A Systematic Review on Metaheuristic Optimization Techniques for Feature Selections in Disease Diagnosis: Open Issues and Challenges

A review of case study on different metaheuristic optimization techniques for disease detection and classification

This framework explores the use of metaheuristic optimization techniques for disease detection, specifically in image segmentation and feature selection to enhance classification performance. The study evaluates five swarm intelligence methods: Artificial Bee Colony (ABC) for image segmentation, Krill Herd Optimization (KHO) for both segmentation and feature selection, Particle Swarm Optimization (PSO) for feature selection, Grey Wolf Optimization (GWO) for feature selection, and Moth-Flame Optimization (MFO) for feature selection. Results demonstrate significant performance improvements, with accuracy increases of 0.9%, 2%, 2.3%, 2.1%, and 4.2%. These gains are attributed to optimized exploration/exploitation, enhanced diversity, and convergence, showing the effectiveness of metaheuristic techniques in disease detection.

FOTELP-VOX-OA: Enhancing radiotherapy planning precision with particle transport simulations and Optimization Algorithms

BACKGROUND AND OBJECTIVE

Accurate tumor targeting with minimal exposure to healthy tissue remains a significant challenge in radiotherapy. Modern techniques like Intensity-Modulated Radiation Therapy and stereotactic radiotherapy increasingly rely on detailed simulations and planning to achieve maximum treatment efficiency. Particle transport simulations play a crucial role in accurately modeling interactions between radiation and biological structures, providing a foundation for advancements in treatment planning. Building on this, FOTELP-VOX-OA is introduced as a novel framework designed to determine the optimal external-beam radiotherapy treatment plan. The primary aim of this study is to integrate the existing FOTELP-VOX framework with various Optimization Algorithms, focusing on estimating the parameters of interest that lead to the optimal radiation dose. While the framework itself is not pathology-specific, ocular melanoma is chosen as a test case due to its requirement for exceptionally precise dose delivery, given the small tumor volume and proximity of critical ocular structures.

METHODS

Particle transport simulations were conducted with FOTELP-VOX software, enabling detailed dose distribution analysis in tissues. Simulated conditions included a detailed biological model of eye melanoma to closely mimic clinical scenarios. The study integrates advanced optimization algorithms, such as Random Search, Tree-structured Parzen Estimator, and Genetic Algorithm, into the FOTELP-VOX framework, creating FOTELP-VOX-OA, to achieve the optimal treatment plan. Additionally, a specialized metric named Total Error was developed to determine the efficiency of the proposed treatment plan, focusing on both the desired tumor dose and minimizing exposure to surrounding tissues.

RESULTS

In the presented case-study, FOTELP-VOX-OA, utilizing the Genetic Algorithm, achieved a Total Error of 1701.52, significantly improving treatment planning compared to a human expert. However, this approach required the longest computation time among all methods. In contrast, the Tree-structured Parzen Estimator within the FOTELP-VOX-OA framework provided a balanced trade-off between speed and accuracy, while the Random Search-based solution was the fastest but also the least accurate.

CONCLUSION

The FOTELP-VOX-OA framework improves radiotherapy precision, reduces risks to surrounding healthy tissues, and achieves better treatment outcomes. This approach demonstrates how particle transport simulations, coupled with optimization techniques, can address critical challenges in radiotherapy planning, paving the way for future applications in other tumor sites and clinical contexts.

Improving human brucellosis susceptibility mapping using effective and simultaneously metaheuristic-based feature selection and hyperparameter tuning

Human Brucellosis, a neglected zoonotic disease, affects 1.6 to 2.1 million people globally each year. In Iran, it has become a significant health concern, with an average annual incidence rate of 19.91 cases per 100,000 people. This study aims to create a reliable Human Brucellosis Susceptibility Map (HBSM) for Mazandaran Province using a hybrid machine learning approach that enhances performance through metaheuristic algorithms for feature and hyperparameter optimization. A transformation function is integrated into these algorithms to reduce computational and time complexities while simultaneously executing feature selection and hyperparameter tuning. Additionally, a two-phase mutation operator is employed to improve the performance of feature selection. The results indicate that the hybrid model of Support Vector Regression-Transformation Mutation Grey Wolf Optimizer (SVR-TMGWO) outperformed other models, achieving RMSE=0.7723, MAE=0.614, MdAE=0.473, and R = 0.536. The predicted HBSM for 2018 identified 68 rural districts in Mazandaran Province as being in the High and Very High susceptibility classes. The susceptibility map can help decision-makers more effectively prevent, control, and manage Human Brucellosis in Mazandaran Province.

