Near real-time subsidence monitoring and AI forecasting with multi-depth extensometers

Purpose: The research examines artificial intelligence technology's (AI) ability to provide real-time medical diagnostics and decision-making solutions for critical care environments. The study targets high-acuity settings such as ICUs and emergency departments to analyze AI's capability to enhance clinical response times and decrease diagnostic delays while improving outcomes for sepsis multi-organ failure and acute respiratory events. Methodology: A systematic literature review utilized PICO-based search terms, which examined PubMed alongside IEEE Xplore and JAMA AI databases. The search query utilized Boolean operators to retrieve results about "real-time AI" combined with "critical care diagnostics" and "emergency care AI" along with "point-of-care AI tools". Peer-reviewed studies published between 2021 and 2024 received priority for evaluation because they assessed AI-based models for real-time monitoring, predictive analytics, and edge AI deployments in critical care settings. The research focused on studies implementing reproducible validation methods using authentic clinical data sets. Findings: Implementing AI models produced significant enhancements in early warning systems and real-time physiological monitoring and emergency diagnostics, surpassing conventional tools in terms of sensitivity and speed of inference. The deployment of edge AI systems in real-time allowed continuous vital sign data integration with lab and imaging inputs, which improved clinical decision-making through latency reduction. The integration of explainable AI frameworks (e.g., SHAP and LIME) within clinical workflows resulted in a 20% enhancement in diagnostic precision and a significant decrease in incorrect alerts according to study-based quantitative benchmarks. A unique contribution to theory, practice, and policy (recommendations): The research builds theoretical knowledge about AI-based temporal modeling in changing clinical environments while demonstrating the practical advantages of implementing real-time AI directly into bedside medical equipment. The research supports a transformation from reactive to anticipatory healthcare practices enabled by AI-based early interventions. The study suggests that regulatory frameworks should be established to guarantee the ethical implementation of AI tools alongside strict clinical validation and system interoperability in critical care settings. The research presents an operational plan that stakeholders can utilize to build reliable, time-sensitive AI systems for medical frontlines.

Currently the settlement and deformation of factory building structure is monitored using total stations and other more conventional measuring instruments, it is difficult to reflect the health of the structure timely and accurately. In order to change the situation, we establish a set of system for real-time monitoring of deformation and safety warning. The system is formed of sensing layer, transport layer and application layer. Sensing layer is composed of static force level and biaxial inclinometer. The system can be used in dynamic real-time factory structure safety monitoring, also applied to other similar structural monitoring. This paper will study the system components and principle, early warning systems grading, calculation of real-time deformation of roof frame, laboratory test scheme and verification. Experiments showed that the system is suitable for the actual factory structure monitoring, while the choice of static force level and biaxial inclinometer of precision to meet the requirements.

Real-time IoT-powered AI system for monitoring and forecasting of air pollution in industrial environment

Machine learning (ML) has become a transformative tool in health informatics, offering significant advancements in how healthcare data is analyzed, interpreted, and utilized for improved patient outcomes. This review provides an overview of the key applications, challenges, and opportunities of ML in health informatics. ML techniques such as predictive analytics, personalized medicine, medical imaging, and clinical decision support systems are increasingly being integrated into healthcare to enhance diagnostic accuracy, optimize treatments, and predict disease progression. Additionally, ML-driven natural language processing (NLP) is being used to extract valuable insights from electronic health records (EHR), while wearable devices and remote monitoring systems leverage ML to support chronic disease management and real-time patient monitoring. Despite its potential, the adoption of ML in health informatics faces several challenges. These include issues with data quality, privacy concerns surrounding sensitive patient information, and the interpretability of complex ML models. Ethical considerations, such as algorithmic bias and fairness, are also critical factors that need to be addressed to ensure equitable healthcare outcomes. However, ML presents numerous opportunities for the future, including advancements in deep learning, integration with the Internet of Things (IoT), and the expansion of telemedicine services. Case studies in areas such as AI-driven radiology and predictive models for personalized treatment demonstrate the effectiveness of ML in improving healthcare delivery. As ML continues to evolve, its role in real-time analytics, AI integration, and global health initiatives is expected to grow. Overcoming current barriers will require collaboration between healthcare providers, technologists, and policymakers to ensure ethical and efficient implementation of ML in health informatics, ultimately enhancing healthcare quality and accessibility across diverse populations.

