Optimizing pulmonary chest x-ray classification with stacked feature ensemble and swin transformer integration Manas Ranjan Mohanty, Pradeep Kumar Mallick, Annapareddy V N Reddy Biomedical Physics and Engineering Express, 2025 This research presents an integrated framework designed to automate the classification of pulmonary chest x-ray images. Leveraging convolutional neural networks (CNNs) with a focus on transformer architectures, the aim is to improve both the accuracy and efficiency of pulmonary chest x-ray image analysis. A central aspect of this approach involves utilizing pre-trained networks such as VGG16, ResNet50, and MobileNetV2 to create a feature ensemble. A notable innovation is the adoption of a stacked ensemble technique, which combines outputs from multiple pre-trained models to generate a comprehensive feature representation. In the feature ensemble approach, each image undergoes individual processing through the three pre-trained networks, and pooled images are extracted just before the flatten layer of each model. Consequently, three pooled images in 2D grayscale format are obtained for each original image. These pooled images serve as samples for creating 3D images resembling RGB images through stacking, intended for classifier input in subsequent analysis stages. By incorporating stacked pooling layers to facilitate feature ensemble, a broader range of features is utilized while effectively managing complexities associated with processing the augmented feature pool. Moreover, the study incorporates the Swin Transformer architecture, known for effectively capturing both local and global features. The Swin Transformer architecture is further optimized using the artificial hummingbird algorithm (AHA). By fine-tuning hyperparameters such as patch size, multi-layer perceptron (MLP) ratio, and channel numbers, the AHA optimization technique aims to maximize classification accuracy. The proposed integrated framework, featuring the AHA-optimized Swin Transformer classifier utilizing stacked features, is evaluated using three diverse chest x-ray datasets—VinDr-CXR, PediCXR, and MIMIC-CXR. The observed accuracies of 98.874%, 98.528%, and 98.958% respectively, underscore the robustness and generalizability of the developed model across various clinical scenarios and imaging conditions.
Adaptive KOA: Leaf disease classification using hyperspectral images for Internet of Things (IoT)-based sustainable agriculture Pradeep Kumar Mallick, Annapareddy V N Reddy, Manas Ranjan Mohanty, Biswajit Sahoo, Alok Kumar Jagadev Esic 2025 5th International Conference on Emerging Systems and Intelligent Computing Proceedings, 2025 In recent times, the crops as well as the agriculture management is one of the important field to watch out, therefore the image processing business provide more benefit to the crops along with support precaution needs. In this research, an adaptive Kookaburra Optimization Algorithm (Adaptive KOA) is implemented for leaf disease classification using hyperspectral images for Sustainable Agriculture in IoT.At first, the IoT modelis replicated and the routing is conductedto transmit the data from source to the Base Station (BS). Moreover, in BS the following steps are conducted. The input hyperspectral image (HSI)is subjected to a band selection module. The band selection is done using the implemented Adaptive Kookaburra Optimization Algorithm (Adaptive KOA). The developed Adaptive KOA is devised by incorporating Adaptive concept with Kookaburra Optimization Algorithm (KOA). The Selected bands are fed into the leaf segmentation module, where the Deep Embedded Clustering (DEC) technique is utilized for segmentation. Fusion of segmented image from different bands is carried out by the majority voting method. Finally, Leaf disease classification is carried out based on the 3D-Convolutional Neural Network (3DCNN) into normal and abnormal cases. The developed Adaptive KOA-3DCNN method has the maximum value for accuracy as 0.917%, highest value for True negative rate(TNR)as 0.927%, highest value for True positive rate (TPR) as 0.907%.
An empirical hybridized Siamese network using hypercube natural aggregation algorithm for handling imbalance data learning Subhashree Rout, Pradeep Kumar Mallick, Annapareddy V. N. Reddy, Meshal Alharbi, Ahmed Alkhayyat Expert Systems, 2024 Dealing with imbalanced data is a common challenge in machine learning, where one class has significantly fewer examples than another. Successfully addressing this challenge requires careful consideration of the data, algorithm, and evaluation metrics to ensure that the model accurately predicts the minority class. In this study, we present a hybrid approach called Siamese‐HYNAA, which combines a Siamese network and a population‐based optimizer hypercube natural aggregation algorithm (HYNAA) to generate candidate solutions for augmenting the minority class. We collected 10 imbalanced datasets ranging from 1.81 to 8.78 imbalanced ratios and built solution pairs based on correctly predicted candidate solutions using support vector machine (SVM). We then fed these solutions to the Siamese network, which employs a one‐shot learning approach to improve predictions with fewer candidate solutions. However, we found that SVM predicted only a small number of minority class samples accurately, prompting us to optimize the number of candidate solution pairs using HYNAA to generate more synthetic samples for the Siamese network. We evaluated our proposed strategy against basic SMOTE and our previous work, SMOTE‐PSOEV, using various performance measures, including ROC‐AUC learning curves, sensitivity, specificity, accuracy, Characteristic stability index, balanced accuracy, F1‐score, informedness, markedness, and execution time. Our results indicate that Siamese‐HYNAA generates promising results for imbalanced data.
An efficient brain tumor classification using MRI images with hybrid deep intelligence model Annapareddy V. N. Reddy, Pradeep Kumar Mallick, B. Srinivasa Rao, Phaneendra Kanakamedala Imaging Science Journal, 2024 The area of the brain affected by a brain tumour can be identified using the tumour’s shape, size, location, and border. This study seeks to develop a novel system of classification for brain tumours through pre-processing, segmentation, feature extraction, and tumour classification. An improved median filter will be applied to the input image in this initial phase to improve it. In this step, the image is segmented using a U-net model. Then, characteristics based on the Median Binary Pattern (MBP), the loop, the modified Local Gabor Directional Pattern (LGDiP), and the tumour size are retrieved. A hybrid model that fuses DBN and Bi-LSTM is presented to classify cancers. The optimal weights for both classifiers will be tuned during training to improve the classification performance. For this, BMEBEO (Blue Monkey Extended Bald Eagle Optimization) is proposed, which is a hybrid optimization technique. The suggested model obtains the maximum F-measure of 96.16%%.
Automatic COVID Protocols-Based Human Entry Check System Annapareddy V. N. Reddy, Chinthalapudi Siva Vara Prasad, Oleti Prathyusha, Duddu Sai Praveen Kumar, Jangam Sneha Madhuri Smart Innovation Systems and Technologies, 2023
On optimization efficiency of scalability and availability of cloud-based software services using scale rate limiting algorithm Annapareddy V N Reddy, A. Arun Kumar, Nookala Venu, R. Vijaya Kumar Reddy Measurement Sensors, 2022 Internet applications nowadays combine globally shared resources into a single software platform. It's a difficult technology issue to supply reports for the resource consumption among those World Wide Web applications. The formulation and simulation of spread levels were introduced in this report. Spread frequency operates together again to impose a worldwide speed restriction over revenue aggregated at several locations allowing for the synchronized monitoring of an internet company's activity. Assures that traffic delays mass transit streams act as passing through a unique, common limitation network. We describe two models overall and the other TCP-optimized—that permits network operators to expressly balance away transmission cost with network correctness, speed, and scaling. All these approaches could speed restrict 1000s of streams with little expense (less than 3% in the tested configuration). We show that with us TCP-centric architecture could scale to hundreds of servers whilst remaining resilient towards both outages and transmission postponement, suitable for countrywide telecom operators.
A survey on applications and performance of deep convolution neural network architecture for image segmentation disease classification from MRI images Journal of Advanced Research in Dynamical and Control Systems, 2018
