Search

Abstract The emerging expanding scope of the Internet of Things (IoT) necessitates robust intrusion detection systems (IDS) to mitigate security risks effectively. However, existing approaches often struggle with adaptability to emerging threats and fail to account for IoT-specific complexities. To address these challenges, this study proposes a novel approach by hybridizing convolutional neural network (CNN) and gated recurrent unit (GRU) architectures tailored for IoT intrusion detection. This hybrid model excels in capturing intricate features and learning relational aspects crucial in IoT security. Moreover, we integrate the feature-weighted synthetic minority oversampling technique (FW-SMOTE) to handle imbalanced datasets, which commonly afflict intrusion detection tasks. Validation using the IoTID20 dataset, designed to emulate IoT environments, yields exceptional results with 99.60% accuracy in attack detection, surpassing existing benchmarks. Additionally, evaluation on the network domain dataset, UNSW-NB15, demonstrates robust performance with 99.16% accuracy, highlighting the model’s applicability across diverse datasets. This innovative approach not only addresses current limitations in IoT intrusion detection but also establishes new benchmarks in terms of accuracy and adaptability. The findings underscore its potential as a versatile and effective solution for safeguarding IoT ecosystems against evolving security threats.

Data temperature is a response to the ever-growing amount of data. These data have to be stored, but they have been observed that only a small portion of the data are accessed more frequently at any one time. This leads to the concept of hot and cold data. Cold data can be migrated away from high-performance nodes to free up performance for higher priority data. Existing studies classify hot and cold data primarily on the basis of data age and usage frequency. We present this as a limitation in the current implementation of data temperature. This is due to the fact that age automatically assumes that all new data have priority and that usage is purely reactive. We propose new variables and conditions that influence smarter decision-making on what are hot or cold data and allow greater user control over data location and their movement. We identify new metadata variables and user-defined variables to extend the current data temperature value. We further establish rules and conditions for limiting unnecessary movement of the data, which helps to prevent wasted input output (I/O) costs. We also propose a hybrid algorithm that combines existing variables and new variables and conditions into a single data temperature. The proposed system provides higher accuracy, increases performance, and gives greater user control for optimal positioning of data within multi-tiered storage solutions.

The convergence of smart technologies and predictive modelling in prisons presents an exciting opportunity to revolutionize the monitoring of inmate behaviour, allowing for the early detection of signs of distress and the effective mitigation of suicide risks. While machine learning algorithms have been extensively employed in predicting suicidal behaviour, a critical aspect that has often been overlooked is the interoperability of these models. Most of the work done on model interpretations for suicide predictions often limits itself to feature reduction and highlighting important contributing features only. To address this research gap, we used Anchor explanations for creating human-readable statements based on simple rules, which, to our knowledge, have never been used before for suicide prediction models. We also overcome the limitation of anchor explanations, which create weak rules on high-dimensionality datasets, by first reducing data features with the help of SHapley Additive exPlanations (SHAP). We further reduce data features through anchor interpretations for the final ensemble model of XGBoost and random forest. Our results indicate significant improvement when compared with state-of-the-art models, having an accuracy and precision of 98.6% and 98.9%, respectively. The F1-score for the best suicide ideation model appeared to be 96.7%.

One of the challenges in image processing involves single image deraining (SID). Existing methods do not exploit images at multi-scales and often overlook spatial and channel information. These methods fail to account for different rain conditions, leading to challenges in effectively removing a wide range of rain streak patterns, such as diverse directional or dense streaks. Moreover, they often result in the loss of fine texture details or a blurred background in the process of eliminating rain streaks from images captured in heavy rain. Mostly state-of-the-art (SOTA) deraining models achieve higher performance in removing rain from rainy images but at the expense of a high number of parameters, which results in computational complexity and memory requirements. Furthermore, they also do not consider high-level visioned evaluation metrics to perform deep evaluations of the proposed models. In this paper, a simple lightweight network with few parameters and relatively shallow depth is proposed by fusing the traditional Gaussian-Laplacian pyramid technique with the attention module. We propose a novel attention-based lightweight pyramid network for image deraining and detection (LPN-IDD) to achieve better deraining performance. The proposed model includes a dual attention module integrated with the Gaussian-Laplacian pyramids network. In LPN-IDD, different levels of Laplacian pyramids can extract multi-scale features to adapt to different shapes and types of rain streaks. Residual and recursive blocks are used in each subnetwork along with dual attention blocks to resist the occlusion or texture feature while suppressing unnecessary features. Extensive experimentation performed on the SID synthetic and real-world datasets demonstrate the effectiveness of the proposed model in image deraining tasks. For deep evaluation of the proposed methodology, object detection models including faster-RCNN, YOLO-V3, YOLO-V7 are used along with full reference evaluation metrics.