Application of Metaheuristics for Optimizing Predictive Models in iHealth: A Case Study on Hypotension Prediction in Dialysis Patients

Intradialytic hypotension (IDH) is a critical complication in patients with chronic kidney disease undergoing dialysis, affecting both patient safety and treatment efficacy. This study examines the application of advanced machine learning techniques, combined with metaheuristic optimization methods, to improve predictive models for intradialytic hypotension (IDH) in hemodialysis patients. Given the critical nature of IDH, which can lead to significant complications during dialysis, the development of effective predictive tools is vital for improving patient safety and outcomes. Dialysis session data from 758 patients collected between January 2016 and October 2019 were analyzed. Particle Swarm Optimization, Grey Wolf Optimizer, Pendulum Search Algorithm, and Whale Optimization Algorithm were employed to reduce the feature space, removing approximately 45% of clinical and analytical variables while maintaining high recall for the minority class of patients experiencing hypotension. Among the evaluated models, the XGBoost classifier showed superior performance, achieving a macro F-score of 0.745 with a recall of 0.756 and a precision of 0.718. These results highlight the effectiveness of the combined approach for early identification of patients at risk for IDH, minimizing false negatives, and improving clinical decision-making in nephrology.

Voice pathology detection using optimized convolutional neural networks and explainable artificial intelligence-based analysis

This article proposes a noninvasive computer-aided assessment approach based on optimized convolutional neural network for healthy and pathological voice detection. Firstly, the input voice samples are first transformed into mel-spectrogram time-frequency visual representations and fed for training the CNN model. The time-frequency image captures inherent speech variations beneficial for healthy and pathological voice sample detection. The weights and biases of trained CNN network are further optimized using artificial bee colony (ABC) optimization algorithm resulting in optimum CNN network employed for testing unseen data. The proposed approach is evaluated using three popular and publicly available datasets: SVD, AVPD and VOICED. Experimental results emphasize that proposed ABC optimized CNN model shows improved accuracy performance by 1.02% compared to conventional CNN network illustrating data-independent discriminative representation ability. Finally, gradient-weighted class activation mapping (Grad-CAM) explainable artificial intelligence (XAI) is utilized to make the decision understandable.

Automating cancer diagnosis using advanced deep learning techniques for multi-cancer image classification

Cancer detection poses a significant challenge for researchers and clinical experts due to its status as the leading cause of global mortality. Early detection is crucial, but traditional cancer detection methods often rely on invasive procedures and time-consuming analyses, creating a demand for more efficient and accurate solutions. This paper addresses these challenges by utilizing automated cancer detection through AI-based techniques, specifically focusing on deep learning models. Convolutional Neural Networks (CNNs), including DenseNet121, DenseNet201, Xception, InceptionV3, MobileNetV2, NASNetLarge, NASNetMobile, InceptionResNetV2, VGG19, and ResNet152V2, are evaluated on image datasets for seven types of cancer: brain, oral, breast, kidney, Acute Lymphocytic Leukemia, lung and colon, and cervical cancer. Initially, images undergo segmentation techniques, proceeded by contour feature extraction where parameters such as perimeter, area, and epsilon are computed. The models are rigorously evaluated, with DenseNet121 achieving the highest validation accuracy as 99.94%, 0.0017 as loss, and the lowest Root Mean Square Error (RMSE) values as 0.036056 for training and 0.045826 for validation. These results revealed the capability of AI-based techniques in improving cancer detection accuracy, with DenseNet121 emerging as the most effective model in this study.

Altruistic seagull optimization algorithm enables selection of radiomic features for predicting benign and malignant pulmonary nodules

Accurately differentiating indeterminate pulmonary nodules remains a significant challenge in clinical practice. This challenge becomes increasingly formidable when dealing with the vast radiomic features obtained from low-dose computed tomography, a lung cancer screening technique being rolling out in many areas of the world. Consequently, this study proposed the Altruistic Seagull Optimization Algorithm (AltSOA) for the selection of radiomic features in predicting the malignancy risk of pulmonary nodules. This innovative approach incorporated altruism into the traditional seagull optimization algorithm to seek a global optimal solution. A multi-objective fitness function was designed for training the pulmonary nodule prediction model, aiming to use fewer radiomic features while ensuring prediction performance. Among global radiomic features, the AltSOA identified 11 interested features, including the gray level co-occurrence matrix. This automatically selected panel of radiomic features enabled precise prediction (area under the curve = 0.8383 (95 % confidence interval 0.7862-0.8863)) of the malignancy risk of pulmonary nodules, surpassing the proficiency of radiologists. Furthermore, the interpretability, clinical utility, and generalizability of the pulmonary nodule prediction model were thoroughly discussed. All results consistently underscore the superiority of the AltSOA in predicting the malignancy risk of pulmonary nodules. And the proposed malignant risk prediction model for pulmonary nodules holds promise for enhancing existing lung cancer screening methods. The supporting source codes of this work can be found at: https://github.com/zzl2022/PBMPN.

