Feature selection of pre-trained shallow CNN using the QLESCA optimizer: COVID-19 detection as a case study

Hamad, Qusay Shihab; Samma, Hussein; Suandi, Shahrel Azmin

doi:10.1007/s10489-022-04446-8

Cited by 11 publications

(6 citation statements)

References 89 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It has been analyzed that certain limitations of classical computations can be addressed by applying quantum technologies (QT). Therefore, various studies incorporate quantum technologies such as quantum transfer learning and quantumbased chameleon swarm quantum annealing for FE and FS methodologies [27][28][29]. In 2023, research has been published that shows how a quantum computation (QC) based feature selection method can reduce the model complexity and improve the ML interpretability.…”

Section: Related Workmentioning

confidence: 99%

Quantum-enhanced hybrid feature engineering in thoracic CT image analysis for state-of-the-art nodule classification: an advanced lung cancer assessment

Shivwanshi,

Nirala

2024

Biomed. Phys. Eng. Express

View full text Add to dashboard Cite

The intricate nature of lung cancer treatment poses considerable challenges upon diagnosis. Early detection plays a pivotal role in mitigating its escalating global mortality rates. Consequently, there are pressing demands for robust and dependable early detection and diagnostic systems. However, the technological limitations and complexity of the disease make it challenging to implement an efficient lung cancer screening system. AI-based CT image analysis techniques are showing significant contributions to the development of computer-assisted detection (CAD) systems for lung cancer screening. Various existing research groups are working on implementing CT image analysis systems for assessing and classifying lung cancer. However, the complexity of different structures inside the CT image is high and comprehension of significant information inherited by them is more complex even after applying advanced feature extraction and feature selection techniques. Traditional and classical feature selection techniques may struggle to capture complex interdependencies between features. They may get stuck in local optima and sometimes require additional exploration strategies. When applied to a prominent feature space, traditional techniques may also struggle with combinatorial optimization problems. This paper proposed a methodology to overcome the existing challenges by applying feature extraction using Vision Transformer (FexViT) and Feature selection using the Quadratic unconstrained binary optimization (FSelQUBO) technique. This algorithm shows better performance when compared with other existing techniques. The proposed methodology showed better performance as compared to other existing techniques when evaluated by applying necessary output measures, such as accuracy, Area under roc (receiver operating characteristics) curve, precision, sensitivity, and specificity, obtained as 94.28%, 99.10%, 96.17%, 90.16% and 97.46%. The further advancement of CAD systems is essential to meet the demand for more reliable detection and diagnosis of cancer, which can be addressed by leading the proposed quantum computation and growing AI-based technology ahead.

show abstract

Section: Related Workmentioning

confidence: 99%

Quantum-enhanced hybrid feature engineering in thoracic CT image analysis for state-of-the-art nodule classification: an advanced lung cancer assessment

Shivwanshi,

Nirala

2024

Biomed. Phys. Eng. Express

View full text Add to dashboard Cite

show abstract

“…Afterwards, each dimension of the search agent is compared to the threshold value, which is a random value between 0 and 1 and is based on Eq. (15). The binary value corresponding to each dimension is calculated accordingly.…”

Section: The Binary Modified Cat and Mouse Based Optimizer (Bmcmbo)mentioning

confidence: 99%

“…The applications of ML techniques in the diagnosis and treatment of all diseases are not identical. For instance, ML methods in data analysis are more commonly used to diagnose, monitor, and treat patients with cancer, heart problems, kidney diseases, and COVID-19 due to the high incidence rate, high mortality rate, expensive tests, and some other reasons [14][15][16]. The process of learning about health conditions and treatments through the analysis of medical records is known as medical data mining, which is a type of ML technique [17].…”

Section: Introductionmentioning

confidence: 99%

Binary modified Cat and Mouse based Optimizer for medical feature selection: A COVID-19 case study

