SRADHO: statistical reduction approach with deep hyper optimization for disease classification using artificial intelligence

Scientific Reports, Год журнала: 2025, Номер 15(1)

Опубликована: Янв. 7, 2025

Artificial Intelligence techniques are being used to analyse vast amounts of medical data and assist in the accurate early diagnosis diseases. The common brain related diseases faced by most people which affects structure function brain. neural networks have been extensively for disease prediction due their ability learn complex patterns relationships from large datasets. However, there some problems like over-fitting, under-fitting, vanishing gradient increased elapsed time occurred course analysis results performance degradation model. Therefore, a perception is much essential avoiding over-fitting under-fitting. This empirical study presents statistical reduction approach along with deep hyper optimization (SRADHO) technique better feature selection classification reduced time. Deep combines learning models hyperparameter tuning automatically identify relevant features, optimizing model accuracy reducing dimensionality. SRADHO calibrate weight, bias select optimal number hyperparameters hidden layer using Bayesian approach. uses probabilistic efficiently search space, identifying configurations that maximize while minimizing evaluations. Three benchmark datasets classifier logistic regression, decision tree, random forest, K-nearest neighbour, support vector machine Naïve Bayes experimentation. proposed algorithm achieves 98.2% accuracy, 97.2% precision rate, 98.3% recall rate 98.1% F1-Score value 0.3% error rate. execution 12 s.

Язык: Английский

---

Enhanced cardiovascular disease prediction through self-improved Aquila optimized feature selection in quantum neural network & LSTM model

Frontiers in Medicine, Год журнала: 2024, Номер 11

Опубликована: Июнь 20, 2024

Introduction Cardiovascular disease (CVD) stands as a pervasive catalyst for illness and mortality on global scale, underscoring the imperative sophisticated prediction methodologies within ambit of healthcare data analysis. The vast volume medical available necessitates effective mining techniques to extract valuable insights decision-making prediction. While machine learning algorithms are commonly employed CVD diagnosis prediction, high dimensionality datasets poses performance challenge. Methods This research paper presents novel hybrid model predicting CVD, focusing an optimal feature set. proposed encompasses four main stages namely: preprocessing, extraction, selection (FS), classification. Initially, preprocessing eliminates missing duplicate values. Subsequently, extraction is performed address issues, utilizing measures such central tendency, qualitative variation, degree dispersion, symmetrical uncertainty. FS optimized using self-improved Aquila optimization approach. Finally, hybridized combining long short-term memory quantum neural network trained selected features. An algorithm devised optimize LSTM model’s weights. Performance evaluation approach conducted against existing models specific measures. Results Far dataset-1, accuracy-96.69%, sensitivity-96.62%, specifity-96.77%, precision-96.03%, recall-97.86%, F1-score-96.84%, MCC-96.37%, NPV-96.25%, FPR-3.2%, FNR-3.37% dataset-2, accuracy-95.54%, sensitivity-95.86%, specifity-94.51%, F1-score-96.94%, MCC-93.03%, NPV-94.66%, FPR-5.4%, FNR-4.1%. findings this study contribute improved by efficient with Discussion We have proven that our method accurately predicts cardiovascular unmatched precision conducting extensive experiments validating methodology large dataset patient demographics clinical factors. QNN frameworks tuning increase forecast accuracy reveal risk-related physiological pathways. Our shows how advanced computational tools may alter sickness management, contributing emerging field in healthcare. used revolutionary produced significant advances

Язык: Английский

---

Multimodal Skin Cancer Prediction: Integrating Dermoscopic Images and Clinical Metadata with Transfer Learning

The Open Bioinformatics Journal, Год журнала: 2025, Номер 18(1)

Опубликована: Янв. 28, 2025

Background Skin cancers exist as the most pervasive in world; to increase survival rates, early prediction has become more predominant. Many conventional techniques frequently depend on visual review of clinical information and dermoscopic illustrations. In recent technological developments, enthralling algorithms combining modalities are used for increasing diagnosis accuracy deep learning. Methods Our research proposes a multi-faceted approach skin cancer that incorporates metadata with visuals. The pre-trained convolutional neural networks, like EfficientNetB3, were images along transfer learning excavate some attributes this study. Moreover, TabNet was processing metadata, including age, gender, medical history. features obtained from both fusion integrated enhance accuracy. benchmark datasets, ISIC 2018, 2019, HAM10000, assess model. Results proposed system achieved 98.69% classification cancer, surpassing model snapshots data. convergence substantially enhanced resilience, demonstrating importance multimodal lesion diagnosis. Conclusion This focused mainly efficiency integrating visuals using prediction. offers promising tool improving diagnostic accuracy, further could explore its application other fields requiring data integration.

