Birla Institute of Technology (Deoghar)

Achieving optimal mechanical performance in wire arc additive manufacturing (WAAM) of high-strength lowalloy (HSLA) steels is critical for the structural integrity of load-bearing and pressure-retaining components. This study compares the effects of high-capacity Spray Arc and voltage-controlled Pulsed Arc gas metal arc welding (GMAW) modes on the microstructure, residual stress distribution, and mechanical behavior of AM70 HSLA steel, an alloy formulated for enhanced arc stability and deoxidation. Comprehensive characterization using optical/electron microscopy, X-ray diffraction, mechanical testing, and fractography reveals distinct process-structure-property relationships. Pulsed Arc mode produces a refined acicular ferrite matrix with more uniform crystallite size and elemental homogeneity, leading to superior hardness (236-246 HV5 vs. 223-235 HV5), tensile strength (743-793 MPa vs. 687-710 MPa), and Charpy impact toughness (78.5-103 J vs. 59.3-78.2 J) compared to Spray Arc. Additionally, compressive residual stresses are more uniformly distributed under Pulsed Arc (-134 to -288 MPa), whereas Spray Arc introduces steeper gradients (-66 to -311 MPa), which could affect long-term structural performance. Although Spray Arc yields higher ductility (34.7-36.4 % vs. 29.1-32.9 %), Pulsed Arc offers a better balance of strength and toughness. Fractographic analysis confirms ductile failure modes in both cases, with finer dimple morphology observed in Pulsed Arc samples. These findings demonstrate the potential of Pulsed Arc WAAM with AM70 steel for manufacturing pressure-resilient and structurally reliable HSLA steel components.

Timely and precise detection of breast cancer is critical for better treatment outcomes. This study intends to improve the effectiveness and interpretability of breast cancer classification by introducing a novel ETCapsNet model and integrating a comprehensive set of Explainable AI algorithms that employ information from multiple imaging modalities. ETCapsNet combines EfficientNetv2 Small for robust feature extraction, a custom Multi-Head Transformer Block to capture crucial long-range spatial dependencies for improved contextual understanding, and a Capsule Network to preserve spatial hierarchies within the multimodal image data effectively. The model was thoroughly evaluated on a comprehensive dataset that included all three imaging modalities. To offer clinical utility and transparency, we developed MammXAI, an interactive XAI dashboard that integrates Grad-CAM, Grad-CAM++, ScoreCAM, SmoothGrad, Integrated Gradient, Occlusion Sensitivity, PDA, and LIME. ETCapsNet performed well and consistently across all three imaging modalities, obtaining an overall accuracy of 0.996 and an average loss of 0.009, revealing superior performance of our proposed hybrid architecture compared to VGG16-19, ResNet50, and novel ConvNeXtTiny. These results highlight the model’s effectiveness for accurate breast cancer detection and its potential applicability to other medical imaging challenges. This research presents a novel multimodal framework for breast cancer classification, significantly enhanced by the XAI-integrated MammXAI dashboard. This work promises to speed up breast image evaluation, reduce diagnostic errors, and improve overall clinical efficiency by providing medical practitioners with interpretable visualizations and explanations of the algorithm’s decision-making process using techniques such as Grad-CAM, Integrated Gradients, and other explainable AI methods.

Antibiotic resistance (AR) in environmental bacteria poses a critical challenge to public health, driven largely by the dissemination of antibiotic resistance genes (ARGs) via mobile genetic elements such as plasmids. In our previously reported work, environmental bacterial isolates were obtained from diverse aquatic sources contaminated by hospital effluents, and their antibiotic susceptibility profiles were determined. Species-level identification of isolates as Aeromonas spp. was achieved through 16S rRNA sequencing, revealing a diverse microbiota harbouring multidrug resistance (MDR) traits. In the present study, plasmids sequence was reconstructed using the plasmidSPAdes assembler. The assembled plasmid sequences were interrogated against the Comprehensive Antibiotic Resistance Database (CARD), enabling the identification of ARGs and associated accessory genes. Comparative analysis of the five reconstructed plasmid sequences was conducted through network analysis, which allowed the segregation of ARGs shared across all plasmids from those uniquely present. Representative ARGs identified as common were selected for experimental validation. PCR amplification and sequencing detected these target ARGs within the plasmids of different isolates. Subsequently, their abundance was quantified using absolute qPCR, providing precise copy number/μL. This integrated approach from plasmid reconstruction to ARG network mapping and quantitative validation offers a robust framework for understanding the distribution and prevalence of plasmid borne ARGs in environmental bacteria. The findings underscore the pivotal role of plasmids in ARG dissemination and highlight the importance of coupling in silico analysis with experimental confirmation for comprehensive resistance surveillance.

This study reports the hydrothermal synthesis of PdX/CDs nanocomposites with varying palladium contents, confirmed by inductively coupled plasma mass spectrometry (ICP-MS) analysis, and their comprehensive physicochemical and electrochemical characterization for ethanol oxidation reaction (EOR) in alkaline media. The carbon dots (CDs), synthesized from D-galactose and L-histidine, serve as a porous, conductive support that enhances Pd nanoparticle dispersion and electronic conductivity. Among the catalysts tested, the Pd2/CDs nanocomposite exhibited superior performance, delivering a high anodic peak current density of 26 mA cm(-2), a low charge transfer resistance of 450 Omega, and a small Tafel slope of 61.97 mV dec(-1). Electrochemical stability was demonstrated through chronoamperometric tests, highlighting the catalyst's durability under alkaline conditions. Enhanced activity is attributed to the increased electrochemically active surface area, suppressed nanoparticle agglomeration, and efficient charge transport facilitated by the CDs. These findings position Pd2/CDs as a promising, cost-effective electrocatalyst alternative to platinum for direct ethanol fuel cells (DEFCs), addressing key challenges such as catalyst poisoning and slow kinetics while leveraging the advantageous structural and electronic properties of both Pd nanoparticles and carbon dots.

Biodegradable and biocompatible hydrogels were prepared using starch and hydroxypropyl methylcellulose (HPMC) as natural polymers, and citric acid as a natural crosslinking agent, through microwave oven irradiation. The hydrogel's equilibrium degree of swelling, FTIR, FESEM, optical contact angle, hemocompatibility, and in-vitro degradation were evaluated using the soil biodegradability method. The hydrogel exhibited appreciable swelling behavior at pH 1.2, 7, and 8. FTIR and FESEM studies confirmed the formation of a hydrogel, while the optical contact angle revealed the interaction between the hydrogel and water. The hemolytic value for hydrogel was 1.03, and hydrolytic degradation was observed due to soil microbes, causing the hydrogel's observable weight loss. The prepared hydrogel was biocompatible and had a longer shelf life.