Search Results (214)

Wireless sensor networks (WSNs) contend with the critical challenge of balancing energy conservation against data transmission delay, a trade-off that protocols such as PEGASIS—while being strong in energy efficiency—fail to manage optimally due to resulting high latency, unbalanced load distribution, and suboptimal cluster formation. To address these limitations, this paper introduces the Enhanced Multi-Objective PEGASIS (EMO-PEGASIS) protocol, which is designed and implemented using a dual-phase machine learning strategy. This multi-objective approach works in two stages. First, it utilises K-means clustering to achieve robust spatial partitioning of the network. Second, it employs K-Nearest Neighbours (K-NN) classification to enable adaptive and intelligent routing. The simulation was performed using MATLAB R2025a, and the results show that EMO-PEGASIS addresses this multi-objective optimisation problem. The proposed EMO-PEGASIS protocol achieves a 45% reduction in average energy consumption, a 38% decrease in end-to-end delay, and a 67% increase in network lifetime compared to the original PEGASIS protocol. Additionally, EMO-PEGASIS demonstrates enhanced stability and effective load balancing under heterogeneous network configurations, while maintaining an excellent packet delivery ratio of 96.8%. These findings underscore the effectiveness of integrating machine learning techniques, which ultimately yield enhanced performance and enable reliable multi-objective optimisation within energy- and delay-constrained WSN environments. Full article

Background/Objectives: Missing data in clinical observational studies, such as out-of-hospital cardiac arrest (OHCA) registries, can compromise statistical validity. Single imputation methods are simple alternatives to complete-case analysis (CCA) but do not account for imputation uncertainty. Multiple imputation (MI) is the standard for handling missing-at-random (MAR) data, yet its implementation remains challenging. This study evaluated the performance of MI in association analysis compared with CCA and single imputation methods. Methods: Using a simulation framework with real-world Singapore OHCA registry data (N = 13,274 complete cases), we artificially introduced 20%, 30%, and 40% missingness under MAR. MI was implemented using predictive mean matching (PMM), random forest (RF), and classification and regression trees (CART) algorithms, with 5–20 imputations. Performance was assessed based on bias and precision in a logistic regression model evaluating the association between alert issuance and bystander CPR. Results: CART outperformed PMM, providing more accurate β coefficients and stable CIs across missingness levels. Although K-Nearest Neighbours (KNN) produced similar point estimates, it underestimated imputation uncertainty. PMM showed larger bias, wider and less stable CIs, and in some settings performed similarly to CCA. MI methods produced wider CIs than single imputation, appropriately capturing imputation uncertainty. Increasing the number of imputations had minimal impact on point estimates but modestly narrowed CIs. Conclusions: MI performance depends strongly on the chosen algorithm. CART and RF methods offered the most robust and consistent results for OHCA data, whereas PMM may not be optimal and should be selected with caution. MI using tree-based methods (CART/RF) remains the preferred strategy for generating reliable conclusions in OHCA research. Full article

Environmental monitoring systems generate large volumes of multivariate time series data from heterogeneous sensors, including those measuring soil, weather, and air quality parameters. However, sensor malfunctions and transmission failures frequently lead to missing values, compromising the performance of downstream analytical and predictive models. To address this challenge, this study presents a comprehensive and systematic evaluation of previously proposed hybrid architecture that interleaves Long Short-Term Memory (LSTM) layers with a Multi-Head Attention mechanism in a “sandwiched” setting (LSTM–Attention–LSTM) for robust multivariate data imputation in environmental IoT datasets. The first LSTM layer captures short-term temporal dependencies, the attention layer emphasises long-range relationships among correlated features, and the second LSTM layer re-integrates these enriched representations into a coherent temporal sequence. The model is evaluated using multiple environmental datasets of soil temperature, meteorological (precipitation, temperature, wind speed, humidity), and air quality data across missingness levels ranging from 10% to 90%. Performance is compared against baseline methods, including K-Nearest Neighbour (KNN) and Bidirectional Recurrent Imputation for Time Series (BRITS). Across all datasets, the Hybrid model consistently outperforms baseline methods, achieving MAE reductions exceeding 50% and reaching over 80% in several scenarios, along with RMSE reductions of up to approximately 85%, particularly under moderate to high missingness conditions. An ablation study further examines the contribution of each layer to overall model performance. Results demonstrate that the proposed Hybrid model achieves superior accuracy and robustness across datasets, confirming its effectiveness for environmental sensor data imputation under varying missing data conditions. Full article