Facial paralysis is a debilitating condition that weakens or damages facial muscles resulting in asymmetric or abnormal facial movements. To aid in the diagnosis and rehabilitation of facial paralysis, researchers have developed machine learning and deep learning computer-aided diagnosis systems. However, machine learning models have limitations as they rely on facial landmark techniques and manual face palsy region extraction methods to obtain spatial information. Moreover, deep learning models need large, labelled datasets for training whereas existing available facial paralysis datasets are small and restricted. This presents significant challenges, including difficulties in data acquisition, insufficient patient numbers, and inadequate diversity within the datasets. These limitations can potentially restrict the generalizability of these models and introduce biases in the resulting outcomes. In this study, we propose an approach for the diagnosis and grading of facial paralysis comprised of two datasets, one from MEEI (Massachusetts Eye and Ear Infirmary) videos of patients and the other from the YFP (YouTube Face Palsy) dataset. The model uses a transfer learning approach to fine-tune the VGGFace model, which is pre-trained on facial images, on the prepared datasets for facial paralysis. The resultant model was subsequently renamed as FP-VGGFace for the purpose of this research. Additionally, two more pre-trained models on facial images, ResNet50 and VGG16, are also fine-tuned for the facial paralysis task. This was undertaken to conduct a performance comparison of multiple models on the prepared dataset. The findings indicate that the models exhibit high accuracy, benefiting from pre-training on a diverse dataset that enables the capture of spatial information from facial images. The FP-VGGFace model achieves the best accuracy (99.3%) and F1-score (99.3%) surpassing all benchmark models. This study underscores the potential of utilizing pre-trained deep learning models for the diagnosis and rehabilitation of facial paralysis.

Due to the advancement of digital twin (DT) technology, Health 4.0 applications have become reality and starting to take roots. In this article, we focus on intracranial hemorrhage (ICH) which is a life-threatening emergency that needs immediate diagnosis and treatment. ICH is caused by bleeding inside the skull or brain. Radiologists typically examine computed tomography (CT) scans of the patients to determine the ICH and its subtype. But the manual assessment of the CT scan is a complex and time-consuming task. The existing pre-trained convolutional neural network (CNN) models are state-of-the-art for ICH classification. However, they employ poor feature extraction techniques which hinder overall model performance. Furthermore, they suffer from the curse of dimensionality and use redundant and noisy features. The problem of imbalanced data is also crucial for achieving model generalization. This paper proposes a hybrid attention-based ResNet architecture for ICH detection and classification. An attention mechanism allows the model to focus on a specific region and extract relevant features. Principal component analysis (PCA) is used for dimensionality reduction and redundant feature removal whereas deep convolutional generative adversarial network (DCGAN) is used for resolving the class imbalance problem. The proposed model is evaluated using the dataset assembled during the Radiologist Society of North America (RSNA) ICH detection challenge 2019. The results show that our proposed model outperforms existing state-of-the-art models in terms of accuracy and F1-score. ICH classification achieved accuracies of 99.2%, 97.1%, 96.7%, 96.7% and 96.1%, for detecting epidural hemorrhage (EH), intraparenchymal hemorrhage (IH), intraventricular hemorrhage (IVH), subdural hemorrhage (SH), and subarachnoid hemorrhage (SAH) subtypes respectively. The F1-score of 96.1% for EH subtype is also best when compared with the benchmark models.

Person re-identification has inspired a lot of interest due to its significance in intelligent video surveillance. It is a difficult task due to the presence of critical challenges such as changes in appearance, misalignment, occlusion and background noise. Batch drop block layer (BDB) has been used recently in person re-identification by exploiting the feature erasing procedure. However, BDB drops a block of features randomly, resulting in the loss of contextual information, which makes the model difficult to train. Also, due to the random dropping of features, large area of discriminative information may lose during training, resulting in low efficiency and performance. To address this problem and to improve the model representation power, we propose a novel, lightweight, self-adaptive bottleneck attention module with a self-attention branch to improve the model performance by reducing the parameter overhead with negligible computation cost. The proposed approach entails bottleneck attention module (BAM) which is incorporated between ResNet layers to remove the background noise and to nominate the high-level semantic part. Further, dilated convolutions with batch normalization are used to tackle the contextual information loss problem and to avoid overfitting. In addition, an informative global branch is used which captures the global representation of the network, and the attention branch entails the multiscale local salient information. Two types of loss functions including softmax and batch hard triplet are used in the training process for each branch, forcing the network to encapsulate the common attribute within the similar identity and to maintain distance between distinct individuals. Compared with BDB, our network improves the mAp to 88.1%, and Rank-1 gets 96.3% for the market-1501 dataset. The results on Cuhk-03-Detected dataset showed 79.2% mAp score, with 81.4 %, Rank-1, whereas on Cuhk-03-labelled dataset, a mAP score of 81.3% and a Rank-1 score of 83.3% is achieved. Experiments reveal that ResNet model with addition of multiple BAM layers performs consistently over the state-of-the-art datasets using softmax and batch hard triplet loss.