Optimizing cancer classification: a hybrid RDO-XGBoost approach for feature selection and predictive insights

Cancer Immunology Immunotherapy, Journal Year: 2024, Volume and Issue: 73(12)

Published: Oct. 9, 2024

The identification of relevant biomarkers from high-dimensional cancer data remains a significant challenge due to the complexity and heterogeneity inherent in various types. Conventional feature selection methods often struggle effectively navigate vast solution space while maintaining high predictive accuracy. In response these challenges, we introduce novel approach that integrates Random Drift Optimization (RDO) with XGBoost, specifically designed enhance performance classification tasks. Our proposed framework not only improves accuracy but also offers valuable insights into underlying biological mechanisms driving progression. Through comprehensive experiments conducted on real-world datasets, including Central Nervous System (CNS), Leukemia, Breast, Ovarian cancers, demonstrate efficacy our method identifying smaller subset unique genes. This results significantly improved efficiency When compared popular classifiers such as Support Vector Machine, K-Nearest Neighbor, Naive Bayes, consistently outperforms models terms both F-measure metrics. For instance, achieved an 97.24% CNS dataset, 99.14% 95.21% Ovarian, 87.62% Breast cancer, showcasing its robustness effectiveness across different types data. These underline potential RDO-XGBoost promising for analysis, offering enhanced insights.

Language: Английский

---

Greylag goose optimization and multilayer perceptron for enhancing lung cancer classification

Scientific Reports, Journal Year: 2024, Volume and Issue: 14(1)

Published: Oct. 10, 2024

Lung cancer is an important global health problem, and it defined by abnormal growth of the cells in tissues lung, mostly leading to significant morbidity mortality. Its timely identification correct staging are very for proper therapy prognosis. Different computational methods have been used enhance precision lung classification, among which optimization algorithms such as Greylag Goose Optimization (GGO) employed. These purpose improving performance machine learning models that presented with a large amount complex data, selecting most features. As per data preparation one steps, contains operations scaling, normalization, handling gap factor ensure reasonable reliable input data. In this domain, use GGO includes refining feature selection, mainly focuses on enhancing classification accuracy compared other binary format algorithms, like bSC, bMVO, bPSO, bWOA, bGWO, bFOA. The efficiency bGGO algorithm choosing optimal features improved indicator possible application method field diagnosis. achieved highest MLP model at 98.4%. selection results were assessed using statistical analysis, utilized Wilcoxon signed-rank test ANOVA. also accompanied set graphical illustrations ensured adequacy adopted hybrid (GGO + MLP).

Language: Английский

---

Augmenting Cardiovascular Disease Prediction Through CWCF Integration Leveraging Harris Hawks Search in Deep Belief Networks

Cognitive Computation, Journal Year: 2025, Volume and Issue: 17(1)

Published: Jan. 25, 2025

Language: Английский

---

Skin cancer classification based on an optimized convolutional neural network and multicriteria decision-making

Scientific Reports, Journal Year: 2024, Volume and Issue: 14(1)

Published: July 27, 2024

Abstract Skin cancer is a type of disease in which abnormal alterations skin characteristics can be detected. It treated if it detected early. Many artificial intelligence-based models have been developed for detection and classification. Considering the development numerous according to various scenarios selecting optimum model was rarely considered previous works. This study aimed develop classification select model. Convolutional neural networks (CNNs) form AlexNet, Inception V3, MobileNet V2, ResNet 50 were used feature extraction. Feature reduction carried out using two algorithms grey wolf optimizer (GWO) addition original features. images classified into four classes based on six machine learning (ML) classifiers. As result, 51 with different combinations CNN algorithms, without GWO ML To best results, multicriteria decision-making approach utilized rank alternatives by perimeter similarity (RAPS). Model training testing conducted International Imaging Collaboration (ISIC) 2017 dataset. Based nine evaluation metrics RAPS method, AlexNet algorithm classical yielded model, achieving accuracy 94.5%. work presents first benchmarking many models. not only reduces time spent but also improves accuracy. The method has proven its robustness problem

Language: Английский

---

Hybrid feature selection module for improving performance of software vulnerability severity prediction model on textual dataset

Computing, Journal Year: 2025, Volume and Issue: 107(2)

