Intelligent anomaly detection in medical IoT systems using Coral Reefs Optimization and Aquila enhanced CatBoost

The rapid growth of metropolitan populations and large public gatherings has increased the need for intelligent surveillance systems capable of ensuring public safety. This study proposes a novel framework, Intelligent Anomaly Detection and Localization with Crowd Density Estimation (IADL-CDE), which jointly addresses crowd density classification and anomaly detection in real-time. Leveraging the ShanghaiTech dataset for crowd density estimation and the UCSD Pedestrian Database for anomaly detection, the framework integrates a Robust Maximum Correntropy Kalman Filter (RMCKF) for effective image pre-processing, followed by a Self-Modulating Convolutional Neural Network (SMCNN) that accurately classifies crowd density levels into sparse, medium, and dense categories. Subsequently, a Pixel-Associated Auto Encoder (PAAE) is employed to detect and localize anomalies such as cycling, skateboarding, and unauthorized vehicles in pedestrian environments. Experimental results demonstrate that the proposed IADL-CDE model achieves significant improvements in accuracy, precision, recall, F1-Score and Cohen’s Kappa Coefficient compared to the baseline methods. These findings highlight the model’s effectiveness and robustness in enhancing public safety through intelligent crowd monitoring and anomaly detection.

Extreme learning machine (ELM) is a novel and fast learning method to train single layer feed-forward networks. However due to the demand for larger number of hidden neurons, the prediction speed of ELM is not fast enough. An evolutionary based ELM with differential evolution (DE) has been proposed to reduce the prediction time of original ELM. But it may still get stuck at local optima. In this paper, a novel algorithm hybridizing DE and metaheuristic coral reef optimization (CRO), which is called differential evolution coral reef optimization (DECRO), is proposed to balance the explorative power and exploitive power to reach better performance. The thought and the implement of DECRO algorithm are discussed in this article with detail. DE, CRO and DECRO are applied to ELM training respectively. Experimental results show that DECRO-ELM can reduce the prediction time of original ELM, and obtain better performance for training ELM than both DE and CRO.

Image registration is process of alignment of two or more images. It is used to get more information about the obtained image which is useful for the analysis of the disease. In the proposed method we use three different types of image registration. They are demon registration, image registration by mutual information and difference method. In these methods the alignment is achieved by changing some parameters manually. The degree of alignment of two images is directly proportional to the amount of information obtained. In order to maximize the alignment an optimization algorithm is used. The conventional image registration methods are constrained by many limitations. Hence we use a bio-inspired meta-heuristics and high performance Coral Reef optimization with Substrate Layer (CRO-SL) algorithm. CRO-SL is an advanced method of coral reef optimization based on natural behavior of coral reef. The image registration process comprises of various steps like transformation of registering image, evaluation of performance metrics and finding the optimized value for transformation. A uni-model affine transformation is used in the proposed method. The experimental results show that CRO-SL is a very efficient approach in case of alignment of image in higher degree.

It is important to ensure the proper functioning of any given system since every continuous process of tracking or accumulating data forms a dataset which is essential in any field of study. The main goal of this project is to develop an approach that detects anomalies without providing any labeled training data. The two techniques of machine learning that have been considered in this work are Isolation Forest and DBSCAN. The process begins with data cleaning and scaling before applying outlier detection using Isolation Forest. After identifying the outliers, the remaining data within each cluster is considered normal, while everything outside is deemed anomalous. Outlier detection is performed using a modified version of DBSCAN, and the quality of the clusters created is evaluated by calculating the Silhouette score. Different models were visually compared in the synthetic data, and images showed various forms of how the data was processed and the results analyzed. Findings revealed that the developed technique is useful in detecting anomalies in large complicated datasets and in various real-world scenarios such as fraudulent activities, security systems, and monitoring systems in intelligent urban environments.

The accurate prediction of the friction angle of clays is crucial for assessing slope stability in engineering applications. This study addresses the importance of estimating the friction angle and presents the development of four soft computing models: YJ-FPA-MLPnet, YJ-CRO-MLPnet, YJ-ACOC-MLPnet, and YJ-CSA-MLPnet. First of all, the Yeo-Johnson (YJ) transformation technique was used to stabilize the variance of data and make it more suitable for parametric statistical models that assume normality and equal variances. This technique is expected to improve the accuracy of friction angle prediction models. The friction angle prediction models then utilized multi-layer perceptron neural networks (MLPnet) and metaheuristic optimization algorithms to further enhance performance, including flower pollination algorithm (FPA), coral reefs optimization (CRO), ant colony optimization continuous (ACOC), and cuckoo search algorithm (CSA). The prediction models without the YJ technique, i.e. FPA-MLPnet, CRO-MLPnet, ACOC-MLPnet, and CSA-MLPnet, were then compared to those with the YJ technique, i.e. YJ-FPA-MLPnet, YJ-CRO-MLPnet, YJ-ACOC-MLPnet, and YJ-CSA-MLPnet. Among these, the YJ-CRO-MLPnet model demonstrated superior reliability, achieving an accuracy of up to 83% in predicting the friction angle of clay in practical engineering scenarios. This improvement is significant, as it represents an increase from 1.3% to approximately 20% compared to the models that did not utilize the YJ transformation technique.

