Balakrishnan Ramadoss

This study presents a radiomics-based framework for liver cancer analysis, integrating imaging techniques with Explainable Artificial Intelligence (XAI) methods. The workflow involves collecting imaging data, extracting radiomics features to quantify tumor characteristics, and training Machine Learning (ML) models with Shapley Additive Explanations (SHAP) and Local Interpretable Model-agnostic Explanations (LIME) to enhance interpretability. Its results demonstrate improved predictive performance, with significant imaging biomarkers identified for disease progression and classification. The integration of XAI ensures model transparency, allowing clinicians to derive actionable insights and support personalized treatment planning. This approach aim to bridge the gap between complex algorithms and clinical decision-making, advancing liver cancer diagnosis and care.

Liver cancer has significantly high mortality, especially in regions such as Africa and Asia. Early detection enhances treatment options, but indications are frequently not apparent until advanced stages. This research introduces an explainable AI (XAI) approach using a cascaded Convolutional Neural Network (CNN) combined with Gray Level Co-occurrence Matrix (GLCM)-based texture features to segregate non-cancerous from malicious tumors. The CLD system was used for assessment, and the approach was examined using the TCIA dataset, demonstrating higher accuracy and interpretability compared to prevailing techniques. XAI methods, such as feature importance and model visualization, were employed to provide details on the decision-making process of the model, ensuring transparency and reliability in clinical applications.

Pneumonia is a severe respiratory disease with potentially life-threatening consequences if not promptly diagnosed and treated. Chest X-rays are commonly employed for pneumonia detection, but interpreting the images can pose challenges. This study explores the efficacy of four popular transfer learning models, namely VGG16, ResNet, InceptionNet, and DenseNet, alongside a custom CNN model for this task. The model performance is evaluated using Mean Absolute Error (MAE) as the performance metric. The findings reveal that VGG16 outperforms the other transfer learning models, achieving the lowest MAE (66.19). To optimize the model training process, a distributed training strategy utilizing TensorFlow's TPU (Tensor Processing Unit) strategy is implemented. The custom CNN model is parallelized using TPU's multiple instances available over the cloud, enabling efficient computation parallelization and significantly reducing model training times. The experimental results demonstrate a remarkable decrease of 68.36% and 54.74% in model training times for the CNN model when trained using TPU compared to training on a CPU and GPU, respectively.

The main objective of online anomaly detection is to identify abnormal/unusual behavior such as network intrusions, malware infections, over utilized system resources due to design defects etc from real time data stream. Terrabytes of performance data generated in cloud data centers is a well accepted example of such data stream in real time. In this paper, we propose an online anomaly detection framework using non-parametric statistical technique in cloud data center. In order to determine the accuracy of the proposed work, we experiments it to data collected from RUBis cloud testbed and Yahoo Cloud Serving Benchmark (YCSB). Our experimental results shows the greater accuracy in terms of True Positive Rate (TPR), False Positive Rate (FPR), True Negative Rate (TNR) and False Negative Rate (FNR).

Attribute-based encryption (ABE) is used for achieving data confidentiality and access control in cloud environments. Most often ABE schemes are constructed using bilinear pairing which has a higher computational complexity, making algorithms inefficient to some extent. The motivation of this paper is on achieving user privacy during the interaction with attribute authorities by improving the efficiency of ABE schemes in terms of computational complexity. As a result the aim of this paper is two-fold; firstly, to propose an efficient Tate pairing algorithm based on multi-base number representation system using point halving (TP-MBNR-PH) with bases 1/2, 3, and 5 to reduce the cost of bilinear pairing operations and, secondly, the TP-MBNR-PH algorithm is applied in decentralized KP-ABE to compare its computational costs for encryption and decryption with existing schemes.

Secure Data Communication using File Hierarchy Attribute Based Encryption in Wireless Body Area Networks