Search

Early detection and proper treatment of epilepsy is essential and meaningful to those who suffer from this disease. The adoption of deep learning (DL) techniques for automated epileptic seizure detection using electroencephalography (EEG) signals has shown great potential in making the most appropriate and fast medical decisions. However, DL algorithms have high computational complexity and suffer low accuracy with imbalanced medical data in multi seizure-classification task. Motivated from the aforementioned challenges, we present a simple and effective hybrid DL approach for epileptic seizure detection in EEG signals. Specifically, first we use a K-means Synthetic minority oversampling technique (SMOTE) to balance the sampling data. Second, we integrate a 1D Convolutional Neural Network (CNN) with a Bidirectional Long Short-Term Memory (BiLSTM) network based on Truncated Backpropagation Through Time (TBPTT) to efficiently extract spatial and temporal sequence information while reducing computational complexity. Finally, the proposed DL architecture uses softmax and sigmoid classifiers at the classification layer to perform multi and binary seizure-classification tasks. In addition, the 10-fold cross-validation technique is performed to show the significance of the proposed DL approach. Experimental results using the publicly available UCI epileptic seizure recognition data set shows better performance in terms of precision, sensitivity, specificity, and F1-score over some baseline DL algorithms and recent state-of-the-art techniques. Post-print / Final draft

The current century has witnessed a prodigious expansion in scientific innovations contributing toward the betterment of humanity. The astonishing advancements in digital communications have facilitated various spheres of our lifestyle including the manufacturing sector. A plethora of cutting-edge technologies are rubbing shoulders to revolutionize manufacturing trends. Distributed system communications introduce a new concept of digital collaboration among a diversified range of scattered communication nodes. The individual resources of multiple nodes are intelligently integrated to formulate an aggregated impact that yields phenomenal productivity. The smart connectivity among various heterogeneous nodes may familiarize the network with an extended variety of potential security threats. The literature environs a bulk of security solutions proposed to overcome these challenges. This research study provides a comprehensive elaboration of these security threats along with the security practices designed to encounter such activities.

Breast cancer (BC) is a type of tumor that develops in the breast cells and is one of the most common cancers in women. Women are also at risk from BC, the second most life-threatening disease after lung cancer. The early diagnosis and classification of BC are very important. Furthermore, manual detection is time-consuming, laborious work, and, possibility of pathologist errors, and incorrect classification. To address the above highlighted issues, this paper presents a hybrid deep learning (CNN-GRU) model for the automatic detection of BC-IDC (+,−) using whole slide images (WSIs) of the well-known PCam Kaggle dataset. In this research, the proposed model used different layers of architectures of CNNs and GRU to detect breast IDC (+,−) cancer. The validation tests for quantitative results were carried out using each performance measure (accuracy (Acc), precision (Prec), sensitivity (Sens), specificity (Spec), AUC and F1-Score. The proposed model shows the best performance measures (accuracy 86.21%, precision 85.50%, sensitivity 85.60%, specificity 84.71%, F1-score 88%, while AUC 0.89 which overcomes the pathologist’s error and miss classification problem. Additionally, the efficiency of the proposed hybrid model was tested and compared with CNN-BiLSTM, CNN-LSTM, and current machine learning and deep learning (ML/DL) models, which indicated that the proposed hybrid model is more robust than recent ML/DL approaches.

E-health has grown into a billion-dollar industry in the last decade. Its device’s high throughput makes it an obvious target for cyberattacks, and these environments desperately need protection. In this scientific study, we presented an artificial intelligence (AI)-driven software-defined networking (SDN)-enabled intrusion detection system (IDS) to address increasing cyber threats in the E-health and internet of medical things (IoMT) environments. AI’s success in various fields, including big data and intrusion detection systems, has prompted us to develop a flexible and cost-effective approach to protect such critical environments from cyberattacks. We present a hybrid model consisting of long short-term memory (LSTM) and gated recurrent unit (GRU). The proposed model was thoroughly evaluated using the publicly available CICDDoS2019 dataset and conventional evaluation measures. Furthermore, for proper validation, the proposed framework is compared with relevant classifiers, such as cu-GRU+ DNN and cu-BLSTM. We have further compared the proposed model with existing literature to prove its efficacy. Lastly, 10-fold cross-validation is also used to verify that our results are unbiased. The proposed approach has bypassed the current literature with extraordinary performance ramifications such as 99.01% accuracy, 99.04% precision, 98.80 percent recall, and 99.12% F1-score.

