Saad Mohamed Darwish

Prof. Dr. Saad Mohamed Darwish

Alexandria University
orcid

Research Focus & Metrics

Academic Career Length

19Years (2008-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
29 July 2020 in Symmetry
Verified
Journal Article
Open access
PDF file
HKHassan Ahmed Khalil
Hassan Ahmed Khalil
Saad Mohamed DarwishSaad Mohamed Darwish
Saad Mohamed Darwish
YIYasmine M. Ibrahim
Yasmine M. Ibrahim
OHOsama Hassan
Osama Hassan
Accurate brain tumor segmentation from 3D Magnetic Resonance Imaging (3D-MRI) is an important method for obtaining information required for diagnosis and disease therapy planning. Variation in the brain tumor’s size, structure, and form is one of the main challenges in tumor segmentation, and selecting the initial contour plays a significant role in reducing the segmentation error and the number of iterations in the level set method. To overcome this issue, this paper suggests a two-step dragonfly algorithm (DA) clustering technique to extract initial contour points accurately. The brain is extracted from the head in the preprocessing step, then tumor edges are extracted using the two-step DA, and these extracted edges are used as an initial contour for the MRI sequence. Lastly, the tumor region is extracted from all volume slices using a level set segmentation method. The results of applying the proposed technique on 3D-MRI images from the multimodal brain tumor segmentation challenge (BRATS) 2017 dataset show that the proposed method for brain tumor segmentation is comparable to the state-of-the-art methods.

4 November 2023 in Stochastic Environmental Research and Risk Assessment
Verified
Journal Article
Open access
PDF file
MAMohammed S. Aly
Mohammed S. Aly
Saad Mohamed DarwishSaad Mohamed Darwish
Saad Mohamed Darwish
AAAhmed A. Aly
Ahmed A. Aly
Accurate reference evapotranspiration (ET0) estimation has an effective role in reducing water losses and raising the efficiency of irrigation water management. The complicated nature of the evapotranspiration process is illustrated in the amount of meteorological variables required to estimate ET0. Incomplete meteorological data is the most significant challenge that confronts ET0 estimation. For this reason, different machine learning techniques have been employed to predict ET0, but the complicated structures and architectures of many of them make ET0 estimation very difficult. For these challenges, ensemble learning techniques are frequently employed for estimating ET0, particularly when there is a shortage of meteorological data. This paper introduces a powerful super learner ensemble technique for ET0 estimation, where four machine learning models: Extra Tree Regressor, Support Vector Regressor, K-Nearest Neighbor and AdaBoost Regression represent the base learners and their outcomes used as training data for the meta learner. Overcoming the overfitting problem that affects most other ensemble methods is a significant advantage of this cross-validation theory-based approach. Super learner performances were compared with the base learners for their forecasting capabilities through different statistical standards, where the results revealed that the super learner has better accuracy than the base learners, where different combinations of variables have been used whereas Coefficient of Determination (R2) ranged from 0.9279 to 0.9994 and Mean Squared Error (MSE) ranged from 0.0026 to 0.3289 mm/day but for the base learners R2 ranged from 0.5592 to 0.9977, and MSE ranged from 0.0896 to 2.0118 mm/day therefore, super learner is highly recommended for ET0 prediction with limited meteorological data.

BABassant A. Abdelfattah
Bassant A. Abdelfattah
Saad Mohamed DarwishSaad Mohamed Darwish
Saad Mohamed Darwish
SESaleh Mesbah ELkaffas
Saleh Mesbah ELkaffas
Social media platforms have allowed many people to publicly express and disseminate their opinions. A topic of considerable interest among researchers is the impact of social media on predicting the stock market. Positive or negative feedback about a company or service can potentially impact its stock price. Nevertheless, the prediction of stock market movement using sentiment analysis (SA) encounters hurdles stemming from the imprecisions observed in SA techniques demonstrated in prior studies, which overlook the uncertainty inherent in the data and consequently directly undermine the credibility of stock market indicators. In this paper, we proposed a novel model to enhance the prediction of stock market movements using SA by improving the process of SA using neutrosophic logic (NL), which accurately classifies tweets by handling uncertain and indeterminate data. For the prediction model, we use the result of sentiment analysis and historical stock market data as input for a deep learning algorithm called long short-term memory (LSTM) to predict the stock movement after a specific number of days. The results of this study demonstrated a predictive accuracy that surpasses the accuracy rate of previous studies in predicting stock price fluctuations when using the same dataset.