Machine learning with multimodal neuroimaging data to classify stages of Alzheimer's disease: a systematic review and meta-analysis

In recent years, Alzheimer's disease (AD) has been a serious threat to human health. Researchers and clinicians alike encounter a significant obstacle when trying to accurately identify and classify AD stages. Several studies have shown that multimodal neuroimaging input can assist in providing valuable insights into the structural and functional changes in the brain related to AD. Machine learning (ML) algorithms can accurately categorize AD phases by identifying patterns and linkages in multimodal neuroimaging data using powerful computational methods. This study aims to assess the contribution of ML methods to the accurate classification of the stages of AD using multimodal neuroimaging data. A systematic search is carried out in IEEE Xplore, Science Direct/Elsevier, ACM DigitalLibrary, and PubMed databases with forward snowballing performed on Google Scholar. The quantitative analysis used 47 studies. The explainable analysis was performed on the classification algorithm and fusion methods used in the selected studies. The pooled sensitivity and specificity, including diagnostic efficiency, were evaluated by conducting a meta-analysis based on a bivariate model with the hierarchical summary receiver operating characteristics (ROC) curve of multimodal neuroimaging data and ML methods in the classification of AD stages. Wilcoxon signed-rank test is further used to statistically compare the accuracy scores of the existing models. With a 95% confidence interval of 78.87-87.71%, the combined sensitivity for separating participants with mild cognitive impairment (MCI) from healthy control (NC) participants was 83.77%; for separating participants with AD from NC, it was 94.60% (90.76%, 96.89%); for separating participants with progressive MCI (pMCI) from stable MCI (sMCI), it was 80.41% (74.73%, 85.06%). With a 95% confidence interval (78.87%, 87.71%), the Pooled sensitivity for distinguishing mild cognitive impairment (MCI) from healthy control (NC) participants was 83.77%, with a 95% confidence interval (90.76%, 96.89%), the Pooled sensitivity for distinguishing AD from NC was 94.60%, likewise (MCI) from healthy control (NC) participants was 83.77% progressive MCI (pMCI) from stable MCI (sMCI) was 80.41% (74.73%, 85.06%), and early MCI (EMCI) from NC was 86.63% (82.43%, 89.95%). Pooled specificity for differentiating MCI from NC was 79.16% (70.97%, 87.71%), AD from NC was 93.49% (91.60%, 94.90%), pMCI from sMCI was 81.44% (76.32%, 85.66%), and EMCI from NC was 85.68% (81.62%, 88.96%). The Wilcoxon signed rank test showed a low P-value across all the classification tasks. Multimodal neuroimaging data with ML is a promising future in classifying the stages of AD but more research is required to increase the validity of its application in clinical practice.

Hybrid whale algorithm with evolutionary strategies and filtering for high-dimensional optimization: Application to microarray cancer data

The standard whale algorithm is prone to suboptimal results and inefficiencies in high-dimensional search spaces. Therefore, examining the whale optimization algorithm components is critical. The computer-generated initial populations often exhibit an uneven distribution in the solution space, leading to low diversity. We propose a fusion of this algorithm with a discrete recombinant evolutionary strategy to enhance initialization diversity. We conduct simulation experiments and compare the proposed algorithm with the original WOA on thirteen benchmark test functions. Simulation experiments on unimodal or multimodal benchmarks verified the better performance of the proposed RESHWOA, such as accuracy, minimum mean, and low standard deviation rate. Furthermore, we performed two data reduction techniques, Bhattacharya distance and signal-to-noise ratio. Support Vector Machine (SVM) excels in dealing with high-dimensional datasets and numerical features. When users optimize the parameters, they can significantly improve the SVM's performance, even though it already works well with its default settings. We applied RESHWOA and WOA methods on six microarray cancer datasets to optimize the SVM parameters. The exhaustive examination and detailed results demonstrate that the new structure has addressed WOA's main shortcomings. We conclude that the proposed RESHWOA performed significantly better than the WOA.