Parizi

2024

Preprint

View full text Add to dashboard Cite

Recent technological advances in medical diagnosis have led to the generation of high-dimensional datasets. The presence of redundant and irrelevant features in these datasets can have adverse effects on the performance of machine learning (ML) methods and reduce the accuracy of their results. Therefore, feature selection (FS), i.e., a popular preprocessing method in ML, is used to select the optimal subsets of features to improve the accuracy of ML methods. This performance enhancement is more crucial while addressing high-dimensional medical issues. Since FS is a multiobjective binary optimization problem, it is necessary to develop efficient FS algorithms. Although metaheuristic algorithms (MAs) have been widely used for FS in medicine, they face different challenges in most applications, e.g., a lack of sufficient effectiveness and scalability to select the most effective features in small and large medical datasets. The cat and mouse-based optimizer (CMBO) is a novel MA based on the natural competitive behavior of cats and mice. Despite its acceptable performance in a variety of problems, the CMBO faces various challenges such as limited exploitation abilities, an unbalanced search mechanism, and high fluctuation in solutions to complex problems, e.g., FS. This paper proposes a modified and binary version of the CMBO called the BMCMBO to enhance the performance in selecting effective features from medical datasets. The BMCMBO involves significant modifications to the method of updating the positions of search agents, the method of selecting mice, the effect of the positional information of the most optimal member of the population, and the addition of the adaptive step size. These modifications are meant to improve the exploitation abilities, boost the accuracy of the solutions, and balance the search process when dealing with the FS problem in medical datasets. The performance of the proposed algorithm on 12 real medical datasets was compared with the performance of the most effective MA and CMBO variants. The statistical results demonstrated that BMCMBO was more effective than other evaluated methods. In addition, the BMCMBO algorithm was employed to select features and diagnose COVID-19 in a real case study. The proposed algorithm identified healthy and infected COVID-19 correctly samples with an accuracy of 98.4\%, demonstrating its superiority.

show abstract

“…Building on the success of CNNs, recent studies have utilized pre-trained deep CNNs like VGG19 [30], [31] to obtain generic visual representations [32], [33]. Researchers suggested multidomain visual neural models to capture the inherent traits of fabricated news images more effectively.…”

Section: ) Single-modality Based Classification Approaches Using Imag...mentioning

confidence: 99%

Utilizing Ensemble Learning for Detecting Multi-Modal Fake News

Luqman,

Faheem,

Ramay

et al. 2024

IEEE Access

View full text Add to dashboard Cite

The spread of fake news has become a critical problem in recent years due extensive use of social media platforms. False stories can go viral quickly, reaching millions of people before they can be mocked, i.e., a false story claiming that a celebrity has died when he/she is still alive. Therefore, detecting fake news is essential for maintaining the integrity of information and controlling misinformation, social and political polarization, media ethics, and security threats. From this perspective, we propose an ensemble learning-based detection of multi-modal fake news. First, it exploits a publicly available dataset Fakeddit consisting of over 1 million samples of fake news. Next, it leverages Natural Language Processing (NLP) techniques for preprocessing textual information of news. Then, it gauges the sentiment from the text of each news. After that, it generates embeddings for text and images of the corresponding news by leveraging Visual Bidirectional Encoder Representations from Transformers (V-BERT), respectively. Finally, it passes the embeddings to the deep learning ensemble model for training and testing. The 10-fold evaluation technique is used to check the performance of the proposed approach. The evaluation results are significant and outperform the state-of-the-art approaches with the performance improvement of 12.57%, 9.70%, 18.15%, 12.58%, 0.10, and 3.07 in accuracy, precision, recall, F1-score, Matthews Correlation Coefficient (MCC), and Odds Ratio (OR), respectively.

show abstract

Feature selection of pre-trained shallow CNN using the QLESCA optimizer: COVID-19 detection as a case study

Cited by 11 publications

References 89 publications

Quantum-enhanced hybrid feature engineering in thoracic CT image analysis for state-of-the-art nodule classification: an advanced lung cancer assessment

Quantum-enhanced hybrid feature engineering in thoracic CT image analysis for state-of-the-art nodule classification: an advanced lung cancer assessment

Binary modified Cat and Mouse based Optimizer for medical feature selection: A COVID-19 case study

Utilizing Ensemble Learning for Detecting Multi-Modal Fake News

Contact Info

Product

Resources

About