Applied Data Science and Analysis

Harnessing the Tide of Innovation: The Dual Faces of Generative AI in Applied Sciences; Letter to Editor

Wed, 10 Jan 2024 00:00:00 +0000

Advancements in Artificial Intelligence (AI) and emerging generative capabilities added paradoxical aspects. One aspect is its positive impact and limitless power it brings to users. On the other hand, concerns about the misuse of this powerful tool have consistently increased [1]. AI advancements affect all domains and sectors as they evolve in their applicable nature in the applied sciences. The more powerful AI the more influence it has on the model workflow within the specific domain and its applied field [2]. This dual nature of generative AI ignited a wide discussion on implementation and produced a debate according to the latest employed tools and technologies by scientists and researchers.

Generalized Time Domain Prediction Model for Motor imagery-based Wheelchair Movement Control

Thu, 20 Jun 2024 00:00:00 +0000

Principally, Brain-computer interface (BCI-MI)-based wheelchair control is an appropriate method for completely paralyzed people with healthy brains. In a BCI-based wheelchair control system, the procedure of pattern recognition in terms of preprocessing, feature extraction, and classification plays a significant role in avoiding recognition errors, which can lead to the initiation of the wrong command that will put the user in unsafe condition. Therefore, the objective of this research is to develop a time domain generic pattern recognition model (GPRM) for two classes of EEG-MI signals for use in a wheelchair control system. The advantage of such a GPRM is that it has a model that can also be applied for unknown subjects and is not suitable for only one subject. This GPRM has been developed, evaluated, and validated by utilizing two datasets, namely, the BCI Competition IV and Emotive EPOC datasets. Initially, fifteen time windows were investigated with seven machine learning methods to determine the optimal time window as well as the best classification method with strong generalizability. Evidently, the experimental results of this study revealed that the duration of the EEG-MI signal in the range of 4 to 6 s (4-6 s) had a high impact on the classification accuracy when the signal features were extracted using five statistical methods. Additionally, the results showed that there was a one-second latency after each command cue when utilizing the eight-second EEG-MI signal recorded by the Graz protocol applied in this study. This one second latency is inevitable because it is practically impossible for the subjects to imagine their MI hand movement instantly. Therefore, at least one second is required for subjects to prepare to initiate their motor imagery hand movement. Practically, the five statistical methods are efficient and viable for decoding the EEG-MI signal in the time domain. Evidently, the GPRM model based on the LR classifier showed generalizability by attaining an impressive percentage classification accuracy of 90%, which was validated on the Emotive EPOC dataset. Overall, the findings suggest that the GPRM developed in this study is highly adaptable and is recommended for use in real-time applications of EEG-MI-based wheelchair control systems.

An Innovative Method of Malicious Code Injection Attacks on Websites

Hussein Alnabulsi , Rafiqul Islam , Izzat Alsmadi , Savitri Bevinakoppa — Mon, 20 May 2024 00:00:00 +0000

This paper provides a model to identify website vulnerability to Code Injection Attacks (CIAs). The proposed model identifies vulnerabilities to CIA of various websites, to check vulnerable to CIAs. The lack of existing models in providing checking against code injection has motivated this paper to present a new and enhanced model against web code injection attacks that uses SQL injections and Cross-Site Script (XSS) injections. This paper previews a self-checking protection model which enables web administrators to know whether their current protection program is adequate, or whether a website needs stronger protection against CIAs. The Automated Injection’s model is to check vulnerable to cod injection. The checking methodology consists of many intrusion methods that the attacker may use to launch code injection attacks. Methodology can give a high precision of CIA vulnerability checking for a website compared with other approaches (the minimum accuracy different between proposed approach and other approaches is 3.15%). CIAs can be a serious problem for vulnerable websites including stealing, deleting, or altering important data. Extensive experiments are conducted and compared with existing research [e.g. 1, 5, and 9] to study the effectiveness of the proposed model that can check whether a website is vulnerable to CIAs. The performance of the suggested approach has been tested on SQL injections and XSS injections. The studies showed that the detection rate of our model is 95.27%, and the false positive rate is 5.55%.

Deep Transfer Learning Model for EEG Biometric Decoding

Rasha A. Aljanabi, Z.T. Al-Qaysi , M. S Suzani — Wed, 28 Feb 2024 00:00:00 +0000

In automated systems, biometric systems can be used for efficient and unique identification and authentication of individuals without requiring users to carry or remember any physical tokens or passwords. Biometric systems are a rapidly developing and promising technology domain. in contrasting with conventional methods like password IDs. Biometrics refer to biological measures or physical traits that can be employed to identify and authenticate individuals. The motivation to employ brain activity as a biometric identifier in automatic identification systems has increased substantially in recent years. with a specific focus on data obtained through electroencephalography (EEG). Numerous investigations have revealed the existence of discriminative characteristics in brain signals captured during different types of cognitive tasks. However, because of their high dimensional and nonstationary properties, EEG signals are inherently complex, which means that both feature extraction and classification methods must take this into consideration. In this study, a hybridization method that combined a classical classifier with a pre-trained convolutional neural network (CNN) and the short-time Fourier transform (STFT) spectrum was employed. For tasks such as subject identification and lock and unlock classification, we employed a hybrid model in mobile biometric authentication to decode two-class motor imagery (MI) signals. This was accomplished by building nine distinct hybrid models using nine potential classifiers, primarily classification algorithms, from which the best one was finally selected. The experimental portion of this study involved, in practice, six experiments. For biometric authentication tasks, the first experiment tries to create a hybrid model. In order to accomplish this, nine hybrid models were constructed using nine potential classifiers, which are largely classification methods. Comparing the RF-VGG19 model to other models, it is evident that the former performed better. As a result, it was chosen as the method for mobile biometric authentication. The performance RF-VGG19 model is validated using the second experiment. The third experiment attempts for verifying the RF-VGG19 model's performance. The fourth experiment performs the lock and unlock classification process with an average accuracy of 91.0% using the RF-VGG19 model. The fifth experiment was performed to verify the accuracy and effectiveness of the RF-VGG19 model in performing the lock and unlock task. The mean accuracy achieved was 94.40%. Validating the RF-VGG19 model for the lock and unlock task using a different dataset (unseen data) was the goal of the sixth experiment, which achieved an accuracy of 92.8%. This indicates the hybrid model assesses the left and right hands' ability to decode the MI signal. Consequently, The RF-VGG19 model can aid the BCI-MI community by simplifying the implementation of the mobile biometric authentication requirement, specifically in subject identification and lock and unlock classification.

