionally along axons toward neural spheroids in live imaging assays and infiltrate spheroids extensively in endpoint assays, unlike non-cancerous neural stem cells. A proof-of-principle drug screen identified PF 573228 (FAK inhibitor) and motixafortide (CXCR4 inhibitor) as potent suppressors of GBM20 and GBM1 infiltration, respectively. Bulk RNA sequencing revealed gene expression profiles correlating with invasive behaviour and drug sensitivity. This platform offers a valuable model for studying glioblastoma infiltration along axons and provides proof-of-principle that migration can serve as a measurable and actionable phenotype to screen therapeutic vulnerabilities in glioblastoma.

Modelling and economic evaluation of a guar gum–enhanced electrocoagulation process optimized by BBD with sludge characterization for DB183 dye removal