Novel statistical time series data augmentation and machine learning based classification of unobtrusive respiration data for respiration Digital Twin model

Digital Twin (DT), a concept of Healthcare (4.0), represents the subject's biological properties and characteristics in a digital model. DT can help in monitoring respiratory failures, enabling timely interventions, personalized treatment plans to improve healthcare, and decision-support for healthcare professionals. Large-scale implementation of DT technology requires extensive patient data for accurate monitoring and decision-making with Machine Learning (ML) and Deep Learning (DL). Initial respiration data was collected unobtrusively with the ESP32 Wi-Fi Channel State Information (CSI) sensor. Due to limited respiration data availability, the paper proposes a novel statistical time series data augmentation method for generating larger synthetic respiration data. To ensure accuracy and validity in the augmentation method, correlation methods (Pearson, Spearman, and Kendall) are implemented to provide a comparative analysis of experimental and synthetic datasets. Data processing methodologies of denoising (smoothing and filtering) and dimensionality reduction with Principal Component Analysis (PCA) are implemented to estimate a patient's Breaths Per Minute (BPM) from raw respiration sensor data and the synthetic version. The methodology provided the BPM estimation accuracy of 92.3% from raw respiration data. It was observed that out of 27 supervised classifications with k-fold cross-validation, the Bagged Tree ensemble algorithm provided the best ML-supervised classification. In the case of binary-class and multi-class, the Bagged Tree ensemble showed accuracies of 89.2% and 83.7% respectively with combined real and synthetic respiration dataset with the larger synthetic dataset. Overall, this provides a blueprint of methodologies for the development of the respiration DT model.

A Comprehensive Analysis of Deep Learning-Based Approaches for Prediction and Prognosis of Infectious Diseases

Artificial intelligence is the most powerful and promising tool for the present analytic technologies. It can provide real-time insights into disease spread and predict new pandemic epicenters by processing massive amount of data. The main aim of the paper is to detect and classify multiple infectious diseases using deep learning models. The work is conducted by using 29,252 images of COVID-19, Middle East Respiratory Syndrome Coronavirus, Pneumonia, normal, Severe Acute Respiratory Syndrome, tuberculosis, viral pneumonia, and lung opacity which has been collected from various disease datasets. These datasets are used to train the deep learning models such as EfficientNetB0, EfficientNetB1, EfficientNetB2, EfficientNetB3, NASNetLarge, DenseNet169, ResNet152V2, and InceptionResNetV2. The images have been initially graphically represented using exploratory data analysis to study the pixel intensity and find anomalies by extracting the color channels in an RGB histogram. Later, the dataset has been pre-processed to remove noisy signals using image augmentation and contrast enhancement techniques. Further, feature extraction techniques such as morphological values of contour features and Otsu thresholding have been applied to extract the feature. The models have been evaluated on the basis of various parameters, and it has been discovered that during the testing phase, the InceptionResNetV2 model generated the highest accuracy of 88%, best loss value of 0.399, and root mean square error of 0.63.

A Systematic Review of the Whale Optimization Algorithm: Theoretical Foundation, Improvements, and Hybridizations

Despite the simplicity of the whale optimization algorithm (WOA) and its success in solving some optimization problems, it faces many issues. Thus, WOA has attracted scholars' attention, and researchers frequently prefer to employ and improve it to address real-world application optimization problems. As a result, many WOA variations have been developed, usually using two main approaches improvement and hybridization. However, no comprehensive study critically reviews and analyzes WOA and its variants to find effective techniques and algorithms and develop more successful variants. Therefore, in this paper, first, the WOA is critically analyzed, then the last 5 years' developments of WOA are systematically reviewed. To do this, a new adapted PRISMA methodology is introduced to select eligible papers, including three main stages: identification, evaluation, and reporting. The evaluation stage was improved using three screening steps and strict inclusion criteria to select a reasonable number of eligible papers. Ultimately, 59 improved WOA and 57 hybrid WOA variants published by reputable publishers, including Springer, Elsevier, and IEEE, were selected as eligible papers. Effective techniques for improving and successful algorithms for hybridizing eligible WOA variants are described. The eligible WOA are reviewed in continuous, binary, single-objective, and multi/many-objective categories. The distribution of eligible WOA variants regarding their publisher, journal, application, and authors' country was visualized. It is also concluded that most papers in this area lack a comprehensive comparison with previous WOA variants and are usually compared only with other algorithms. Finally, some future directions are suggested.