In this paper, we implemented a drone service system for facility inspection. The effectiveness of the service system was verified by applying it to the defect inspection of an actual structure. The service system configuration for facility inspection is as follows: 3D modeling of structures, development of drones for facility inspection, real-time streaming of 5G-based 6K-size images, AI processing, and real-time monitoring based on still images. It is difficult for humans to approach special structures such as sports stadiums, bridges (piers), and high-rise structures for safety management. In addition, there are risks during inspection. Safety inspection using drones can reduce accidents, inspection cost, and time. In addition, the application of real-time image streaming and AI analysis can further reduce time and improve efficiency. The developed drone is used to build a test bed in a sports stadium, and the service model is applied in real time.

Predictive analytics is transforming the maintenance and reliability of satellite telecommunications infrastructure, offering proactive solutions to prevent downtime and enhance operational efficiency. This conceptual review explores key strategies and technological advancements driving the adoption of predictive analytics in this field. The integration of IoT devices and sensors enables real-time monitoring, providing valuable data on equipment performance and environmental conditions. Advanced algorithms, such as AI and ML, analyze this data to predict equipment failures and optimize maintenance schedules. These technologies improve the accuracy of predictive models, allowing companies to reduce downtime and improve overall infrastructure reliability. Challenges include data privacy and security concerns, as well as the integration of predictive analytics into existing maintenance processes. Companies must invest in specialized skills and expertise to implement predictive analytics successfully. Looking ahead, emerging technologies like real-time analytics and AI will continue to shape the future of predictive analytics in satellite telecommunications. Standardized practices, collaboration with industry partners, and a focus on data quality are essential for companies to harness the full potential of predictive analytics. In conclusion, predictive analytics is a game-changer for the maintenance and reliability of satellite telecommunications infrastructure. By adopting predictive analytics, companies can optimize maintenance processes, reduce downtime, and improve overall infrastructure reliability. Keywords: Predictive Analytics, Infrastructure, Telecommunications, Satellite, Reliability.

In the age of big data, business intelligence technology is pivotal in enhancing user experiences and driving innovation across industries. This paper focuses on Alibaba Group, a trailblazer in e-commerce, to examine the transformative role of business intelligence. This paper investigates Alibabas cutting-edge application of business intelligence technology, focusing on intelligent recommendation systems, personalized marketing strategies, and efficient supply chain management. The recommendation system harnesses data analysis to provide tailored product suggestions, boosting user satisfaction and sales. Data-driven marketing strategies enable Alibaba to create personalized promotions and coupons, enhancing user experiences and building loyalty. Intelligent supply chain management employs real-time monitoring, optimized transportation, and data-driven decisions to ensure timely deliveries and cost efficiency. A case study of Alibaba's "Singles Day Global Shopping Festival" illustrates how business intelligence technology creates a dynamic, data-powered shopping event. Every decision during this event is informed by real-time analysis and AI insights, enabling swift responses to evolving consumer needs. In summary, business intelligence, a driving force in the age of big data, is at the heart of Alibabas success. Alibaba has tapped the potential of business intelligence by enhancing the user experience and facilitating data-driven decision-making.

