International Research Journal of Multidisciplinary Technovation

Novel Attention-Tuned CNN for Emotion and Stress Recognition Using EEG in Cognitive Therapy Sessions

Sat, 03 Jan 2026 00:00:00 +0000

Emotional and stress responses are essential for interpreting cognition and mental health in therapy. Traditional models insufficiently seize the complicated temporal-spatial correlations in EEG data, resulting in imprecise assessments of patients' emotional states. This research presents a Novel Attention-Tuned deep learning (DL) model for recognizing emotions and stress by real-time EEG signals gathered during Cognitive Therapy Sessions, intending to improve recognition accuracy and the interpretability of neural activity connected to psychological states. 2000 EEG data were collected, which undergoes preprocessing employing adaptive median filtering to eliminate noise and artifacts. Discrete Wavelet Transform (DWT) using features is extracted to both temporal and spectral characteristics of the brain’s activity patterns. The See-See Partidge Chicks-driven Attention-Tuned Convolutional Neural Network (SSPC-Att-CNN) model is used to recognize human emotions and stress levels by examining EEG data. This architecture integrates convolutional layers for spatial feature learning with an attention mechanism that adaptively concentrates on the most informative EEG regions corresponding to emotional arousal and stress variations. A SSPC Optimization (SSPCO) algorithm is employed for feature selection, enhancing computational efficacy and robustness. Experimental evaluation was conducted on python 3.10 for standard EEG datasets which demonstrates that the proposed AT-CNN significantly outperforms conventional DL models in recognizing emotional valence in accuracy of 82.10 %, arousal in the accuracy of 97.89%, and stress intensity levels. The system provides real-time cognitive monitoring, giving therapist’s objective insights into emotional regulation and stress patterns. This improves Artificial Intelligence (AI)-assisted psychological assessment tools and promotes a data-driven understanding of emotions in therapy.

Hybrid BIRCH-ACO and PSO-MST Strategies for Energy-Aware Data Aggregation in Software-Defined WSNs

Chitra R, Sudarmani R — Wed, 10 Dec 2025 00:00:00 +0000

In a Wireless Sensor Network (WSN), dozens or hundreds of battery-driven sensors communicate with one another. Batteries have to be replaced frequently when nodes are deployed in unattended environments. Internet of Things (IoT) applications are becoming increasingly scalable and energy-efficient, making energy-efficient data aggregation a critical research focus. As part of this study, two hybrid data aggregation frameworks are presented and evaluated in order to optimize energy consumption and network performance. In the first framework, hierarchical clustering is performed using BIRCH (Balanced Iterative Reduction and Clustering Using Hierarchies), while mobile base station shunting is performed using Ant Colony Optimization (ACO). Using Particle Swarm Optimization (PSO), optimal cluster heads and base stations can be placed, and routing paths can be optimized using the Minimum Spanning Tree (MST) algorithm. Software-defined WSNs reduce computational overhead and improve adaptability by utilizing a software-defined architecture. According to a comparison of energy efficiency, network lifetime, control overhead, and data latency metrics, both approaches outperform traditional static clustering methods significantly; however, the BIRCH and ACO model excels in adaptive clustering and load distribution, while the PSO and MST model provides the best path optimization and the least amount of delay in data transmission.

Study of Streptomyces extract as an effective inhibitor against microbiologically influenced corrosion by Pseudomonas aeruginosa biofilm on API 5L X52 steel

Bouchakour F, Driche E, Unsal T, Fares C, Taskın O.S, Aksu A, Sebaihia M — Tue, 13 Jan 2026 00:00:00 +0000

This research investigates the effect of a Streptomyces-derived extract (SE) on microbiologically influenced corrosion (MIC) induced by Pseudomonas aeruginosa biofilms on API 5L X52 steel. Various amounts of SE were evaluated in P. aeruginosa cultures by 7-day immersion assays. SE was extracted from Actinobacteria strains found in a soil sample from the Béchar-Kenadsa region of the Sahara. Gas chromatography-mass spectrometry (GC-MS) was utilized to analyze the chemical composition of the extract. This facilitated the identification of specific components that inhibit corrosion. The anticorrosion efficacy of SE against biofilm-induced corrosion under microbiological influence (CMI) of API 5L X52 steel caused by P. aeruginosa was assessed by potentiodynamic polarization (PDP) and electrochemical impedance spectroscopy (EIS). Scanning electron microscopy (SEM) was employed to examine the surface morphology. The findings indicate that the incorporation of SE significantly inhibits the development of P. aeruginosa biofilms on the coupons. This inhibitory activity leads to a significant reduction in the rate of microbiologically induced corrosion, commonly attributed to bacterial colonization.. In silico investigations validated that the discovered compounds exhibit a strong affinity for the corrosive bacterium P. aeruginosa, elucidating their inhibitory function.