Landslide risk assessment along high-altitude transportation corridors is critical for infrastructure resilience. This study presents an integrated framework combining Small Baseline Subset Interferometric Synthetic Aperture Radar (SBAS-InSAR) deformation data and machine learning (ML) to systematically identify and assess landslide susceptibility along the entire 245.5 km West Kunlun Highway. We first compiled a landslide inventory through visual interpretation and SBAS-InSAR analysis. Subsequently, fourteen causative factors were selected to construct and compare six ML models: random forest (RF), K-nearest neighbours (KNN), artificial neural network (ANN), gradient boosting decision trees (GBDT), support vector machine (SVM), and logistical regression (LR). Research findings indicate that along the Hotan–Kangziva Highway in the Western Kunlun Mountains, there exist 21 potential risk points for small-scale landslides, 12 for medium-scale landslides, and 5 for large-scale landslides, with hazard identification accuracy reaching 80%. The random forest model demonstrated outstanding performance, classifying areas with 5.10%, 4.55% and 4.96% probability as extremely high, high and medium susceptibility, respectively. This work provides a robust methodology and a high-accuracy assessment tool for landslide risk management in the data-scarce Western Kunlun Mountains. Full article

The spice Ferula assa-foetida L., also known as asafoetida, is widely recognized for its medicinal and culinary applications. The non-native status of the plant and the prolonged dormancy of its seeds pose significant challenges for large-scale cultivation in India. In vitro organogenesis offers an effective solution to these obstacles. Establishing reliable in vitro regeneration protocols requires standardized statistical methods to evaluate univariate and multivariate data for optimizing specific traits. However, these methods have limitations when handling complex, nonlinear inputs, often producing large prediction errors that reduce the reliability of trait optimization. This study developed an in vitro regeneration system for F. assa-foetida L. and identified optimal PGRs for somatic embryogenesis and shoot organogenesis through image-based morphological analysis. Predictive models were created using DL and ML algorithms. Calli induced from leaf explants was cultured on the Murashige and Skoog medium supplemented with various combinations and concentrations of thidiazuron (TDZ), 6-benzylaminopurine (BAP), and α-naphthaleneacetic acid (NAA), as experimental variables. Seven ML approaches, namely random forest (RF), support vector machine (SVM), k-nearest neighbours (kNN), decision tree (DT), extreme gradient boosting (XG Boost), naïve bayes, and logistic regression, alongside five DL models—convolutional neural network (CNN), MobileNet, region-based convolutional neural network (RCNN), residual neural network (ResNet), and visual geometry group (VGG19)—were employed to predict the best PGRs for somatic embryogenesis and shoot organogenesis. Among them, the convolutional neural network (CNN) achieved the highest accuracy (87%), outperforming baseline ML models such as logistic regression and decision tree (82%). This pioneering study in F. assa-foetida L. presents an AI-driven, image-based framework for predicting optimal PGRs, offering a scalable approach to enhance micropropagation in endangered medicinal plants. Full article

Land surface temperature (LST) is a critical variable for understanding energy exchanges and water balance at the Earth’s surface, as well as for calculating turbulent heat flux and long-wave radiation at the surface–atmosphere interface. Remote sensing techniques, particularly using satellite platforms like Landsat 8 OLI/TIRS and Sentinel-2A, have facilitated detailed LST mapping. Sentinel-2 offers high spatial and temporal resolution multispectral data, but it lacks thermal infrared bands, which Landsat 8 can provide a 30 m resolution with less frequent revisits compared to Sentinel-2. This study employs Sentinel-2 spectral indices as independent variables and Landsat 8-derived LST data as the target variable within a machine-learning framework, enabling LST prediction at a 10 m resolution. This method applies grid search-based hyperparameter-tuned machine learning algorithms—Random Forest (RF), Gradient Boosting Machine (GBM), Support Vector Machine (SVM), and k-Nearest Neighbours (kNN)—to model complex nonlinear relationships between the spectral indices (NDVI, NDWI, NDBI, and BSI) and LST. Grid search, combined with cross-validation, enhanced the model’s prediction accuracy for both pre- and post-monsoon seasons. This approach surpasses earlier methods that either employed untuned models or failed to integrate Sentinel-2 data. This study demonstrates that capturing urban thermal dynamics at fine spatial and temporal scales, combined with tuned machine learning models, can enhance the capability of urban heat island monitoring, climate adaptation planning, and sustainable environmental management models. Full article