The article discusses the possibilities of using medical diagnostic equipment for measuring blood pressure, heart rate, saturation and body temperature, purchased within the framework of the Horizon Europe international project «WIDE AcrossEU» (101158561), to improve the effectiveness of the educational process and conduct scientific research in the specialty G22 «Biomedical Engineering». The experience of using the ChoiceMMed medical monitor in laboratory work, training courses, student project activities and during scientific research is analysed. The prospects for the use of the equipment in interdisciplinary educational programmes, in particular in the fields of telemedicine, medical IoT systems, biomedical signal processing, modelling of adaptive responses of the body and the creation of intelligent health monitoring systems, have been identified. The role of modern diagnostic equipment in shaping the competencies of future biomedical engineers and developing the scientific and educational infrastructure of higher education institutions has been outlined.

Relational Database Management Systems (RDBMSs) are foundational to operations across finance, healthcare, and government sectors. However, they remain susceptible to advanced threats such as polymorphic SQL injection, insider misuse, and timing-based data exfiltration.Traditional intrusion detection systems (IDSs), including signature-based and static machine learning models, often struggle to detect evolving and obfuscated attacks, resulting in high false-positive rates and limited adaptability.This paper presents the Triple Loop Learning (TLL) framework—a recursive, multi-tiered architecture that integrates supervised learning, reinforcement learning (RL), and meta-learning for real-time anomaly detection in SQL-based environments. To the best of our knowledge, TLL is the first framework to unify these three learning paradigms into a recursive, explainable, and low-latency anomaly detection system specifically tailored for structured SQL workloads.TLL comprises three interdependent loops: (i) an operational loop utilizing SQL-aware autoencoders for fine-grained anomaly detection, (ii) a tactical loop leveraging Deep Q-Networks (DQN) for adaptive threshold calibration, and (iii) a strategic loop based on Model-Agnostic Meta-Learning (MAML) for rapid cross-domain adaptation.TLL was evaluated on a composite dataset comprising the UNSW-NB15 benchmark, anonymized SQL logs from financial and healthcare institutions, and adversarial samples generated with MAD-GAN. The framework achieved an F1 score of 0.94, an AUC of 0.976, and an average inference latency of 84 ms.The results demonstrate that TLL enables robust, low-latency detection of zero-day and obfuscated SQL threats. Its explain-ability is enhanced through SHAP-based attributions, and its deployment is version-controlled and reproducible, ensuring traceability, compliance, and production readiness.

Telemedicine for chronic disease management is extending to the home due to the coronavirus pandemic and patients with intractable diseases, such as amyotrophic lateral sclerosis (ALS) disease, are returning to their homes. Sophisticated alarm delivery for condition changes of patients and burden reduction of their families are foundations of telemedicine. This paper discusses reliable and flexible alarm delivery system for caregivers including patient's family and introduces a buddy system which manage clinically necessary alarms from highly managed medical devices in the system. We discuss alarm analysis and reduction, zero-admin security, 920MHz radio integration, and degeneration operation in alarm delivery. We design a buddy system and an alarm delivery system taking advantage of the buddy and evaluate the alarm traffic when 920MHz wireless network is used. We, then, investigate the system with multi-hop 920MHz configuration intended for floor-wide alarm delivery system and raise issues of the system for further works. We conclude with our further work for interface standardization of medical buddy IoT systems.

Coral Reef Optimization with substrate layers for medical Image Registration

Heatwaves heavily affect European public health, society and economy. A full understanding of the drivers behind the occurrence and intensity of heatwaves (HW) is one of the priorities of H2020 CLimate INTelligence (CLINT) project. Particular attention is given to the detection and attribution of HWs in future climate projections. However, it is important to assess the capability of climate models to thoroughly describe relationships between the drivers and the occurrence and intensity of HWs. For this reason, a feature selection framework, based on the Coral Reef Optimization (Salcedo-Sanz et al., 2014) has been developed. This has been applied to ERA5 summer data, using as a target the Lake Como HW occurrence and, as candidate predictors, time series of weather variables calculated on clustered areas on European and global scales. The same algorithm has been applied to historical climate simulations included in CMIP6. The comparison of the results of these two steps has first focused on the similarities in maximum temperature and HW trends in the target region of Lake Como. Then, the selected drivers in each historical climate simulation have been evaluated, using ERA5 results as benchmark. Thanks to this, the models that better resemble the statistical properties and teleconnections described by the reanalysis have been identified. This set of models will be considered for the detection and attribution of heatwaves in future climate projections under different emission scenarios. In this upcoming phase the goal will be to analyse changes in the relationship between the drivers and HW occurrence and intensity, giving an insight about possible future evolutions in heatwaves frequency and magnitude.