Язык: Английский

---

EBHOA-EMobileNetV2: a hybrid system based on efficient feature selection and classification for cardiovascular disease diagnosis

Computer Methods in Biomechanics & Biomedical Engineering, Год журнала: 2025, Номер unknown, С. 1 - 23

Опубликована: Фев. 19, 2025

The accurate prediction of cardiovascular disease (CVD) or heart is an essential and challenging task to treat a patient efficiently before occurring attack. Many deep learning machine frameworks have been developed recently predict in intelligent healthcare. However, lack data-recognized appropriate methodologies meant that most existing strategies failed improve accuracy. This paper presents healthcare framework based on model detect disease, motivated by present issues. Initially, the proposed system compiles data from multiple publicly accessible sources. To quality dataset, effective pre-processing techniques are used including (i) interquartile range (IQR) method identify eliminate outliers; (ii) standardization technique handle missing values; (iii) 'K-Means SMOTE' oversampling address issue class imbalance. Using Enhanced Binary Grasshopper Optimization Algorithm (EBHOA), dataset's features chosen. Finally, presence absence CVD predicted using MobileNetV2 (EMobileNetV2) model. Training evaluation approach were conducted UCI Heart Disease Framingham Study datasets. We obtained excellent results comparing with recent methods. beats current approaches concerning performance metrics, according experimental results. For research achieves higher accuracy 98.78%, precision 99%, recall 99% F1 score 99%. 99.39%, 99.50%, learning-based classification combined feature selection yielded best innovative has potential enhance consistency prediction, which would be advantageous for clinical practice care.

Язык: Английский

---

IntelliNet: intelligent deep net architecture for efficient cardiovascular disease prediction

Multimedia Tools and Applications, Год журнала: 2025, Номер unknown

Опубликована: Март 25, 2025

---

Securing Cloud Computing Environment via Optimal Deep Learning-based Intrusion Detection Systems

Опубликована: Май 17, 2024

Язык: Английский

---

Deep Learning Model and Multi-Modal Late Fusion For Predicting Adverse Events Following Cardiothoracic Surgery in the ICU Using STS Data and Time Series Intraoperative Data

medRxiv (Cold Spring Harbor Laboratory), Год журнала: 2024, Номер unknown

Опубликована: Сен. 5, 2024

Abstract Accurate prediction of post-operative adverse events following cardiothoracic surgery is crucial for timely interventions, potentially improving patient outcomes and reducing healthcare costs. By leveraging advanced deep learning techniques, this study highlights the transformative potential incorporating intraoperative variables into predictive analytics models to enhance postoperative care patients in ICU. We developed anticipating using a dataset from Society Thoracic Surgeons’ database ( 4 ) data. Our perform late fusion by integrating static data intra-operative time-series data, utilizing Fully Connected Neural Networks (FCNN) long short-term memory (LSTM) networks, respectively. The hybrid model was validated through five-fold cross-validation, demonstrating robust performance with mean AUC 0.93, Sensitivity 0.83 Specificity 0.89. This work represents significant step forward proactive management cardio thoracic ICU effectively predicting associated mortality post operative period.

Язык: Английский

---

Hybrid Learning Approach for Automated Identification and Categorization of Cardiovascular Disorders

International Journal of Electrical and Electronics Research, Год журнала: 2024, Номер 12(4), С. 1301 - 1323

Опубликована: Ноя. 30, 2024

ardio Vascular Diseases (CVDs) pose an important global health challenge, contributing substantially to mortality rates worldwide. Electrocardiography (ECG) is a necessary diagnostic tool in the detection of CVDs. Manual analysis by medical experts, for ECG interpretation, laborious and subject interobserver variability. To overcome these limitations, automated categorization technique has gained prominence, enabling efficient CVDs classification. The major focus this work utilize deep learning (DL) approach identification using signals. presented incorporates two hybrid models: one-dimensional convolutional neural network (1D-CNN) with Recurrent Hopfield Neural Network (1DCNN-RHNN) Residual (1D-CNN-ResNet), obtain features from raw data categorize them into different groups that correlate CVD situation. 1D-CNN-RHNN model achieved classification accuracy 96.62% 4-class normal, coronary artery disease (CAD), myocardial infarction (MI), congestive heart failure (CHF) 1DCNN-ResNet 95.75% 5-class CAD, MI, CHF cardiomyopathy. proposed model's functionality validated data, its outcomes are evaluated various measures. Experimental findings demonstrate models outperform other existing approaches categorizing multiple classes. Our suggested might potentially help doctors screen signals capable being verified larger databases.

Язык: Английский

---