Deep Attention-Driven Multi-Scale AI Framework for Automated Breast Histopathology Analysis Ravi Kumar. M, N.Krishnaveni, Gotte Ranjith Kumar, Rajesh Kumar Tripathi, Anandakumar Haldorai, Veeraswamy Ammisetty 2026 International Conference on ICT and Photonics Ictp 2026 Advancing ICT Photonics for A Smarter Sustainable World Proceedings, 2026 Breast cancer diagnosis based on histopathological examination is a critical process but a very time-consuming one, which requires expert interpretation, commonly hampered by inter-observer variability and high complexity of tissue heterogeneity. To overcome these limitations, this study recommends a strong deep learning framework for automatic analysis of breast histopathology images with multi-scale convolutional Neural Networks(CNN)-Transformer feature extractor. The main goal is to improve the diagnostic accuracy, interpretability and clinical applicability with an end-to-end workflow combining the digitization, preprocessing, multi-scale feature extraction, attention-guided tissue classification and intelligent reporting of a whole slide image (WSI). The methodology utilizes stain normalization, artifact correction, as well as region of interest extraction that are followed by a hybrid CNN-Transformer model based on the extraction of both local cellular morphology and global tissue context. Attention mechanisms are dedicated to diagnostically interesting areas, and an ongoing learning cycle allows the adaptation of models based on the feedback provided by the pathologist. Experimental validation using BreaKHis dataset containing multi magnification images of breast histopathology (40x-400x) showed high performance of the proposed framework with 94.2% accuracy, 0.96 AUC, 0.92 F1-score, 93.5% sensitivity and 94.8% specificity as compared to benchmark models such as SAMASK-CLTR and ResNet-SVM. Overall, the results confirm that the proposed multi-scale artificial intelligence (AI) pathology framework is an efficient, explainable and scalable solution to improve the precision of diagnoses and assist clinical decision-making in digital breast pathology.
Strategic of Wind Farms Re-Powering the Market-Driven Assessment Across Integrated Energy Systems Ravi Kumar Mugadhanam, R Sujitha, Sumitra Sureliya, Ananda Kumar Haldorai, Puneet, Takveer Singh 4th IEEE International Conference on Power Electronics and Iot Applications in Renewable Energy and Its Control Parc 2026, 2026 With the expanding uptake of combined energy power systems, there is mounting attention to renewable energy sources and storage technologies. This research centers on addressing the issues posed by the variability and instability inherent in wind energy, which hamper its market competitiveness and operability reliability. In an effort to resolve these challenges, a new analytical approach was established to examine maturity of wind markets and determine viability of repowering old wind farms at local, national, and international levels. On the basis of systematic criteria, the research detects unevenness in the progression of markets within various U.S. states, indicating premature aging in a number of wind farms, particularly in Texas, where lifespans in operation have been below the estimated twenty-year threshold. These results underscore the pressing necessity of strategic repowering measures and offer insight into the wider development of wind energy infrastructure within consolidated energy systems.