Industry 4.0 intends to address a fast-changing and challenging manufacturing environment with diverse demands, short order lead-time and product life cycle, limited capacities, and highly complex process technologies. A smart manufacturing ecosystem integrated with Industry 4.0 technologies, such as advanced automation, AI, machine learning, big data analytics, and Internet of Things, is capable of performing real-time monitoring and optimization of manufacturing processes in various aspects from high level strategic resource and production planning down to real-time equipment-level smart dispatching and predictive maintenance. By fully using real-time data and AI, the ecosystem is able to help manufacturers shorten production and R&D processes, increase production capacity, reduce production cost, guarantee product quality, and improve product yield. It is suitable to help not only high-tech industries such as semiconductor wafer fabrication, but also conventional labor-intensive sectors. This talk illustrates the transformation and improvement of manufacturing activities by using Industry 4.0 technologies through real-life application examples from the semiconductor manufacturing sector.

Taking the mining subsidence as research object, a real-time monitoring scheme was designed. On this basis, a virtual instrument monitoring system based on LabVIEW was constructed. The data acquisition instrument collect real-time data through monitoring of subsidence and subsidence slope deformation by this system in range of mining influence, then control and transform it. After this, the data was transmitted to the computer terminal through wireless transmission equipment, and monitoring will be got after analysis and prediction by computer terminal. This result can provide mining of this area with real-time advice and guidance, to reduce the harms brought by mining subsidence and unnecessary economic losses.

Usually the settlement monitoring system was mainly applied in buildings, roads and so on in the city, which cannot meet the requirements of complex environments. While the train rail vibration monitoring, wind power platform settlement monitoring，etc are facing severe environmental interference. In order to ensure that rail, power tower platform and other field facilities to long-term and stable operation, there is a set of real-time monitoring system to monitor the settlement of the platform is particularly important. This paper proposes a new kind of high precision and micro displacement measurement technology based on laser and CCD devices, with the help of FPGA and MCU, we can complete the function of high-speed data acquisition and real-time monitoring easily. Compared with some traditional methods, this technology of measurement take great advantages in high-speed data acquisition, real-time performance, high-precision measurement and strong adaptability to complex environment, etc.

There are numerous monitoring technologies available today, owing to the rapid advancements in technology and the increasing demand for safety and security in forests. Real-time monitoring with AI cameras, which are commonly utilized for creating and updating real-time features through surveillance, stands out as one of the most effective monitoring solutions. The objective of this current research is to monitor various risk zones within the Periyar Tiger Reserve by integrating real-time AI camera with geographic data. AI cameras were strategically placed using spatial analysis techniques. Leveraging Geographic Information System (GIS) technology, the system facilitates the spatiotemporal management of multiple cameras and their associated data. The spatial distribution and monitoring range of the AI cameras are depicted on the GIS map, along with the layout densities of the cameras and other pertinent information stored in a geospatial database. Additionally, merging various risk areas identified from past incidents with the camera locations enhances the system's capability to establish accurate topological connections between cameras and other points of interest. The results revealed that only 13% of the risk zone was observable from the nine available Real Time Monitoring towers. However, with the addition of 51 more towers, the visibility of the risk zone would increase to 40%. The remaining 15% of the risk zones were not visible through the existing infrastructure. To address this visibility gap, if permitted by the Wildlife Protection Act, wired communication may need to be implemented instead of wireless for monitoring these areas.

Patient flow management for hospitals is a key effort to reduce hospital wait times and to optimally allot resources. In this research we focus on using Machine Learning algorithms like reinforcement learning, genetic algorithms, deep learning etc. to drive efficiency in hospitals implement. The predictive models were developed based on real hospital datasets, and are used to improve patient scheduling, bed management, and prognosis of hospital stay. Results from experimental work showed that waiting times of patients can be reduced by 37.5%, and the bed occupancy efficiency can be improved by 29% with AI-driven scheduling and optimized resource allocation. Furthermore, predictive models produced an 87.2% accuracy prediction of patient hospital stay durations above traditional statistical methods by 18%. The responses of the models were compared to related works and their flexible nature to evolving healthcare environments was noted. Although large scale implementation remains a challenge, key barriers continue to include data privacy as well as system integration and clinician acceptance. This study brings out the importance of improved cybersecurity frameworks and real-time AI interpetability to facilitate hospital integration seamlessly. Second, future research should capitalize upon real-time monitoring and the blockchain based security integration with the real time decision support systems provided by AI. AI’s ability to transform healthcare with more effective, data driven and patient needs responsive patient flow management is underlined by these findings.