Treatment-induced ovarian function loss is a significant concern for many young patients with breast cancer. Accurately predicting this risk is crucial for counselling young patients and informing their fertility-related decision-making. However, current risk prediction models for treatment-related ovarian function loss have limitations. To provide a broader representation of patient cohorts and improve feature selection, we combined retrospective data from six datasets within the FoRECAsT (Infertility after Cancer Predictor) databank, including 2679 pre-menopausal women diagnosed with breast cancer. This combined dataset presented notable missingness, prompting us to employ cross imputation using the k-nearest neighbours (KNN) machine learning (ML) algorithm. Employing Lasso regression, we developed an ML model to forecast the risk of treatment-related amenorrhea as a surrogate marker of ovarian function loss at 12 months after starting chemotherapy. Our model identified 20 variables significantly associated with risk of developing amenorrhea. Internal validation resulted in an area under the receiver operating characteristic curve (AUC) of 0.820 (95% CI: 0.817–0.823), while external validation with another dataset demonstrated an AUC of 0.743 (95% CI: 0.666–0.818). A cutoff of 0.20 was chosen to achieve higher sensitivity in validation, as false negatives—patients incorrectly classified as likely to regain menses—could miss timely opportunities for fertility preservation if desired. At this threshold, internal validation yielded sensitivity and precision rates of 91.3% and 61.7%, respectively, while external validation showed 92.9% and 60.0%. Leveraging ML methodologies, we not only devised a model for personalised risk prediction of amenorrhea, demonstrating substantial enhancements over existing models but also showcased a robust framework for maximally harnessing available data sources. Full article

Rapid sand filtration is a critical step in the water treatment process, as its effectiveness directly impacts the supply of safe drinking water. However, optimising filtration processes in water treatment plants (WTPs) presents a significant challenge due to the varying operational parameters and conditions. This study applies explainable machine learning to enhance insights into predicting direct filtration operations at the Ålesund WTP in Norway. Three baseline models (Multiple Linear Regression, Support Vector Regression, and K-Nearest Neighbour (KNN)) and three ensemble models (Random Forest (RF), Extra Trees (ET), and XGBoost) were optimised using the GridSearchCV algorithm and implemented on seven filter units to predict their filtered water turbidity. The results indicate that ML models can reliably predict filtered water turbidity in WTPs, with Extra Trees models achieving the highest predictive performance (R² = 0.92). ET, RF, and KNN ranked as the three top-performing models using Alternative Technique for Order of Preference by Similarity to Ideal Solution (A-TOPSIS) ranking for the suite of algorithms used. The feature importance analysis ranked the filter runtime, flow rate, and bed level. SHAP interpretation of the best model provided actionable insights, revealing how operational adjustments during the ripening stage can help mitigate filter breakthroughs. These findings offer valuable guidance for plant operators and highlight the benefits of explainable machine learning in water quality management. Full article

Poor-quality milk and the use of chemicals in it can lead to serious health problems, including various diseases and, in some cases, even death for those who consume it. In our society, using such products or contaminated milk that contains chemicals or is of bad quality, often with water or other adulterants, is very common. Based on previous research and existing models, we have improved the process to better and more accurately predict milk quality by using a voting system. This system uses four different algorithms: KNN (K Nearest Neighbour), Naïve Bayes, Random Forest, and Decision Tree. We applied these models to a dataset with almost 1000 samples. To enhance performance, we used brute-force feature selection and a voting process to make accurate decisions. All these procedures were implemented in RapidMiner Studio, resulting in an overall accuracy of 99.69%. Full article