CROSS-LAYER ATTENTION ADAPTATION FOR REAL-TIME NEURAL INFERENCE IN EMBEDDED DEVICES Journal of Theoretical and Applied Information Technology, 2026
Design and Verification of AHB to I2C Protocol Krithika A, Ravikumar M, Nagesh K N, Veerappa Chikkagoudar 2025 IEEE International Conference on Communication Networks and Computing Cnc 2025, 2025 Transfer of data between low-speed external devices and high-speed system buses is frequently necessary for modern System-on-Chip (SoC) platforms. The Advanced High-performance Bus (AHB) provides parallel, high-throughput data transfers suitable for processors and memory units; However, the Inter Integrated Circuit (I2C) network protocol provides a straightforward two-wire serial interface that is frequently utilized for control devices, sensors, and memories. Direct data exchange between these protocols is not possible due to their architectural differences. The implementation and validation of an AHB-to I2C networking connection using Verilog HDL are presented in this paper. The suggested design functions as an AHB slave to record parallel transactions and transform them into serial data that is compatible with I2C for peripheral communication. The simulation and implementation in the Synopsys environment were used to validate the design, which was written in Verilog HDL. The findings accurately verify that the AHB to I2C protocol interface is a good fit for SoC applications requiring communication between low-speed peripheral modules and high-speed cores
A Smart Wi-Fi Enabled IoT Framework for Fishermen Tracking and Communication Ravi Kumar M, Shivani M, Sahana VM, Shivaranjini S, Sridhar N 2nd International Conference on Electronics Computing Communication and Control Technology Iceccc 2025, 2025 For fisherman who may travel great distances and spend days or weeks on the open ocean, the unpredictability of maritime weather poses a number of difficulties. The inability to clearly differentiate international borders, which is even worse by inclement weather, is a significant concern that has resulted in route diversions and safety issues. To tackle these issues. the fisherman and the base station (Navy personnel) must communicate effectively. Research aiming at creating a cutting-edge marine system to improve fishermen safety and security is covered in this work. Real-time communication, automated alarms for excessive sea wave levels and ship vibrations during cyclones, and emergency signaling capabilities to the base station are all features of the suggested system. The device would also help officials provide prompt backup assistance to fishermen who are having trouble. The research seeks to protect fishermen livelihoods and enhance marine security by integrating these characteristics. It would also make it possible for officials to keep an eye on the sea conditions and promptly assist fishermen who are in trouble. This work aims to protect a safe and sustainable fishing environment, improve marine security, and protect fishermen livelihoods by combining these qualities with state-of-the-art technology. For any angler, navigating in maritime environments is a crucial Components of any hunting expedition.
Smart Energy Management Leveraging Twin Adaptive Pulse Coupled Networks for Dynamic Energy Optimization in IoT-based Electrical WSN Ravi Kumar M, B. Md. Irfan, Sugunadevi C, Umang Soni, Madhu B K, Ramya Maranan Proceedings of 5th International Conference on Trends in Material Science and Inventive Materials Ictmim 2025, 2025 Internet of Things (IoT)-driven Wireless Sensor Networks (WSNs) undergo fast growth hence requiring sophisticated energy optimization methods to keep the networks operational longer with reliable data handling. Traditional energy management practices lead to early node failure combined with inefficient network routes and non-even energy distribution which blocks network development and operational performance expansion. The three main factors that cause WSNs to be energy inefficient are improper node positioning along with excessive routing overhead and uneven distribution of power consumption across sensor nodes. The current network management approaches do not provide adequate dynamic energy distribution which results in premature network failure. This research establishes "Smart Energy Management leveraging Twin Adaptive Pulse Coupled Networks for Dynamic Energy Optimization in IoT-Based WSN (MG-TwinAPC-ReP)" to address these challenges. The proposed framework MG-TwinAPC-ReP features four layers which (1) strategic node deployment coverage, (2) Cluster-Based Routing Protocol Using Modified Greylag-Goose Optimization, and (3) Energy management through adaptive load balancing using Twin Adaptive Pulse Coupled Network's dual synchronization model and (4) uses Reformed Poplar Optimization to optimize networking parameters. Experimental results indicate remarkable performance capabilities which lead to a 99.82% increase in network lifetime and 99.74% energy conservation together with 99.91% reliable data delivery and 99.65% reduced latency compared to traditional IoT-WSN systems. The proposed scalable self-adapting energy-efficient WSN model provides an optimal solution for smart cities together with healthcare and agriculture and industrial IoT applications which drives sustainable IoT-driven WSN deployment into the future.
