An amalgamation of deep neural networks optimized with Salp swarm algorithm for cervical cancer detection

Computers & Electrical Engineering, Journal Year: 2025, Volume and Issue: 123, P. 110106 - 110106

Published: Jan. 28, 2025

Language: Английский

---

An Improved Ensemble-Based Cardiovascular Disease Detection System with Chi-Square Feature Selection

Computers, Journal Year: 2024, Volume and Issue: 13(6), P. 126 - 126

Published: May 22, 2024

Cardiovascular disease (CVD) is a leading cause of death globally; therefore, early detection CVD crucial. Many intelligent technologies, including deep learning and machine (ML), are being integrated into healthcare systems for prediction. This paper uses voting ensemble ML with chi-square feature selection to detect early. Our approach involved applying multiple classifiers, naïve Bayes, random forest, logistic regression (LR), k-nearest neighbor. These classifiers were evaluated through metrics accuracy, specificity, sensitivity, F1-score, confusion matrix, area under the curve (AUC). We created an model by combining predictions from different mechanism, whose performance was then measured against individual classifiers. Furthermore, we applied method 303 records across 13 clinical features in Cleveland cardiac dataset identify 5 most important features. improved overall accuracy our reduced computational load considerably more than 50%. Demonstrating superior effectiveness, achieved remarkable 92.11%, representing average improvement 2.95% over single highest classifier (LR). results indicate as viable practical improve

Language: Английский

---

A lightweight SEL for attack detection in IoT/IIoT networks

Journal of Network and Computer Applications, Journal Year: 2024, Volume and Issue: 230, P. 103980 - 103980

Published: July 26, 2024

Language: Английский

---

Decoding the black box: Explainable AI (XAI) for cancer diagnosis, prognosis, and treatment planning-A state-of-the art systematic review

International Journal of Medical Informatics, Journal Year: 2024, Volume and Issue: 193, P. 105689 - 105689

Published: Nov. 4, 2024

Language: Английский

---

Ensemble Machine Learning-Based Approach to Predict Cervical Cancer with Hyperparameter Tuning and Model Explainability

Deleted Journal, Journal Year: 2025, Volume and Issue: unknown

Published: Jan. 10, 2025

Cervical cancer remains the top killer of women at a young age in world, 85% cases are detected low-income countries. Preventive measures and therapeutic response enhanced if potential hazards identified early. This research belongs to this field by introducing an end-to-end prediction model based on individual medical records early screening data thus emphasizing discovery meaningful predictors. In order overcome issues with feature selection class imbalances, our study creates ensemble framework that blends Random Forest Logistic Regression techniques. addition achieving astounding accuracy 99.75%, guarantees transparency its decision-making processes utilizing sophisticated machine learning algorithms conjunction interpretability tools like SHAP LIME, which is essential for applications healthcare. The creation extensive method combines several classifiers, advanced techniques locating important predictive factors, help healthcare professionals better understand complex predictions some research's main investments. By offering accurate comprehensible risk assessments, novel has revolutionize clinical enhance cervical cavity identification. promotes development more proactive individualized methods fusing cutting-edge computational technology diagnostics, improving health outcomes everywhere.

Language: Английский

---

Interpretable artificial intelligence (AI) for cervical cancer risk analysis leveraging stacking ensemble and expert knowledge

Digital Health, Journal Year: 2025, Volume and Issue: 11

Published: March 1, 2025

Objectives This study develops a machine learning (ML)-based cervical cancer prediction system emphasizing explainability. A hybrid feature selection method is proposed to enhance predictive accuracy and stability, alongside evaluation of multiple classification algorithms. The integration explainable artificial intelligence (XAI) techniques ensures transparency interpretability in model decisions. Methods approach combining correlation-based recursive elimination introduced. An ensemble integrating random forest, extreme gradient boosting, logistic regression compared against eight classical ML Generative methods, such as variational autoencoders generative teaching networks, were evaluated but showed suboptimal performance. research integrates global local XAI techniques, including individual contributions tree-based explanations, interpret effects data balancing on performance are examined stabilize precision, recall, F1 scores. Classical models without preprocessing achieve 95-96% exhibit instability. Results strategies significantly creating robust model. achieves 98% with an area under the curve 99.50%, outperforming other models. Domain experts validate critical contributing features, confirming practical relevance. Incorporating domain knowledge increases transparency, making predictions interpretable trustworthy for clinical use. Conclusion Hybrid combined substantially improves reliability. supporting trustworthiness, demonstrating significant potential decision-making.