20 July 2021 in IEEE Access
Verified
Journal Article
Open access
IEIbrahim A. Abd El-Moghith
Ibrahim A. Abd El-Moghith
Saad Mohamed DarwishSaad Mohamed Darwish
Saad Mohamed Darwish
Routing is a critical process in Wireless Sensor Networks (WSNs) since it is responsible for data transmission to base stations. Routing attacks are capable of completely destroying and degrading the function of WSNs. A trustworthy routing system is critical for ensuring routing security and WSN efficiency. Numerous studies have been conducted to increase trust between routing nodes, including cryptographic techniques, and centralized routing decisions. Nonetheless, the majority of routing methods are impractical in practice, since it is difficult to identify untrusted activities of routing nodes effectively. Meanwhile, there is no efficient method of preventing malicious node attacks. As a result of these issues, this article offers a trusted routing method that combines deep-chain and Markov Decision Processes (MDPs) in order to enhance the routing security and efficiency of WSNs. To authenticate the process of transmitting the node, the proposed approach utilizes a Proof of Authority (PoA) method inside the blockchain network. The validation group necessary for proofing is selected using a deep learning methodology that focuses on the properties of each node. MDPs are then used to choose the appropriate next hop as a forwarding node capable of transferring messages simply and securely. According to testing data, our routing system still performs well in a 50% malicious node routing environment when compared to existing routing algorithms.

19 July 2022 in IEEE Access
Verified
Journal Article
Open access
Saad Mohamed DarwishSaad Mohamed Darwish
Saad Mohamed Darwish
REReham M. Essa
Reham M. Essa
MOMohamed A. Osman
Mohamed A. Osman
AIAhmed A. Ismail
Ahmed A. Ismail
Protecting the privacy of healthcare information is an important part of encouraging data custodians to give accurate records so that mining may proceed with confidence. The application of association rule mining in healthcare data has been widespread to this point in time. Most applications focus on positive association rules, ignoring the negative consequences of particular diagnostic techniques. When it comes to bridging divergent diseases and drugs, negative association rules may give more helpful information than positive ones. This is especially true when it comes to physicians and social organizations (e.g., a certain symptom will not arise when certain symptoms exist). Data mining in healthcare must be done in a way that protects the identity of patients, especially when dealing with sensitive information. However, revealing this information puts it at risk of attack. Healthcare data privacy protection has lately been addressed by technologies that disrupt data (data sanitization) and reconstruct aggregate distributions in the interest of doing research in data mining. In this study, metaheuristic-based data sanitization for healthcare data mining is investigated in order to keep patient privacy protected. It is hoped that by using the Tabu-genetic algorithm as an optimization tool, the suggested technique chooses item sets to be sanitized (modified) from transactions that satisfy sensitive negative criteria with the goal of minimizing changes to the original database. Experiments with benchmark healthcare datasets show that the suggested privacy preserving data mining (PPDM) method outperforms existing algorithms in terms of Hiding Failure (HF), Artificial Rule Generation (AR), and Lost Rules (LR).

9 November 2021 in IEEE Access
Verified
Journal Article
Open access
WMWael A. Mahrous
Wael A. Mahrous
MFMahmoud Farouk
Mahmoud Farouk
Saad Mohamed DarwishSaad Mohamed Darwish
Saad Mohamed Darwish
Due to businesses’ growing use of IoT services in their day-to-day operations and the increased use of smart devices, digital forensic investigations involving such systems will need increasingly sophisticated digital evidence collection and processing. The majority of IoT systems are composed of disparate software and hardware components, which may pose security and privacy concerns. Recently, blockchain technology was presented as one of the options for achieving IoT security via the use of an immutable ledger, a decentralized architecture, and strong cryptographic primitives. Integrating blockchain platforms with IoT-based applications, on the other hand, poses a number of difficulties owing to the trustworthiness, integrity, and real-time responsiveness of IoT data. However, certain IoT devices may be incompatible with existing blockchain-based IoT forensic methods for additional digital evidence processing owing to their usage of conventional hash. A critical feature of cryptographic hash functions is that even if just one bit of the input is altered, the output acts pseudo-randomly, making it impossible to identify identical files. However, in the field of computer forensics, it is essential to locate comparable files (e.g., various versions of a file); therefore, we need a hash function that preserves similarity. It is getting more difficult to establish how forensic investigators might utilize traces from such devices. To effectively deal with IoT digital forensics applications, this article presents an improved blockchain-based IoT digital forensics architecture that use fuzzy hash to construct the Blockchain’s Merkle tree in addition to the conventional hash for authentication. Fuzzy hashing enables the identification of potentially damning documents that might otherwise remain undiscovered using conventional hashing techniques. By comparing blocks/files to all nodes in the blockchain network using fuzzy hash similarity, the digital forensics investigator will be able to verify their authenticity. To support the proof of concept, we simulated the suggested model.