Deep Learning based Intelligent Spectrum Sensing Framework Optimizing Dynamic Radio Resource Allocation Ravi Kumar M, Mahesh Kumar A. S, Abbas Thajeel Rhaif Alsahlanee, Bhargav H K, Amit Barve, Abhijit Mitra 4th IEEE International Conference on Distributed Computing and Electrical Circuits and Electronics Icdcece 2025, 2025 Efficient spectrum utilization in today's wireless communications requires intelligent spectrum sensing because dynamic spectrum access depends on it for resource allocation. This study presents a new deep learning framework for spectrum sensing which uses convolutional neural networks together with long short-term memory networks. Decision-making processes in real-time employ hybrid architecture which analyses both spectrum data spatial patterns as well as its temporal evolution through reinforcement learning mechanisms. The spectrum sensing framework using deep learning achieved 97.8% accuracy in detecting spectrum holes while reaching 95.3% precision in identifying primary users through its implementation which resulted in a 42% better spectrum utilization than conventional energy detection methods. Under -20dB to 20dB SNR conditions the system maintained steady performance that generated false alarms less than 0.03 times per observation. The proposed system design provides improved spectrum detection capabilities and resource distribution capabilities which makes it applicable for on-going wireless networks in congested urban spaces.
Harnessing Photons for Next-Generation Computational Speed and Efficiency M Ravi Kumar, S. Sivakumar, Sachin Aralikatti, T Raja Santhosh Kumar, K S Chakradhar, Sandeep Gupta 2025 IEEE International Conference on Emerging Technologies and Applications Mpsec Iceta 2025, 2025 In today’s technological age, computers have greatly improved people’s daily lives. Though computer processing speeds are extremely fast when compared to human abilities, they still need to be significantly increased to meet future demands. This is in contrast to the relatively exponential growth in the advancement of other technologies that rely on computers. There is no longer any hope for electrons since humans have driven them to their absolute limit. Photons, on the other hand, may substitute for the slower electrons. Something that can do everything an electron can, but at a million times the speed and with significantly greater dependability in some manner, taking computing to a level nobody could have imagined. This study presents the applications of photonics in the computing industry, discusses their potential as an alternative to electrons, and compares the two from a computational standpoint. It also covers the generalized operation of optical computers based on silicon, the applications of photons, and their critical role in the future.
Enhancing Latency Performance using Non-Local Scalable Quantum Neural Network in Ultra-Dense IoT Networks Beyond 5G Ravi Kumar M, P. S. V. Srinivasa Rao, Jayant Tyagi, Amit Barve, Natrayan L, M. SivaramKrishnan Proceedings of the 4th International Conference on Intelligent Computing Information and Control Systems Icoiics 2025, 2025 In ultra-dense Internet of Things (IoT) over 5G networks, the rapid increase of latency-critical applications like autonomous systems and real-time healthcare has imposed strict demands on low-energy and low-latency computation. Ultra-dense networks are faced with dynamic traffic, large task diversity, and high-density base station deployment, rendering conventional optimization techniques invalid. To mitigate these issues, in this paper, a Enhancing Latency Performance using Non-Local Scalable Quantum Neural Network in Ultra-Dense IoT Networks Beyond 5G (NLS-QNN-SBiA) is suggested. The system initiates with federated data collection from mobile IoT devices and edge nodes, collecting parameters such as task complexity, link quality, and computing capability. The data are fed into a communication-computation model that defines a latency-energy joint optimization problem. Non-Local Scalable Quantum Neural Network (NLS-QNN) uses quantuminspired non-local learning that captures global task relationships to create offloading and resource allocation solutions with low latency, which are then optimized through Swarm Bipolar Algorithm (SBiA). The technique has astounding improvements in key metrics: <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{9 9. 1 2 \%}$</tex> less latency, 99.75% better energy efficiency, 99.36% accuracy of task offloading, and 99.93% resource usage. These are better than state-of-the-art techniques in dynamic environments. In summary, the NLS-QNN-SBiA system offers strong, scalable, and adaptive latency optimization, facilitating real-time responsiveness in next-generation ultra-dense IoT networks and facilitating effective MEC-based orchestration over 5G.