Background and Objectives: Sarcopenia is a progressive loss of skeletal muscle mass and function in elderly adults, posing a significant risk of frailty, falls, and morbidity. The current study designs and evaluates SarcoNet, a novel artificial neural network (ANN)-based classification framework developed in order to classify Sarcopenic from non-Sarcopenic subjects using a comprehensive real-time dataset. Methods: This pilot study involved 30 subjects, who were divided into Sarcopenic and non-Sarcopenic groups based on physician assessment. The collected dataset consists of thirty-one clinical parameters like skeletal muscle mass, which is collected using various equipment such as Body Composition Analyser, along with ten kinetic features which are derived from video-based gait analysis of joint angles obtained during walking on three terrain types such as slope, steps, and parallel path. The performance of the designed ANN-based SarcoNet was benchmarked against the traditional machine learning classifiers utilised including Support Vector Machine (SVM), k-Nearest Neighbours (k-NN), and Random Forest (RF), as well as hard and soft voting ensemble classifiers. Results: SarcoNet achieved the highest overall classification accuracy of about 94%, with a specificity and precision of about 100%, an F1-score of about 92.4%, and an AUC of 0.94, outperforming all other models. The incorporation of lower-limb joint kinetics such as knee flexion, extension, ankle plantarflexion and dorsiflexion significantly enhanced predictive capability of the model and thus reflecting the functional deterioration characteristic of muscles in Sarcopenia. Conclusions: SarcoNet provides a promising AI-driven solution in Sarcopenia diagnosis, especially in low-resource healthcare settings. Future work will focus on improving the dataset, validating the model across diverse populations, and incorporating explainable AI to improve clinical adoption. Full article

The acquisition of a person’s first language is one of their greatest accomplishments. Nevertheless, being fluent in sign language presents challenges for many deaf students who rely on it for communication. Effective communication is essential for both personal and professional interactions and is critical for community engagement. However, the lack of a mutually understood language can be a significant barrier. Estimates indicate that a large portion of New Zealand’s disability population is deaf, with an educational approach predominantly focused on oralism, emphasizing spoken language. This makes it essential to bridge the communication gap between the general public and individuals with speech difficulties. The aim of this project is to develop an application that systematically cycles through each letter and number in New Zealand Sign Language (NZSL), assessing the user’s proficiency. This research investigates various machine learning methods for hand gesture recognition, with a focus on landmark detection. In computer vision, identifying specific points on an object—such as distinct hand landmarks—is a standard approach for feature extraction. Evaluation of this system has been performed using machine learning techniques, including Random Forest (RF) Classifier, k-Nearest Neighbours (KNN), AdaBoost (AB), Naïve Bayes (NB), Support Vector Machine (SVM), Decision Trees (DT), and Logistic Regression (LR). The dataset used for model training and testing consists of approximately 100,000 hand gesture expressions, formatted into a CSV dataset for model training. Full article

Background: Differentiating Parkinson’s disease (PD) from essential tremor (ET) is often challenging, especially in early or atypical cases. Dopamine transporter (DAT) single-photon emission computed tomography (SPECT) with ¹²³I-Ioflupane supports diagnosis, and semi-quantitative tools such as DaTQUANT^® and BasGanV2™ provide objective measures. This study compared their diagnostic performance when integrated with supervised machine learning. Methods: We retrospectively analysed ¹²³I-Ioflupane SPECT scans from 169 patients (133 PD, 36 ET). Semi-quantitative analysis was performed using DaTQUANT^® v2.0 and BasGanV2™ v.2. Classification tree (ClT), k-nearest neighbour (k-NN), and support vector machine (SVM) models were trained and validated with stratified shuffle split (250 iterations). Diagnostic accuracy was compared between the two software packages. Results: All classifiers reliably distinguished PD from ET. DaTQUANT^® consistently achieved higher accuracy than BasGanV2™: 93.8%, 93.2%, and 94.5% for ClT, k-NN, and SVM, respectively, versus 90.9%, 91.7%, and 91.9% for BasGanV2™ (p < 0.001). Sensitivity and specificity were also consistently higher for DaTQUANT^® than BasGanV2. Class imbalance (PD > ET) was addressed using Synthetic Minority Over-sampling Technique (SMOTE). Conclusions: Machine learning analysis of ¹²³I-Ioflupane SPECT enhances differentiation between PD and ET. DaTQUANT^® outperformed BasGanV2™, suggesting greater suitability for AI-driven decision support. These findings support the integration of semi-quantitative and AI-based approaches into clinical workflows and highlight the need for harmonised methodologies in movement disorder imaging. Full article

