MLR Institute of Technology

With high prevalence rates for mental health disorders and the emerging interest in digital behavioural data, computational approaches for depression detection have attracted increasing research attention in recent years. Works available in the literature primarily follow a single-modality approach, using either an audio, text, or visual signal, resulting in inadequate reflection of heterogeneous depressive features and the mitigation of high-dimensional behavioural signals. However, any such approaches typically struggle with the following issues, limiting their practical deployment: modality imbalance, robustness to heterogeneous data sources, and transparency. To overcome such limitations, we present a novel attention-guided multi-source multimodal depression detection framework, named MoodSenseAI, that constructs an ensemble deep learning model from multimodal data, including text, speech, and video. Our framework is based on a DeepMoodNet model with multiple modality-specific encoders (TextMoodEncoder, SpeechMoodEncoder, and FaceMoodEncoder) that learn rich semantic, acoustic, and behavioural representations from independent benchmark datasets for each modality. A modality-specific attention-based fusion module aggregates these embeddings at the representation level, thereby providing context-dependent importance weights that enable adaptive modelling of complementary depression-related patterns across heterogeneous datasets. The experimental results on standard benchmark datasets show that the proposed framework achieves significantly higher accuracy, precision, recall, and AUC-ROC than unimodal baselines and improper fusion strategies (macro accuracy = 94.3

Hybrid composites have attracted increasing attention for medical and prosthetic applications; however, achieving an optimal balance between mechanical strength, fatigue durability, wear resistance, and moisture stability remains a critical challenge for long-term biomedical use. In this context, the present study investigates the influence of Kevlar and Himalayan nettle fibers reinforced with carp fish–derived collagen macromolecules on the overall performance of hybrid polymer composites aimed at human prosthetic applications. Five different composite configurations were fabricated and evaluated: B (100 vol

This study introduces a simple and environmentally friendly hydrothermal method for creating sponge-like nickel vanadate (Ni3V2O8, NVO) nanostructures that are attached to multiwalled carbon nanotubes (MWCNTs) to improve the performance of electrochemical energy storage. The novel in-situ anchoring of NVO on conductive MWCNT frameworks creates a three-dimensional interconnected network that allows ions and electrons to move quickly. XRD, FESEM, TEM, and XPS analyses of the structure and morphology showed that the NVO/MWCNT composites were phase-pure and evenly spread out, with strong interfacial coupling. Electrochemical tests showed that NVO/MWCNT-5 had a much higher specific capacitance of 389 F/g at 2 mA/cm2, which is almost twice as high as that of pure NVO (201 F/g). The NVO/MWCNT-5//AC asymmetric supercapacitor had an energy density of 4.86 Wh/kg and kept 93 % of its capacity after 10,000 cycles. These results confirm the proposed hybrid design, which combines the high pseudocapacitance of NVO with the great conductivity and mechanical strength of MWCNTs. The material's ability to be made in large quantities, its low cost, and its compatibility with the environment make it a good choice for next-generation energy storage in portable electronics, hybrid electric vehicles, and grid-level systems.

The current study aims to look at the Darcy-Forchheimer and bioconvective flow of Casson blood-based trihybrid nanofluid (THNF) through an exponentially expanding surface. The three different nanoparticles, namely, cobalt ferrite (CoFe2O4), molybdenum disulfide (MoS2), and zirconium dioxide (ZrO2) are used to make the THNF. The impacts of viscous dissipation, magnetic field, thermal radiation, Brownian motion, thermophoresis, heat consumption/generation, and thermophoretic particle deposition are also included in this investigation. Using appropriate variables, the set of partial differential equations representing the fluid models are transformed into a system of ordinary differential equations and these equations are numerically solved using the ND solver and the bvp4c approach. Our outcomes are validated through the earlier publication results. Physical traits such as fluid velocity, temperature, nanofluid (NF) concentration, and motile microorganisms are shown graphically. The results show that improving the porosity parameter diminishes the velocity profile. The temperature profile decays when enhancing the value of the Casson parameter. The NF concentration profile suppresses as the thermophoretic particle deposition parameter enhances. The profile of microorganisms declines when enhancing the bioconvective Lewis number. Accelerating the magnetic field parameter makes a reduction in skin friction coefficient. The raise in the radiation parameter improves the heat transmission rate. The larger thermophoresis parameter declines the rate of mass transfer, and the motile microorganisms density diminishes when enlarging the value of the Peclet number. In addition, the long-short term memory model is used to optimize the heat transfer gradient data by training, validating, and testing to determine the data accuracy. The training mean square error (MSE) is 0.001089, 0.000195, 0.000236, and 0.000499, the validation MSE is 0.001665, 0.000647, 0.000629, and 0.000694, and the test MSE is 0.001779, 0.000158, 0.000154, and 0.000269 for Cattaneo-Christov heat and mass flux model (CCHMFM) with suction, CCHMFM with injection, Fourier heat and mass flux model (FHMFM) with suction, and FHMFM with injection, respectively.

In recent years, the widespread adoption of the Internet of Things has played an important role in advancing artificial intelligence by continuously generating large volumes of data used for model training and decision-making processes. The conventional cloud edge computing paradigm faces challenges in handling the massive data generated by Internet of Things. These challenges include high latency, excessive bandwidth usage, limited scalability, and privacy risks. To rectify these limitations, this research proposes a novel Software-defined Wrapper Discriminant Federated learning-Reinforcement Attention Adversarial Regression (SWDF-RAAR) model. Unlike other existing studies, the SWDF-RAAR model jointly addresses privacy analysis and intelligent resource task management within a unified structure. For privacy analysis, the model integrates software-defined networking, linear discriminant analysis with wrapper-style bi-directional removal technique, federated learning, and extreme learning machines. Resource task management is performed using a combination of dynamic perceptrons, scaled dot-product attention, a multilayer perceptron with graph convolution, deep Q-learning aided by generative adversarial networks, and a support vector machine. The main contribution of this model is to develop a scalable, lightweight, and privacy-preserving model for detecting intrusions while enhancing the efficiency of resource task management in cloud edge computing environments. Experimental analysis on various security based datasets and cloud-edge parameters demonstrated that the proposed model attained 98.73% detection accuracy, 93% scalability, 95% quality of service, 96% network efficiency, and 5.3 ms latency.