Published: Feb. 1, 2025

Language: Английский

---

Improving stroke risk prediction by integrating XGBoost, optimized principal component analysis, and explainable artificial intelligence

BMC Medical Informatics and Decision Making, Journal Year: 2025, Volume and Issue: 25(1)

Published: Feb. 7, 2025

The relevance of the study is due to growing number diseases cerebrovascular system, in particular stroke, which one leading causes disability and mortality world. To improve stroke risk prediction models terms efficiency interpretability, we propose integrate modern machine learning algorithms data dimensionality reduction methods, XGBoost optimized principal component analysis (PCA), provide structuring increase processing speed, especially for large datasets. For first time, explainable artificial intelligence (XAI) integrated into PCA process, increases transparency interpretation, providing a better understanding factors medical professionals. proposed approach was tested on two datasets, with accuracy 95% 98%. Cross-validation yielded an average value 0.99, high values Matthew's correlation coefficient (MCC) metrics 0.96 Cohen's Kappa (CK) confirmed generalizability reliability model. speed increased threefold OpenMP parallelization, makes it possible apply practice. Thus, method innovative can potentially forecasting systems healthcare industry.

Language: Английский

---

Vine Bayes classifier based on truncated copula with application to gene expression data

Communications in Statistics - Simulation and Computation, Journal Year: 2025, Volume and Issue: unknown, P. 1 - 19

Published: Feb. 11, 2025

Language: Английский

---

A proximal policy optimisation algorithm-based algorithm for cardiovascular disorders detection

Journal of Medical Engineering & Technology, Journal Year: 2025, Volume and Issue: unknown, P. 1 - 20

Published: March 11, 2025

Cardiovascular diseases (CVDs) significantly impact athletes, impacting the heart and blood vessels. This article introduces a novel method to assess CVD in athletes through an artificial neural network (ANN). The model utilises mutual learning-based bee colony (ML-ABC) algorithm set initial weights proximal policy optimisation (PPO) address imbalanced classification. ML-ABC uses learning enhance process by updating positions of food sources with respect best fitness outcomes two randomly selected individuals. PPO makes updates ANN stable efficient improve model's reliability. Our approach formulates classification problem as series decision-making processes, rewarding every act higher rewards for correctly identifying instances minority class, hence handling class imbalance. We evaluated performance on diversified medical dataset including 26,002 who were examined within Polyclinic Occupational Health Sports Zagreb, further validated NCAA NHANES datasets verify generalisability. findings indicate that our outperforms existing models accuracies 0.88, 0.86 0.82 respective datasets. These results clinical application advance cardiovascular disorder detection methodologies.

Language: Английский

---

Security-aware user authentication based on multimodal biometric data using dilated adaptive RNN with optimal weighted feature fusion

Network Computation in Neural Systems, Journal Year: 2025, Volume and Issue: unknown, P. 1 - 41

Published: April 1, 2025

This work plans to develop a biometric authentication model by the combination of multi-modal inputs like voice, fingerprint, and iris provide high security. At first, spectrogram images, collected input were given Multi-scale Residual Attention Network (RAN) with Atrous Spatial Pyramid Pooling (ASPP) extract best values. These three features are then fed optimal weighted feature fusion, where weight optimization from is done via Enhanced Lichtenberg Algorithm (ELA). into decision-making stage, Dilated Adaptive Recurrent Neural utilized identify individuals, parameters optimized RNN using ELA improve recognition performance. The simulation findings achieved developed multimodal systems validated diverse algorithms over several efficacy metrics accuracy, precision, sensitivity, F1-score, etc. From result analysis, ELA-DARNN-based user system showed higher accuracy 96.01, other models such as 90% than SVM, CNN, CNN-AlexNet, Dil-ARNN be 87.94, 89.88, 93.25, 91.94. Therefore, outcomes explored that offered approach has attained elevated results also effectively supports reduction data theft.

Language: Английский

---

WGCNA and Machine Learning-based Integrative Bioinformatics Analysis for Identifying Key Genes of Colorectal Cancer

IEEE Access, Journal Year: 2024, Volume and Issue: 12, P. 144350 - 144363

Published: Jan. 1, 2024

Language: Английский

---