Hybridizing machine learning with metaheuristics for preventing convergence failures in mechanical models based on compression field theories

This article presents a study on the optimization of parameters of a solid oxide fuel cell (SOFC) by using seven different metaheuristic-based optimization techniques such as Differential Evolutionary algorithm (DE), Genetic Algorithm (GA), Evolutionary Programming (EP), Flower Pollination Algorithm (FPA), Evolutionary Strategies (ES), Memetic Algorithm (MA) and Coral Reefs Optimization (CRO). Optimization of a solid oxide fuel cell which is a complex, non-linear and multimodal problem, is solved by using these algorithms. Voltage-based objective function expressed as minimization of the sum of squared error has been developed in this study. Based on given optimization algorithms, the best value of unknown parameters of solid oxide fuel cell has been reported. Current–voltage, power and efficiency curves are shown in this study to match the estimated values with experimental values. Differential evolutionary algorithm (DE) is found to be best evolutionary algorithm having sum of squared error, mean and standard deviation of 8.33E-06, 1.57E-05 and 6.792E-06 respectively which are minimum among all other algorithm. Differential evolutionary algorithm predicts accurately among all other algorithms, used to find the unknown parameters of solid oxide fuel cell. Box plot shows the minimum median and lowest IQR which further validate the authenticity of DE among all evolutionary based algorithms to find unknown parameters of SOFC used in this study.

In the field of launch vehicles, high launch density, heavy work intensity and large data volume have brought the traditional automatic anomaly detection with incomplete coverage, and complex design of criterion, which influencing performance evaluation of launch vehicle. In order to fully mine the data variation law implied in the massive telemetry data, the intelligent anomaly detection algorithm is proposed and designed as a beneficial supplement to the traditional anomaly detection algorithm, and improve the anomaly detection coverage and efficiency. With the telemetry data generated by the liquid launch vehicle taken as the research object, and the intelligent anomaly detection algorithm is compared with the traditional algorithm. It is shown that the algorithm proposed adapts to the anomaly detection scenarios with frequency anomaly and frame loss, which may be difficult for traditional algorithms to design anomaly detection criteria, and the average anomaly detection performance is improved by nearly 30% and the variance is reduced by 85%, providing a reference for improving anomaly detection system and applying intelligent anomaly detection algorithms in aerospace.

The ministry of transport officially released the implementation plan for the construction of highway long- span bridge structural health monitoring system on march 1, 2021, which is doomed to promote the popularization and application of bridge structure health monitoring. This paper innovatively combines UAV(Unmanned Aerial Vehicle) and IOT(Internet of Things) technology, and carries out practical engineering application in bridge quality detection project. Firstly, based on the research status of bridge health monitoring, the design principles, reference laws and code of intelligent bridge inspection system are sorted out, and a three-layer intelligent bridge detection system based on IOT is designed. The main components include low-power consumption wireless sensor, edge data acquisition device, gateway and health big data analysis center. This paper focuses on the practical application of UAV and IOT technology in intelligent detection of bridge engineering. The innovations include targeted deployment and monitoring of various types of sensors according to the actual bridge structure and on-site conditions, as well as targeted design of UAV flight routing and important detection sites. Finally, this paper discusses the research status of bridge structural damage identification and health status assessment. Although some progress has been made, there are still many difficulties to be overcome for practical engineering applications. Through the introduction of the above practical engineering application cases, it is hoped to provide some guidance for the research of bridge health monitoring and intelligent detection.

In current technological era, surveillance systems generate an enormous volume of video data on a daily basis, making its analysis a difficult task for computer vision experts. Manually searching for unusual events in these massive video streams is a challenging task, since they occur inconsistently and with low probability in real-world surveillance. In contrast, deep learning-based anomaly detection reduces human labour and its decision making ability is comparatively reliable, thereby ensuring public safety. In this paper, we present an efficient deep features-based intelligent anomaly detection framework that can operate in surveillance networks with reduced time complexity. In the proposed framework, we first extract spatiotemporal features from a series of frames by passing each one to a pre-trained Convolutional Neural Network (CNN) model. The features extracted from the sequence of frames are valuable in capturing anomalous events. We then pass the extracted deep features to multi-layer Bi-directional Long Short-term Memory (BD-LSTM) model, which can accurately classify ongoing anomalous/normal events in complex surveillance scenes of smart cities. We performed extensive experiments on various anomaly detection benchmark datasets to validate the functionality of the proposed framework within complex surveillance scenarios. We reported a 3.41% and 8.09% increase in accuracy on UCF-Crime and UCFCrime2Local datasets compared to state-of-the-art methods.