Grey wolf optimization technique with U-shaped and capsule networks-A novel framework for glaucoma diagnosis Govindharaj I, Ramesh T, Poongodai A, Senthilkumar K. P, Udayasankaran P, Ravichandran S Methodsx, 2025 The worldwide prevalence of glaucoma makes it a major reason for blindness thus proper early diagnosis remains essential for preventing major vision deterioration. Current glaucoma screening methods that need expert handling prove to be time-intensive and complicated before yielding appropriate diagnosis and treatment. Our system addresses these difficulties through an automated glaucoma screening platform which combines advanced segmentation methods with classification approaches. A hybrid segmentation method combines Grey Wolf Optimization Algorithm with U-Shaped Networks to obtain precise extraction of the optic disc regions in retinal fundus images. Through GWOA the network achieves optimal segmentation by adopting wolf-inspired behaviors such as circular and jumping movements to identify diverse image textures. The glaucoma classification depends on CapsNet as a deep learning model that provides exceptional image detection to ensure precise diagnosis. The combination of our method delivers 96.01 % segmentation together with classification precision which outstrips traditional approaches while indicating strong capabilities for discovering glaucoma at early stages. This automated diagnosis system elevates clinical accuracy levels through an automated screening method that solves manual process limitations. The detection framework produces better accuracy to improve clinical results in a strong effort to minimize glaucoma-induced blindness worldwide and display its capabilities in real clinical environments.•Hybrid GWOA-UNet++ for precise optic disc segmentation.•CapsNet-based classification for robust glaucoma detection.•Achieved 96.01 % accuracy, surpassing existing methods.
Enhanced diabetic retinopathy detection using U-shaped network and capsule network-driven deep learning Govindharaj I, Poongodai A, Gnanajeyaraman Rajaram, Santhakumar D, Ravichandran S, Vijaya Prabhu R, Udayakumar K, Yazhinian S Methodsx, 2025 Glaucoma, a severe eye disease leading to irreversible vision loss if untreated, remains a significant challenge in healthcare due to the complexity of its detection. Traditional methods rely on clinical examinations of fundus images, assessing features like optic cup and disc sizes, rim thickness, and other ocular deformities. Recent advancements in artificial intelligence have introduced new opportunities for enhancing glaucoma detection. This research explores a hybrid approach combining UNet++ and Capsule Network (CapsNet) architectures for accurate glaucoma diagnosis. UNet++ is employed for semantic segmentation, focusing on defining optic discs and cups, which are crucial for detecting the disease. CapsNet leverages its ability to recognize hierarchical patterns, providing more sensitive detection of glaucomatous changes than conventional Convolutional Neural Networks. Pre-processing of retinal images involves advanced techniques like Histogram Equalization and Contrast Limited Adaptive Histogram Equalization (CLAHE) to enhance image quality. The model is trained and tested on benchmark datasets, showing superior performance in optic cup/disc segmentation and glaucoma detection accuracy compared to existing state-of-the-art models.•Hybrid Model Efficiency: The combined use of UNet++ and CapsNet offers improved accuracy in optic cup and disc segmentation.•Enhanced Image Quality: Application of Histogram Equalization and CLAHE techniques significantly boosts the quality of retinal images.•Superior Performance: The hybrid approach outperforms traditional and contemporary models in glaucoma detection accuracy.
Human Emotion Detection Through Real-Time Facial Expressions Using Deep Learning A Poongodai, V Bhavitha, D Hemanth, G Jahnavi, S Huzma Proceedings of the 6th International Conference on Inventive Research in Computing Applications Icirca 2025, 2025 Facial emotion detection is one of the great steps in the world of artificial intelligence where machines are able to show human emotions with facial detection. The real time system for facial expression recognition using deep learning techniques has proven to have practical applications in healthcare, education and human computer interaction. It integrates two powerful pre trained networks, namely, EfficientNet B3 and ResNet 50 for accurate and efficient emotion classification. Training and evaluation is done on the RAF-DB dataset that has over 15,000 real world facial images and with 7 basic emotions assigned. The proposed system is one of the ones that include essential preprocessing techniques as face detection, alignment, normalization and data augmentation to improve performance. The training accuracy is 87.58 % and the testing accuracy is 82.53 %, meaning that it generalizes strongly. It is designed to be run in real time with input from webcam and OpenCV integration. This proves the suitability of developing intelligent systems that are capable of understanding and responding to human emotional states by means of deep learning.
