Integrating AI into PET/CT and PET/MRI: A paradigm shift in hybrid imaging R. T. Subhalakshmi, M. Nivaashini, G. H. Ram Ganesh, A. Sathishkumar AI Insights on Nuclear Medicine, 2025 Positron emission tomography combined with artificial intelligence is becoming a powerful tool for drug discovery. By analyzing PET imaging data with AI algorithms, researchers can find new drug targets, improve treatment plans, and better understand diseases. PET/CT is a leading cancer imaging method used in clinical practice, while combining MRI's anatomical imaging with PET's functional data offers exciting research opportunities. PET/MRI applications in cardiology, neurology, oncology, and inflammation are also expanding. Advances like Total-Body PET could revolutionize therapeutic imaging, providing deeper insights into human physiology and diseases. Integrating AI, machine learning, and deep learning into PET imaging—from image capture to interpretation—has further improved hybrid imaging techniques like PET/CT and PET/MRI, enhancing their diagnostic and research capabilities.
ALPOA: Adaptive Learning Path Optimization Algorithm for Personalized E-Learning Experiences R.T. Subhalakshmi, S. Geetha, S. Dhanabal, M. Balakrishnan International Journal of Computational and Experimental Science and Engineering, 2024 In this study, we propose the Adaptive Learning Path Optimization Algorithm (ALPOA) to enhance personalized e-learning experiences by tailoring content delivery based on individual learner profiles. ALPOA employs a hybrid optimization framework combining Genetic Algorithm (GA) and Particle Swarm Optimization (PSO) to dynamically adjust learning paths. The algorithm considers multiple factors such as learner proficiency, learning speed, engagement level, and content difficulty. Experimental results demonstrate that ALPOA outperforms traditional static e-learning models, achieving a 25% improvement in learning efficiency, a 30% increase in learner engagement, and a 20% reduction in content redundancy. The model was tested on a dataset of 1,500 learners, showing a 97% accuracy in predicting optimal learning paths and a 15% higher knowledge retention rate compared to benchmark algorithms. ALPOA’s scalability and adaptability make it a promising solution for personalized education systems, fostering improved learning outcomes and satisfaction. Future work will focus on integrating real-time feedback mechanisms and expanding the algorithm to support diverse learning environments.
Deep learning based fusion model for COVID-19 diagnosis and classification using computed tomography images RT Subhalakshmi, S Appavu alias Balamurugan, S Sasikala Concurrent Engineering Research and Applications, 2022 Recently, the COVID-19 pandemic becomes increased in a drastic way, with the availability of a limited quantity of rapid testing kits. Therefore, automated COVID-19 diagnosis models are essential to identify the existence of disease from radiological images. Earlier studies have focused on the development of Artificial Intelligence (AI) techniques using X-ray images on COVID-19 diagnosis. This paper aims to develop a Deep Learning Based MultiModal Fusion technique called DLMMF for COVID-19 diagnosis and classification from Computed Tomography (CT) images. The proposed DLMMF model operates on three main processes namely Weiner Filtering (WF) based pre-processing, feature extraction and classification. The proposed model incorporates the fusion of deep features using VGG16 and Inception v4 models. Finally, Gaussian Naïve Bayes (GNB) based classifier is applied for identifying and classifying the test CT images into distinct class labels. The experimental validation of the DLMMF model takes place using open-source COVID-CT dataset, which comprises a total of 760 CT images. The experimental outcome defined the superior performance with the maximum sensitivity of 96.53%, specificity of 95.81%, accuracy of 96.81% and F-score of 96.73%.
MACHINE LEARNING ALGORITHMS FOR PREDICTION OF DISEASES International Journal of Mechanical Engineering, 2022
A Deep Learning Model Based on Multi-Objective Particle Swarm Optimization for Scene Classification in Unmanned Aerial Vehicles Aghila Rajagopal, Gyanendra Prasad Joshi, A. Ramachandran, R. T. Subhalakshmi, Manju Khari, Sudan Jha, K. Shankar, Jinsang You IEEE Access, 2020 Recently, the increase in inexpensive and compact unmanned aerial vehicles (UAVs) and light-weight imaging sensors has led to an interest in using them in various remote sensing applications. The processes of collecting, calibrating, registering, and processing data from miniature UAVs and interpreting the data semantically are time-consuming. In UAV aerial imagery, learning effective image representations is central to the scene classification process. Earlier approaches to the scene classification process depended on feature coding methods with low-level hand-engineered features or unsupervised feature learning. These methods could produce mid-level image features with restricted representational abilities, which generally yielded mediocre results. The development of convolutional neural networks (CNNs) has made image classification more efficient. Due to the limited resources in UAVs, it is hard to fine-tune the hyperparameters and the trade-offs between classifier results and computation complexity. This paper introduces a new multi-objective optimization model for evolving state-of-the-art deep CNNs for scene classification, which generates the non-dominant solutions in an automated way at the Pareto front. We use a set of two benchmark datasets to test the performance of the scene classification model and make a detailed comparative study. The proposed method attains a very low computational time of 80 sec and maximum accuracy of 97.88% compared to all other methods. The proposed method is found to be appropriate for the effective scene classification of images captured by UAVs.
An obstacle detection and distance sensing algorithm for visually impaired persons International Journal of Scientific and Technology Research, 2019
Active steganalysis based on adapted Lempel-Ziv complexity and approximate entropy estimation G. S. Raman, R. T. Subhalakshmi 2013 IEEE Conference on Information and Communication Technologies ICT 2013, 2013 Steganalysis is the aptitude of identifying stegogrammes that contain a secret message. To test the steganalysis algorithm touching the unusual Stego images to ensure its robustness. This paper proposes an active steganalysis method uses irregularity in the stego image or the specialized algorithm is used to detect the existence of stego image. DCT coefficients are intended for each block. Approximate entropy estimation is used to normalize the randomness in a time series. Lempel-Ziv complexity characterizes the degree of order or disorder and development of spatiotemporal patterns. Moreover, adapted Lempel-Ziv complexity added to decide the block size. Second order statistics estimates the codebook, document to store the message. The proposed steganalysis deduces the properties of the hidden message in high rate embedding stego image. The evaluation scheme has a better performance of evaluating the steganalysis algorithm and can provide a quantitative evaluation criterion for steganalysis algorithm.