Optimal Time Window Selection in the Wavelet Signal Domain for Brain–Computer Interfaces in Wheelchair Steering Control

Sat, 15 Jun 2024 00:00:00 +0000

Background and objective: Principally, the procedure of pattern recognition in terms of segmentation plays a significant role in a BCI-based wheelchair control system for avoiding recognition errors, which can lead to the initiation of the wrong command that will put the user in unsafe situations. Arguably, each subject might have different motor-imagery signal powers at different times in the trial because he or she could start (or end) performing the motor-imagery task at slightly different time intervals due to differences in the complexities his or her brain. Therefore, the primary goal of this research is to develop a generic pattern recognition model (GPRM)-based EEG-MI brain-computer interface for wheelchair steering control. Additionally, having a simplified and well generalized pattern recognition model is essential for EEG-MI based BCI applications. Methods: Initially, bandpass filtering and segmentation using multiple time windows were used for denoising the EEG-MI signal and finding the best duration that contains the MI feature components. Then, feature extraction was performed using five statistical features, namely the minimum, maximum, mean, median, and standard deviation, were used for extracting the MI feature components from the wavelet coefficient. Then, seven machine learning methods were adopted and evaluated to find the best classifiers. Results: The results of the study showed that, the best durations in the time-frequency domain were in the range of (4-7 s). Interestingly, the GPRM model based on the LR classifier was highly accurate, and achieved an impressive classification accuracy of 85.7%.

Advanced Ensemble Classifier Techniques for Predicting Tumor Viability in Osteosarcoma Histological Slide Images

Wed, 29 May 2024 00:00:00 +0000

Background: Osteosarcoma is considered as the primary malignant tumor of the bone, emanating from primitive mesenchymal cells that form osteoid or immature bone. Accurate diagnosis and classification play a key role in management planning to achieve improved patient outcomes. Machine learning techniques may be used to augment and surpass existing conventional methods towards an analysis of medical data.

Methods: In the present study, the combination of feature selection techniques and classification methods was used in the development of predictive models of osteosarcoma cases. The techniques include L1 Regularization (Lasso), Recursive Feature Elimination (RFE), SelectKBest, Tree-based Feature Importance, while the following classification methods were applied: Voting Classifier, Decision Tree, Naive Bayes, Multi-Layer Perceptron, Random Forest, Logistic Regression, AdaBoost, and Gradient Boosting. Some model assessment was done by combining metrics such as accuracy, precision, recall, F1 score, AUC, and V score.

Results: The combination of the Tree-Based Feature Importance for feature selection and Voting Classifier with Decision Tree Classifier proved to be giving a higher performance compared to all other combinations, where such combinations helped in correct classification of positive instances and wonderful minimization of false positives. Other combinations also gave significant performances but slightly less effective, for example, L1 Regularization with the Voting Classifier, RFE with the Voting Classifier.

Conclusion: This work presents strong evidence that advanced machine learning with ensemble classifiers and robust feature selection can result in overall improvement of the diagnostic accuracy and robustness for the classification of osteosarcoma. Research on class imbalance and computational efficiency will be its future research priority.

Big Data Predictive Analytics for Personalized Medicine: Perspectives and Challenges

Thu, 11 Apr 2024 00:00:00 +0000

The integration of predictive analytics into personalized medicine has become a promising approach for improving patient outcomes and treatment efficacy. This paper provides a review of the field, examining the tools, methodologies, and challenges associated with this advanced statistical methodology. Predictive analytics leverages machine learning algorithms to analyze vast datasets, including Electronic Health Records (EHRs), genomic data, medical imaging, and real-time data from wearable devices. The review explores key tools such as the Hadoop Distributed File System (HDFS), Apache Spark, and Apache Hive, which facilitate scalable storage, efficient data processing, and comprehensive data analysis. Key challenges identified include managing the immense volume of healthcare data, ensuring data quality and integration, and addressing privacy and security concerns. The paper also highlights the difficulties in achieving real-time data processing and integrating predictive insights into clinical practice. Effective data governance and ethical considerations are critical to maintaining trust and transparency. The strategic use of big data tools, combined with investment in skill development and interdisciplinary collaboration, is essential for harnessing the full potential of predictive analytics in personalized medicine. By overcoming these challenges, healthcare providers can enhance patient care, optimize resource management, and drive medical discoveries, ultimately revolutionizing healthcare delivery on a global scale.