Improvised Breast Cancer Detection Using Cnn With Particle Swarm Optimization (Pso) A. Poongodai, Sadlapalli Pavani, Shaik Mohammed Khalil, Pasham Jaya Prakash, Cheenepalli Reddy Ujwal Proceedings of the 6th International Conference on Inventive Research in Computing Applications Icirca 2025, 2025 Early detection and accurate diagnosis of breast cancer continues to be one of the top common and deadly diseases among women across the globe. We propose in this work, an enhanced breast cancer detection system that synergizes the use of CNN model, MobileNetV2 using Particle Swarm Optimization (PSO) to improve diagnostic performance. Deep features are extracted from mammographic images by MobileNetV2 without any human intervention, and these features are capable of capturing the details that distinguish between benign and malignant tumors. PSO, a bio-inspired optimization algorithm, is used for the purpose of tuning the hyperparameters of CNN efficiently to further refine the model's accuracy and avoid overfitting. The results presented in this hybrid framework show that it greatly increases detection precision, reduces false positive incidence and speeds up convergence of the model. The proposed approach unites deep learning with optimization algorithms to furnish a reliable, robust, and scalable solution to computer aided breast cancer diagnosis which helps the healthcare professionals in making quicker and more correct decisions which would ultimately lead to better patient care, and therefore, better patient results.
Advancing Glaucoma Detection: Synthetic Image Generation via Generative Adversarial Networks and Classification with Pretrained MobileNetV2 Ramprasad S R, Rampriya R, Poongodai A, Govindharaj I, Vimal Raja R, Yazhinian S 2024 International Conference on System Computation Automation and Networking Icscan 2024, 2024 Irreversible vision loss, generally insidious in start and lacking apparent symptoms, is usually attributed to glaucoma. Detecting glaucoma in its early stages is crucial since it may reduce disease progression. Conventional diagnostic approaches, depending on manual assessments, are notably prone to mistakes. Hence, automated glaucoma analysis acquires vital relevance for precise and prompt detection. Moreover, medical picture databases frequently display asymmetries, creating a problem. To solve these challenges, this paper presents a unique framework employing generative adversarial networks (GANs) to generate images, thus addressing dataset imbalances. Specifically., in the context of fundus images, standard approaches like image-to-image translation are applied to build synthetic fundus images and associated vascular networks, attempting to boost overall image quality and capture finer details. Gaussian filtering is originally done to pre-process the raw dataset, eliminating unwanted noise. Subsequently, GANs are applied for dataset balancing, providing synthetic images that boost classification accuracy. Optic cup segmentation is conducted using the Enhanced Level Set Algorithm. Finally, Pretrained MobileNetV2 permits accurate classification of glaucomatous images into normal and pathological categories. Experimental results illustrate the efficacy of our suggested framework, obtaining an accuracy of 98.9%, exceeding existing techniques.
Biological inspired self-organized secure autonomous routing protocol and secured data assured routing in WSN: Hybrid EHO and MBO approach G. Niranjana, A. Poongodai, K. L. S. Soujanya International Journal of Communication Systems, 2022 Summary Nowadays, the telecommunication system has attained remarkable growth, especially wireless sensor network (WSN) based on the development of various electronic devices. WSN signals are not so much solid. Here, the routers are utilized to work appropriately. Wireless networks have low expenses and are simple to install. But the major challenge concerns with providing secure data transmission via the wireless network, because the data are attacked by the hackers and also the parts of information can be stolen during transmission. In this manuscript, a biological inspired self‐organized secure autonomous routing protocol (BIOSARP) and secured data assured routing (SDAR) in WSN by means of combined elephant herding optimization (EHO) and migrating birds optimization (MBO) approach is proposed to the dynamic security of data via WSN. The BIOSARP by means of Clan updating operator in EHO so as to locate the optimal route and SDAR is advanced by updating the trust level to locate a trustworthy neighbour by means of path discovery approach of MBO algorithm. The proposed approach is executed in network simulator 2 (NS‐2) software. The experimental outcomes are analyzed using some performance evaluating metrics like delivery ratio (DR), energy consumption (EC), packet overhead (PO), and accuracy. Finally, the experimental outcomes demonstrate that the proposed approach outperforms other conventional approaches like BIOSARP‐SRTLD and BIOSARP by providing better packet DR, EC, PO, and accuracy.
A Novel Decision Support System for the Prognosis of Parkinson Disease A. Poongodai, Preety Singh, Kls Soujanya, R. Muthukumar 6th International Conference on I Smac Iot in Social Mobile Analytics and Cloud I Smac 2022 Proceedings, 2022 In this paper, Decision Support System using fuzzy logic is designed to track the progression of Parkinson disease (PD). As fuzzy logic can help people make decisions in situations when information is imprecise, incomplete or uncertain, it is widely used. In this proposed system, risk percentage of getting the disease for the unaffected subject (person), stage of the disease (Not having, Mild, Moderate and Advanced) for the given subject, and progressiveness of disease from one stage to another stage are reported. The system includes risk factor analysis to find the risk percentage of getting the disease and progression evaluation to determine the progressiveness and its nature (rapid, benign). With the inclusion of risk factor analysis in the proposed system, it is discovered that the progressiveness of the disease is known with an accuracy of 82.5%