Language: Английский

---

Recent Applications of Explainable AI (XAI): A Systematic Literature Review

Applied Sciences, Journal Year: 2024, Volume and Issue: 14(19), P. 8884 - 8884

Published: Oct. 2, 2024

This systematic literature review employs the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) methodology to investigate recent applications of explainable AI (XAI) over past three years. From an initial pool 664 articles identified through Web Science database, 512 peer-reviewed journal met inclusion criteria—namely, being recent, high-quality XAI application published in English—and were analyzed detail. Both qualitative quantitative statistical techniques used analyze articles: qualitatively by summarizing characteristics included studies based on predefined codes, quantitatively analysis data. These categorized according their domains, techniques, evaluation methods. Health-related particularly prevalent, with a strong focus cancer diagnosis, COVID-19 management, medical imaging. Other significant areas environmental agricultural industrial optimization, cybersecurity, finance, transportation, entertainment. Additionally, emerging law, education, social care highlight XAI’s expanding impact. The reveals predominant use local explanation methods, SHAP LIME, favored its stability mathematical guarantees. However, critical gap results is identified, as most rely anecdotal evidence or expert opinion rather than robust metrics. underscores urgent need standardized frameworks ensure reliability effectiveness applications. Future research should developing comprehensive standards improving interpretability explanations. advancements are essential addressing diverse demands various domains while ensuring trust transparency systems.

Language: Английский

---

A Meta-Learner-Integrated Stacking Voting Ensemble Network for Cervical Malignancy Classification

2021 International Conference on Emerging Smart Computing and Informatics (ESCI), Journal Year: 2024, Volume and Issue: unknown, P. 1 - 5

Published: March 5, 2024

The aggressiveness and death rate of cervical cancer pose a serious threat to woman's health. By identifying treating the affected tissues at initial phases syndrome, complete recovery is possible. Papanicolaou (Pap) test conventional method examining cervix in order screen for cancer. Many networks automated diagnosis have recently been constructed by researchers; however, large size poor accuracy these single models precludes their practical implementation. Our proposal tackle this problem involves utilizing several Inception as base learners integrating outcomes voting ensemble. This technique called Voting-Stacking collective approach. experimental results show its potential reduce screening burden assist pathologists detecting diseases because they outperform state-of-the-art technologies now use. Furthermore, multi-level ensemble framework intended enhance outcome even more. Utilizing publically accessible dataset, our model demonstrated accuracy, precision, recall, FI are 98.30%, 99.30%, 98.49% 99.21 % correspondingly. judgments demonstrate that suggested performs admirably on pap-stained cytology pictures.

Language: Английский

---

Public health nurse perspectives on predicting nonattendance for cervical cancer screening through classification, ensemble, and deep learning models

Public Health Nursing, Journal Year: 2024, Volume and Issue: 41(4), P. 781 - 797

Published: May 17, 2024

Abstract Objectives Women's attendance to cervical cancer screening (CCS) is a major concern for healthcare providers in community. This study aims use the various algorithms that can accurately predict most barriers of women nonattendance CS. Design The real‐time data were collected from presented at OPD primary health centers (PHCs). About 1046 women's regarding and CCS included. In this study, we have used three models, classification, ensemble, deep learning compare specific accuracy AU‐ROC predicting non‐attenders CC. Results current model employs 22 predictors, with soft voting ensemble models showing slightly higher specificity (96%) sensitivity (93%) than weighted averaging. Bagging excels highest (98.49%), (97.3%), ideal (100%) an AUC 0.99. Classification reveal Naive Bayes (97%) but lower (91%) Logistic Regression. Random Forest Neural Network achieve 0.98. learning, LSTM has 95.68%, (97.60%), (93.42%) compared other models. MLP NN showed values Conclusion Employing proved effective screening.

Language: Английский

---

Prediction of Cervical Cancer With Application of Machine Learning Models

Advances in healthcare information systems and administration book series, Journal Year: 2024, Volume and Issue: unknown, P. 211 - 221

Published: June 30, 2024

Cancer accounts for a large number of fatalities each year. Cervical cancer is type that starts in the cervix. . very curable and linked to long survival high quality life when detected early. can be prevented by screening tests, such Pap smear test used identify precancerous stages. Nonetheless, there are few disheartening drawbacks includes its poor slide preparation rate human error. Consequently, computer-aided diagnosis system presented as fix issue. Artificial intelligence has been employed over healthcare industry recently, greatly facilitating accurate widespread use medical networks. plays crucial role early cervical cancer. classified normal or abnormal using deep learning machine techniques. This chapter proposes prediction associating classifiers publicly available data set based on risk factors.

Language: Английский

---