Real-time Prediction System Research Articles

RescueNow: Real-time SOS and Predictive Women's Safety System

This paper presents a comprehensive women’s safety application designed to address the growing concerns of personal safety through technological intervention. The system integrates multiple advanced features such as real-time SOS alerts, geofencing to detect high-risk areas, AI-driven crime predictions, and multiple SOS activation methods, including voice commands, shake detection, and button press. Additionally, the app employs silent recording functionality to gather evidence in distress situations and incorporates secure communication channels for contacting law enforcement. The application also supports community-driven safety ratings and crime reporting. Tested with real-world scenarios, the system demonstrates high accuracy in predictive analytics and provides a seamless, scalable solution for enhancing women’s safety. This project contributes to the field of personal safety technologies by introducing a multi-faceted, AI-enabled framework for real-time risk mitigation. Key Words: Women’s safety, SOS alerts, AI crime prediction, geofencing, real-time GPS tracking, predictive analytics, Firebase Cloud Messaging, Twilio, silent recording, crime reporting.

Retraction Note: Design of a real-time health monitoring and prediction system for table tennis players based on optical detection sensor networks

Retraction Note: Design of a real-time health monitoring and prediction system for table tennis players based on optical detection sensor networks

Real-time identification of epistatic interactions in SARS-CoV-2 from large genome collections

BackgroundThe emergence of the SARS-CoV-2 virus has highlighted the importance of genomic epidemiology in understanding the evolution of pathogens and guiding public health interventions. The Omicron variant in particular has underscored the role of epistasis in the evolution of lineages with both higher infectivity and immune escape, and therefore the necessity to update surveillance pipelines to detect them early on.ResultsIn this study, we apply a method based on mutual information between positions in a multiple sequence alignment, which is capable of scaling up to millions of samples. We show how it can reliably predict known experimentally validated epistatic interactions, even when using as little as 10,000 sequences, which opens the possibility of making it a near real-time prediction system. We test this possibility by modifying the method to account for the sample collection date and apply it retrospectively to multiple sequence alignments for each month between March 2020 and March 2023. We detected a cornerstone epistatic interaction in the Spike protein between codons 498 and 501 as soon as seven samples with a double mutation were present in the dataset, thus demonstrating the method’s sensitivity. We test the ability of the method to make inferences about emerging interactions by testing candidates predicted after March 2023, which we validate experimentally.ConclusionsWe show how known epistatic interaction in SARS-CoV-2 can be detected with high sensitivity, and how emerging ones can be quickly prioritized for experimental validation, an approach that could be implemented downstream of pandemic genome sequencing efforts.

Optimal air quality management using novel dual Mamdani and neuro fuzzy inference system for real-time accurate prediction

Optimal air quality management using novel dual Mamdani and neuro fuzzy inference system for real-time accurate prediction

Implementation of MATLAB image processing and AI for real-time mood prediction

This study proposes an advanced real-time mood prediction system that utilizes Convolutional Neural Networks (CNNs) and webcam-based Region of Interest (ROI) extraction. By employing MATLAB and its Computer Vision Toolbox, the system detects and preprocesses facial regions from live webcam feeds to accurately predict an individual’s mood. The performance of this system is validated using a custom dataset, showcasing its potential for applications in human-computer interaction, psychological assessments, and real-time monitoring. This approach highlights a significant advancement in combining AI and image processing for real-time emotional analysis.

Front-end speech processing system with SVM algorithm in rail passenger flow management

In order to achieve the goal of dynamically adjusting daily passenger flow to effectively control the overall efficiency of the transportation system, this study constructs a real-time monitoring and prediction system for subway passenger flow based on front-end voice processing technology and support vector machine models. The study first conducted a railway passenger flow analysis, and then used a support vector machine model to construct a preliminary prediction system. In order to achieve global optimization, the study also introduced particle swarm optimization algorithm to construct an optimization prediction model based on PSO-SVM. The results show that the proposed PSO-SVM method has undergone 48 iterations of training, and the predicted values closely match the actual passenger flow curve. The maximum RE error is 2%, and the overall prediction error is 98%. The decision coefficient of PSO-SVM is 0.998932. Therefore, this indicates that it has high performance and feasibility in predicting and controlling passenger flow during peak hours of urban rail transit.