The quality of human life is improving day by day and IOT plays a very important role in this improvement. Everything related to internet have some security concerns. This paper aims to improve the security in IOT environments. In any of the IOT networks the unknown and knows flaws can be a backdoor for any adversary. The increase use of such environment results in the increase of zero day cyber-attacks. This paper aims to focus on different models of DL in order to predict the attacks in IOT environments. The main aim of this research is to provide a very best solution for the detection of threats in order to improve the infrastructures of IOT. In this paper different experiments has been conducted and its results has been discussed in order to provide an effective solution

Computer systems and applications are improving day by day and with the advancement in such area it give birth to new cyber-attacks. Man in the Middle attacks (MITM) are one of those attacks. An attack where an outsider or third party enters in between two online users, where both of the users are unaware of it. The malware in such scenario mainly monitors and have the ability to change the information which is classified onsly to these two users. Mainly it is knows as a protocol to an unauthorized user within the system who can access as well as change the information of the system without leaving any trace to the existing users. This issue is critical. This paper aims to the understanding of the MITM and to understand its different categories. Finally this paper aims to present some of mechanism for the prevention of such attacks and to identify some of the future research directions in such area

The capacity and occurrence of new cyber-attacks have shattered in recent years. Such measures have very complicated workflows and comprise multiple illegal actors and organizations. Threat hunting demonstrates the process of proactively searching through networks for threats based on zero-day attacks by repeating the hunting process again and again. Unlike threat intelligence, it uses different automated security tools to collect logs in order to provide a pattern for making new intelligence-based tools by following those logs. According to our research findings about “threat hunting tools” there’s a major flaw that the designed tools are limited to the collection of logs. It works completely on logs for generating new patterns avoiding system’s main memory. Codes written directly to memory fail this process to provide proactive hunting. To overcome this major challenge, we are proposing two distinct methods, either by generating malicious code alerts or by binding memory forensics processes with threat hunting tools to make active hunting possible

The technological advancements of Internet of Things (IoT) has revolutionized traditional Consumer Electronics (CE) into next-generation CE with higher connectivity and intelligence. This connectivity among sensors, actuators, appliances, and other consumer devices enables improved data availability, and provides automatic control in CE network. However, due to the diversity, decentralization, and increase in the number of CE devices the data traffic has increased exponentially. Moreover, the traditional static network infrastructure-based approaches need manual configuration and exclusive management of CE devices. Motivated from the aforementioned challenges, this article presents a novel Software-Defined Networking (SDN)-orchestrated Deep Learning (DL) approach to design an intelligent Intrusion Detection System (IDS) for smart CE network. In this approach, we have first considered SDN architecture as a promising solution that enables reconfiguration over static network infrastructure and handles the distributed architecture of smart CE network by separating the control planes and data planes. Second, an DL-based IDS using Cuda-enabled Bidirectional Long Short-Term Memory (Cu-BLSTM) is designed to identify different attack types in the smart CE network. The simulations results based on CICIDS-2018 dataset support the validation of the proposed approach over some recent state-of-the-art security solutions and confirms it a phenomenal choice for next-generation smart CE network. Post-print / Final draft

In the published publication [...]

The Internet of Things (IoT) has proven to be a billion-dollar industry. Despite offering numerous benefits, the prevalent nature of IoT makes it vulnerable and a possible target for the development of cyber-attacks. The diversity of the IoT, on the one hand, leads to the benefits of the integration of devices into a smart ecosystem, but the heterogeneous nature of the IoT makes it difficult to come up with a single security solution. However, the centralized intelligence and programmability of software-defined networks (SDNs) have made it possible to compose a single and effective security solution to cope with cyber threats and attacks. We present an SDN-enabled architecture leveraging hybrid deep learning detection algorithms for the efficient detection of cyber threats and attacks while considering the resource-constrained IoT devices so that no burden is placed on them. We use a state-of-the-art dataset, CICDDoS 2019, to train our algorithm. The results evaluated by this algorithm achieve high accuracy with a minimal false positive rate (FPR) and testing time. We also perform 10-fold cross-validation, proving our results to be unbiased, and compare our results with current benchmark algorithms.