Air quality significantly affects human health, productivity, and overall well-being. This study applies machine learning techniques to analyse and predict air quality in Hamilton, New Zealand, focusing on particulate matter (PM2.5 and PM10) and environmental factors such as temperature, humidity, wind speed, and wind direction. Data were collected from two monitoring sites (Claudelands and Rotokauri) to explore relationships between variables and evaluate the performance of different predictive models. First, the unsupervised k-means clustering algorithm was used to categorise air quality levels based on data from one or both locations. These cluster labels were then used as target variables in supervised learning models, including random forests, decision trees, support vector machines, and k-nearest neighbours. Model performance was assessed by comparing prediction accuracy for air quality at either Claudelands or Rotokauri. Results show that the random forest (93.6%) and decision tree (91.8%) models outperformed k-nearest neighbours (KNN, 83%) and support vector machine (SVM, 61%) in predicting air quality clusters derived from k-means analysis. The three clusters (very good, good, and moderate) reflected seasonal and urban–semi-urban gradients, while cross-location validation confirmed that models trained at Claudelands generalised effectively to Rotokauri, demonstrating scalability for regional air quality forecasting. These findings highlight the potential of combining clustering with supervised learning to improve air quality predictions. Such methods could support environmental monitoring and inform strategies for mitigating pollution-related health risks in New Zealand cities and beyond. Full article

Background/Objectives: Clinical observational studies often encounter missing data, which complicates association evaluation with reduced bias while accounting for confounders. This is particularly challenging in multi-national registries such as those for out-of-hospital cardiac arrest (OHCA), a time-sensitive medical emergency with low survival rates. While various methods for handling missing data exist, observational studies frequently rely on complete-case analysis, limiting representativeness and potentially introducing bias. Our objective was to evaluate the impact of various single imputation methods on association analysis with OHCA registries. Methods: Using a complete dataset (N = 13,274) from the Pan-Asian Resuscitation Outcomes Study (PAROS) registry (1 January 2016–31 December 2020) as reference, we intentionally introduced missing values into selected variables via a Missing At Random (MAR) mechanism. We then compared statistical and machine learning (ML) single imputation methods to assess the association between bystander cardiopulmonary resuscitation (BCPR) and the issuance of a mobile app alert, adjusting for confounders. The impacts of complete-case analysis (CCA) and single imputation methods on conclusions in OHCA research were evaluated. Results: CCA was suboptimal for handling MAR data, resulting in more biased estimates and wider confidence intervals compared to single imputation methods. The missingness-indicator (MxI) method offered a trade-off between bias and ease of implementation. The K-Nearest Neighbours (KNN) method outperformed other imputation approaches, whereas missForest introduced bias under certain conditions. Conclusions: KNN and MxI are easy to use and better alternatives to CCA for reducing bias in observational studies. This study highlights the importance of selecting appropriate imputation methods to ensure reliable conclusions in OHCA research and has broader implications for other registries facing similar missing data challenges. Full article

Background/Objectives: Age-related macular degeneration (AMD) is the leading cause of visual impairment among the elderly. Optical coherence tomography angiography (OCTA) is a non-invasive imaging modality that enables detailed visualisation of retinal vascular layers. However, clinical assessment of OCTA images is often challenging due to high data volume, pattern variability, and subtle abnormalities. This study aimed to develop automated algorithms to detect and quantify AMD in OCTA images, thereby reducing ophthalmologists’ workload and enhancing diagnostic accuracy. Methods: Two texture-based algorithms were developed to classify OCTA images without relying on segmentation. The first algorithm used whole local texture features, while the second applied principal component analysis ($PCA$) to decorrelate and reduce texture features. Local texture descriptors, including rotation-invariant uniform local binary patterns ($LB{P}^{2riu}$), local binary patterns ($LBP$), and binary robust independent elementary features ($BRIEF$), were combined with machine learning classifiers such as support vector machine (SVM) and K-nearest neighbour (KNN). OCTA datasets from Manchester Royal Eye Hospital and Moorfields Eye Hospital, covering healthy, dry AMD, and wet AMD eyes, were used for evaluation. Results: The first algorithm achieved a mean area under the receiver operating characteristic curve (AUC) of $1.00\pm 0.00$ for distinguishing healthy eyes from wet AMD. The second algorithm showed superior performance in differentiating dry AMD from wet AMD (AUC $0.85\pm 0.02$). Conclusions: The proposed algorithms demonstrate strong potential for rapid and accurate AMD diagnosis in OCTA workflows. By reducing manual image evaluation and associated variability, they may support improved clinical decision-making and patient care. Full article