Subsidence was first identified in the Wilmington Oil Field in the 1940s. The City of Long Beach, Gas and Oil Department (LBGO) has been conducting surface elevation surveying using spirit leveling and, since 2002, the Global Positioning System (GPS). Recently, software for the LBGO Deformation Network of 12 real-time GPS stations was upgraded with new state-of-the-art technologies and solutions to provide more reliable and accurate continuous monitoring to support elevation survey campaigns. Current oil field waterflood management practices as they relate to surface elevation changes and benefits from the software upgrade for Wilmington Oil Field subsidence monitoring and control are discussed. Waterflood operations in the Wilmington Oil Field are managed for effective oil recovery and injection/production balancing while maintaining stable surface elevations for continued regional economic growth. For cost-effective and timelier surface elevation monitoring, a real-time GPS network provides control for all LBGO subsidence monitoring activities. Using mobile GPS survey equipment, quarterly and semiannual elevations of over 240 bench marks are calculated from the 12 permanent GPS stations' elevations. The LBGO GPS network and software allow for realtime elevation monitoring of key locations and automated data processing, with the regional picture updated semiannually. Surface elevation contour maps and individual survey bench mark elevation trends as a function of time allow for timely adjustments to the waterflood operations to maintain effective oil recovery. Real-time GPS monitoring of surface elevations provides the capability for early detection of surface deformation resulting from production/injection imbalances. The necessary operational adjustments can then be identified and implemented to mitigate the situation. The advantage of early detection is in minimizing the extent of irreversible reservoir rock compaction, resulting in more stable future surface elevations and preventing elevation changes detrimental to surface infrastructure.

The fast development of wireless sensor networks and MEMS make it possible to set up today real-time wireless geotechnical monitoring. To handle interferences and noises from the output data, Kalman filter can be selected as a method to achieve a more realistic estimate of the observations. In this paper, a one-day wireless measurement using accelerometers and inclinometers was deployed on top of a tunnel section under construction in order to monitor ground subsidence. The normal vectors of the sensors were firstly obtained with the help of rotation matrices, and then be projected to the plane of longitudinal section, by which the dip angles over time would be obtained via a trigonometric function. Finally, a centralized Kalman filter was applied to estimate the tilt angles of the sensor nodes based on the data from the embedded accelerometer and the inclinometer. Comparing the results from two sensor nodes deployed away and on the track respectively, the passing of the tunnel boring machine can be identified from unusual performances. Using this method, the ground settlement due to excavation can be measured and a real-time monitoring of ground subsidence can be realized.

Monitoring and controlling of tunneling caused ground subsidence is a key issue related to construction safety. Instead of real-time monitoring, manually measurement using total station or leveling instrument is usually conducted at regular intervals. With the rapid development of wireless sensor networks and microelectro mechanical systems, geological engineers are able to carry out a real-time wireless monitoring of ground subsidence through sensor nodes deployed on the ground. In this paper, we provide a study of accelerometer data fusion in the use of ground settlement monitoring based on surface measurements during shield tunneling. Initially, two sets of sensor nodes were placed at the surface above the excavated tunnel; after obtaining the raw data, the normal vectors of the sensors were firstly derived with the help of rotation matrices, then the dip angles over time were obtained from the projection of normal vectors in the longitudinal section; finally, a centralized Kalman filter was applied to estimate the tilt angles of the sensor nodes based on the data from two respective sets of accelerometers. Using this method, the absolute ground settlement can be achieved via a distributed deployment of sensor nodes.