Objective: To evaluate the effectiveness of resection and primary anastomosis (RPA) and RPA with modified foramen colostomy in case of sigmoid twist. Place and Duration: In the Surgical Unit-II of Nishtar Hospital, Multan for one-year duration from August 2019 to August 2020. Methods: 77 patients with acute sigmoid twist were treated. A total of 47 patients underwent RPA or RPA with an orifice modified colostomy. Twenty-five patients received RPA (Group A) and the remaining 22 patients received RPA with Modified Bore Colostomy (Group B). The clinical course and postoperative complications of both groups were compared. Results: Mean hospital stay, wound infection, and mortality did not differ significantly between the groups. The rate of superficial wound infections was higher in group A (32% vs 9.1%). Anastomotic leakage was observed only in group A, with a frequency of 6.3%. The difference was numerically impressive but statistically insignificant. Conclusion: RPA with Modified Bore Colostomy produces satisfactory results. It is easy to implement and may become the method of choice in patients with sigmoid volvulus. More research is needed to further establish its role in treating sigmoid volvulus. Key words: Anastomosis; Primary resection; Surgical technique

Brain tumors (BTs) are spreading very rapidly across the world. Every year, thousands of people die due to deadly brain tumors. Therefore, accurate detection and classification are essential in the treatment of brain tumors. Numerous research techniques have been introduced for BT detection as well as classification based on traditional machine learning (ML) and deep learning (DL). The traditional ML classifiers require hand-crafted features, which is very time-consuming. On the contrary, DL is very robust in feature extraction and has recently been widely used for classification and detection purposes. Therefore, in this work, we propose a hybrid deep learning model called DeepTumorNet for three types of brain tumors (BTs)—glioma, meningioma, and pituitary tumor classification—by adopting a basic convolutional neural network (CNN) architecture. The GoogLeNet architecture of the CNN model was used as a base. While developing the hybrid DeepTumorNet approach, the last 5 layers of GoogLeNet were removed, and 15 new layers were added instead of these 5 layers. Furthermore, we also utilized a leaky ReLU activation function in the feature map to increase the expressiveness of the model. The proposed model was tested on a publicly available research dataset for evaluation purposes, and it obtained 99.67% accuracy, 99.6% precision, 100% recall, and a 99.66% F1-score. The proposed methodology obtained the highest accuracy compared with the state-of-the-art classification results obtained with Alex net, Resnet50, darknet53, Shufflenet, GoogLeNet, SqueezeNet, ResNet101, Exception Net, and MobileNetv2. The proposed model showed its superiority over the existing models for BT classification from the MRI images.

Data storage on local devices provides fast, secure and complete access to the users. However, it needs sufficient storage, which is not feasible for light-weight clients’ environments. In such scenarios, the usage of external devices makes the system vulnerable to data tampering, privacy leakage, and other data security issues. Wireless Sensor Networks (WSNs) consist of resource constrained devices, where external storage is preferred due to the lack of storage capacity in local devices. Therefore, using a centralized storage mechanism in WSNs causes slow data retrieval, which affects further operations on data. Relying on trusted parties also cause certain privacy and trust issues. Various data trading mechanisms are introduced to efficiently utilize the stored data. However, they still lack in certain aspects as discussed above. Therefore, we propose and implement a Secure Incentive-based Data Storage and Trading Model (SIDSTM), which provides a secure and efficient distributed storage mechanism. The proposed scheme uses AES-128 encryption scheme to encrypt the data for privacy purposes. An elliptic curve Diffie-Hellman key exchange is used for securely exchanging the private keys among the network peers. In WSNs, there are fewer data trading models so far. Anyone can access the data if he has access to the hash provided by the IPFS in the InterPlanetary File System (IPFS). Thus, IPFS provides efficient data access and storage mechanism. However, it deletes the data after a certain time interval due to the limited storage. Therefore, the proposed model provides incentives to motivate IPFS’s peers for data storage. It also eliminates the need for the third-party involvement in data trading. The simulations are conducted to prove the effectiveness of the proposed model by evaluating its efficiency and scalability.

Today, supplier selection &evaluation has been an important issue of strategic planning for a long time. Various artificial intelligence (AI) methods have been applied in the methodology of supplier section & evaluation from different researchers. This paper provides a review of past various artificial intelligence methods (AI) from 2010 to 2018 in the area supplier selection & evaluation problems. The review report is based on the literature of fifty articles include six review papers. Our survey presents and tries to find out the most useful and prominent AI method recommendable for the future in the problem of supplier selection &evalution.

The selection of proper suppliers is one of the most complicated works of the purchasing department. Today, supplier selection includes different conflicting objectives. Because of contradictory multi-objective supplier selection is solving by using decision-making technique. This paper is presented a modified genetic algorithm by using a combination of crossover operators, Order crossover (OX), Simulated binary crossover (SBX) to assign the optimal order quantities to each supplier, with criteria of transportation cost, product quality, and delivery time with a quantity discount. The result shows that the modified genetic algorithm is an allocated optimal order for multi vendors with improves quality as well as less computational times.