Accurate detection and diagnosis of brain tumors at early stages is significant for effective treatment. While numerous methods have been developed for tumor detection and classification, several rely on traditional techniques, often resulting in suboptimal performance. In contrast, AI-based deep learning techniques have shown promising results, consistently achieving high accuracy across various tumor types while maintaining model interpretability. Inspired by these advancements, this paper introduces an improved variant of EfficientNet for multi-grade brain tumor detection and classification, addressing the gap between performance and explainability. Our approach extends the capabilities of EfficientNet to classify four tumor types: glioma, meningioma, pituitary tumor, and non-tumor. For enhanced explainability, we incorporate gradient-weighted class activation mapping (Grad-CAM) to improve model interpretability. The input MRI images undergo data augmentation before being passed through the feature extraction phase, where the underlying tumor patterns are learned. Our model achieves an average accuracy of 98.6%, surpassing other state-of-the-art methods on standard datasets while maintaining a substantially reduced parameter count. Furthermore, the explainable AI (XAI) analysis demonstrates the model’s ability to focus on relevant tumor regions, enhancing its interpretability. This accurate and interpretable model for brain tumor classification has the potential to significantly aid clinical decision-making in neuro-oncology.

Deduplication has been largely employed in distributed storage systems to improve space efficiency. Traditional deduplication research ignores the design specifications of shared-nothing distributed storage systems such as no central metadata bottleneck, scalability, and storage rebalancing. Further, deduplication introduces transactional changes, which are prone to errors in the event of a system failure, resulting in inconsistencies in data and deduplication metadata. In this paper, we propose a robust, fault-tolerant and scalable cluster-wide deduplication that can eliminate duplicate copies across the cluster. We design a distributed deduplication metadata shard which guarantees performance scalability while preserving the design constraints of shared-nothing storage systems. The placement of chunks and deduplication metadata is made cluster-wide based on the content fingerprint of chunks. To ensure transactional consistency and garbage identification, we employ a flag-based asynchronous consistency mechanism. We implement the proposed deduplication on Ceph. The evaluation shows high disk-space savings with minimal performance degradation as well as high robustness in the event of sudden server failure.

With the widespread global trend of voice phishing or vishing attacks, the development of effective detection models using artificial intelligence (AI) has been hindered by the lack of high-quality and large volumes of data. This lack of data reflecting a real vishing scenario often leads to imbalanced datasets and biased detection models. Therefore, we present in this paper a data augmentation (DA) method for expanding the imbalanced Korean call content vishing (KorCCVi) dataset to address the existing data asymmetry problem and enhance the performance of Korean vishing detection. The proposed approach for DA involves using the back-translation (BT) method with three different intermediate languages: English, Chinese, and Japanese. The proposed method offers several advantages over the traditional synthetic minority oversampling technique (SMOTE), which is the main technique used to compare with our multilingual BT approach. Using these two DA techniques, several machine learning (ML) and deep learning (DL) models were trained on the original imbalanced dataset, the dataset balanced with SMOTE and its variants, and the dataset augmented with our method. We analyzed the impact of these DA methods on the performance of the models, demonstrated the benefits of each approach, and suggested the most suitable approach. The performance of the trained models was evaluated using the accuracy, precision, recall, and F1-score metrics. The experimental results demonstrated that the proposed multilingual BT method effectively expands the dataset while preserving its contextual and linguistic characteristics. The average performance of the models revealed that those trained on the augmented dataset outperformed the other models. They achieved F1-scores of 98.91% for the back-translated data, 98.14% for the original data, and 97.23% for SMOTE.

Deduplication has been principally employed in distributed storage systems to improve storage space efficiency. Traditional deduplication research ignores the design specifications of shared-nothing distributed storage systems such as no central metadata bottleneck, scalability, and storage rebalancing. Likewise, inline deduplication integration poses serious threats to storage system read/write performance, consistency, and scalability. Mainly, this is due to ineffective and error-prone deduplication metadata, duplicate lookup I/O redirection, and placement of content fingerprints and data chunks. Further, transaction failures after deduplication integration often render inconsistencies in data chunks, deduplication metadata, and garbage data chunks. results in rendering inconsistencies in data chunks, deduplication metadata, and garbage data chunks. In this paper, we propose GRATE, a high-performance inline cluster-wide data deduplication, complying with the design constraints of shared-nothing storage systems. In particular, GRATE eliminates duplicate copies across the cluster for high storage space efficiency without jeopardizing performance. We employ a distributed deduplication metadata shard, which promises high-performance deduplication metadata and duplicate fingerprint lookup I/Os without introducing a single point of failure. The placement of data and deduplication metadata is made cluster-wide based on the content fingerprint of chunks. We decouple the deduplication metadata shard from read I/O path and replace it with a read manifestation object to further speedup read performance. To guarantee deduplication-enabled transaction consistency and efficient garbage identification, we design a flag-based asynchronous consistency scheme, capable of repairing the missing data chunks on duplicate arrival. We design and implement GRATE in Ceph. The evaluation shows an average of 18% performance bandwidth improvement over the content addressable deduplication approach at smaller chunk sizes, i.e., less than 128KB while maintaining high storage space savings.

Large scale machine learning (ML) and deep learning (DL) platforms face challenges when integrated with deduplication enabled storage clusters. In the quest to achieve smart and efficient storage utilization, removal of duplicate data introduces bottlenecks since deduplication alters the I/O transaction layout of the storage system. Therefore, it is critical to address such deduplication overhead for acceleration of ML/DL computation in deduplication storage. Existing state of the art ML/DL storage solutions such as Alluxio and AutoCache adopt non deduplication-aware caching mechanisms, which lacks the much needed performance boost when adopted in deduplication enabled ML/DL clusters. In this paper, we introduce Redup, which eliminates the performance drop caused by enabling deduplication in ML/DL storage clusters. At the core, is a Redup Caching Manager (RDCM), composed of a 2-tier deduplication layout-aware caching mechanism. The RDCM provides an abstraction of the underlying deduplication storage layout to ML/DL applications and provisions a decoupled acceleration of object reconstruction during ML/DL read operations. Our Redup evaluation shows negligible performance drop in ML/DL training performances as compared to a cluster without deduplication, whilst significantly outperforming Alluxio and AutoCache in terms of various performance metrics.