Cross-Domain Transfer Learning with Vision Transformers for Fatigue Recognition from Facial Expressions

Tue, 13 Jan 2026 00:00:00 +0000

Fatigue detection plays a critical role in mission-critical environments such as defense operations, transportation, and industrial control, where sustained attention and alertness are paramount for safety, operational efficiency, and human performance. This study introduces a real-time, non-intrusive fatigue detection system that employs Vision Transformers (ViTs) to identify subtle facial cues associated with fatigue. Unlike traditional methods that rely primarily on overt indicators such as yawning, eye closure, or head nodding, our approach leverages advanced deep learning techniques to capture nuanced micro-expressions and subtle behavioral patterns that are often overlooked. By applying transfer learning from the FER-2013 dataset to the NTHU-DDD dataset, we achieve enhanced model generalization, with ViT-L16 and R50+ViT-B16 architectures attaining classification accuracies of 70.31% and 72.79%, respectively. Visual attention maps reveal that FER-FFR models focus more effectively on critical facial regions, enabling precise feature extraction and interpretability. Furthermore, we introduce a fatigue level indicator that quantifies fatigue progression over consecutive video frames, demonstrating behavior that closely aligns with human fatigue dynamics. The proposed system is robust, scalable, and suitable for deployment in real-world operational settings, providing an automated, reliable, and objective solution for continuous fatigue monitoring, potentially enhancing safety, productivity, and decision-making in high-stakes environments.

BN+SiC Hybrid Nanofluid in Enhanced Microchannels: Numerical Heat Transfer Augmentation Study

Anirban Bose, Arunabha Chanda — Wed, 10 Dec 2025 00:00:00 +0000

This research article numerically investigated the performance of non-oxide hybrid Boron Nitride Silicon Carbide (BN+SiC/water) hybrid nanofluid in high heat flux miniaturized electronic hardware. 3d-CFD model validated with established experimental data on a triangular section, oblique microchannel geometry is used to explore the influence of total particle loading (0.5 to 1.5 %), relative particle proportion, on the Nusselt number (Nu), friction factor (f) and thermal performance factor (TPF). The results compared with the benchmark conventional oxide hybrid nanofluid system (Al₂O₃+CuO/water) and found superior on overall thermo-hydraulic performance. Heat transfer enhancement by 50% with respect to base fluid water, is quite an improvement in thermal enhancement, if we compare with the benchmark oxide system reference of around 37%. It is also observed that relative SiC proportion increases the performance of this system. Nanoparticle size and morphology effects on the thermo-hydraulic performance is also studied in this work. Smaller size particles are found beneficial in a quantitative analysis in the range of 10nm to 90nm average particle diameter. Non spherical high aspect ratio shapes nanoparticles enhance the performance of the nanofluid observed in this study. This study not only introduced a novel advanced heat transfer fluid but also allow the design customization insight for this BN+SiC/water hybrid nanofluid system.

Fruit Crop Recommendations for Indian Agriculture- A Machine Learning Approach Incorporating Environmental and Soil Parameters

Umamaheswari R, Vimala P, Jayalakshmi V, Saravana Mahesan S — Fri, 16 Jan 2026 00:00:00 +0000

Agriculture is crucial to the existence of man, and its productivity in a country plays a significant role in the growth of the economy. Conventionally, farmers used experience to know how to cultivate and the types of crops to grow, but developments in climatic and environmental conditions have, in recent years, complicated the process of selecting the right crop for farmers. This research will answer this question by applying machine learning (ML) methods to make a fruit crop suggestion using soil macronutrients such as Nitrogen, Phosphorus, and Potassium, temperature, humidity, rainfall, and soil pH. Decision Tree (DT), Random Forest (RF), K-Nearest Neighbor (KNN), Naive Bayes (NB), and Support Vector Machine (SVM) were comparatively tested out of 5 ML algorithms. According to experiment results, the predictive accuracy of 95.45 with 5-fold validation of 5-fold validation is the highest, with better ability to interpret features, strong resistance to noise, and high levels of consistency. NPK patterns, climate effects, and correlation-based interactions feature analysis further amplify the accuracy of the model. These findings affirm that ML-based soil-climate fusion has the capability to provide a scalable, interpretable, and effective framework of sustainable fruit crop recommendation in the changing agro-climatic conditions in India.

Multilayer Colour Image Encryption Scheme Based on Discrete Compound Chaotic Map and S-box

Deep Singh, Lalthazuala, Sandeep Kumar, Jatinder Kumar, Amit Paul — Tue, 13 Jan 2026 00:00:00 +0000

In today’s technological age, ensuring the security of data transmitted across unsecured and open channels from one destination to another is an important issue. Strong techniques to protect images during transmission and storage are becoming more and more critical as the digital age advances. The main focus of the presented work is to enhance the protection of digital image data from unapproved sources over these open networks. It is possible that conventional encryption methods may not provide enough protection against contemporary cryptographic attacks. Thus, this study proposes an encryption method that combines the cryptographic properties of S-boxes and Baker’s map together with the chaotic dynamics of the discrete compound chaotic (DCC) map. During encryption, the level of confusion and diffusion is well maintained. The confusion is achieved by employing the chaotic sequence obtained from the DCC map and through the use of Baker’s map. The S-box and a DCC map are utilized for diffusion purposes. Further, to achieve a better scrambling effect, Baker’s map is implemented multiple times (n_b k times). The innovation of the proposed scheme lies in its novel integration of the discrete compound chaotic map (DCC map) with the cryptographic properties of S-boxes and Baker’s map, achieving an enhanced level of diffusion and confusion in encrypted images. Further, the hashSHA−256 utilized to derive initial conditions ensures the dependency of the proposed scheme on the original image, offering a robust defence against differential attacks and providing a more secure framework than traditional encryption techniques. The strong proposed scheme’s capability was demonstrated via the following key results: average entropy value of 7.9971, low correlation coefficients in the encrypted images, high MSE values, average encryption time of 0.2599 seconds, UACI of 33.4765, NPCR of 99.6113, successful decryption without data loss. These statistical and simulation analysis results confirm the scheme’s efficiency, high security, and robustness.