Application of real-time water temperature prediction system in winter for long-distance water diversion projects

ABSTRACT Water diversion projects in high-latitude areas often reduce the risk of ice jams in winter by reducing the water transfer flow, which might cause the waste of water transfer benefits. This paper establishes a real-time prediction system of water temperature in winter, which can predict the change in water temperature by inputting the air temperature forecast data and the current hydraulic data. Taking the middle route of the south-to-north water diversion project as the background, the model parameters calibration and system application testing at different time periods are carried out. The results show that the prediction errors of water temperature for the 1 and 7 days are relatively small, and the prediction errors of water temperature at four observation stations can be controlled within ±0.3 and ±0.6 °C, with the root mean square error (RMSE) ranging from 0.07 to 0.25 and 0.12 to 0.36, respectively. The 15-day water temperature prediction results are greatly affected by air temperature input conditions. The prediction errors for the first 7 days are relatively small, ranging from −0.59 to 0.36 °C, and the errors for the last 8 days increase as the accuracy of the air temperature forecast decreases, ranging from −2.42 to 0.22 °C.

Healthcare facilities management: A novel data-driven model for predictive maintenance of computed tomography equipment

BackgroundThe breakdown of healthcare facilities is a huge challenge for hospitals. Medical images obtained by Computed Tomography (CT) provide information about the patients' physical conditions and play a critical role in diagnosis of disease. To deliver high-quality medical images on time, it is essential to minimize the occurrence frequencies of anomalies and failures of the equipment. MethodsWe extracted the real-time CT equipment status time series data such as oil temperature, of three equipment, between May 19, 2020, and May 19, 2021. Tube arcing is treated as the classification label. We propose a dictionary-based data-driven model SAX-HCBOP, where the two methods, Histogram-based Information Gain Binning (HIGB) and Coefficient improved Bag of Pattern (CoBOP), are implemented to transform the data into the bag-of-words paradigm. We compare our model to the existing predictive maintenance models based on statistical and time series classification algorithms. ResultsThe results show that the Accuracy, Recall, Precision and F1-score of the proposed model achieve 0.904, 0.747, 0.417, 0.535, respectively. The oil temperature is identified as the most important feature. The proposed model is superior to other models in predicting CT equipment anomalies. In addition, experiments on the public dataset also demonstrate the effectiveness of the proposed model. ConclusionsThe two proposed methods can improve the performance of the dictionary-based time series classification methods in predictive maintenance. In addition, based on the proposed real-time anomaly prediction system, the model assists hospitals in making accurate healthcare facilities maintenance decisions.

RETRACTED ARTICLE: Design of a real-time health monitoring and prediction system for table tennis players based on optical detection sensor networks

RETRACTED ARTICLE: Design of a real-time health monitoring and prediction system for table tennis players based on optical detection sensor networks

Real-Time Predictive Maintenance System of Industrial Equipment without Historical Failure Data

Real-Time Predictive Maintenance System of Industrial Equipment without Historical Failure Data

Research on the Application of Intelligent Diagnosis of Major Equipment Failures in Coal Mines

Abstract Currently, the implementation of a coal mine production safety monitoring system, capable of real-time monitoring of the safety condition of large coal mine equipment and timely dispatch management, is crucial for the efficient and safe operation of coal mine enterprises. This paper designs the coal mine major equipment failure process based on the rolling bearing’s vibration mechanism, analyzes the major equipment’s failure signal using both time domain and frequency domain analysis methods, and employs the intelligent fuzzy control method to identify the major equipment failure. We have designed and practically applied a real-time, multi-threaded, and comprehensive fault diagnosis and prediction system for coal mine major equipment, combining information from coal mine safety, production, and operation. The results show that the normal state’s time-domain amplitude approximates (-0.2, 0.2), and the time-domain waveforms all display periodic shock signals when the bearing sustains damage. The faulty vibration amplitude of the inner ring is within 0.08 ms 2 and the faulty vibration amplitude of the outer ring is within 0.44 ms 2, and the diagnostic accuracy using the method of this paper reaches 97.2%. This study demonstrates the effectiveness of remote monitoring and fault analysis of major equipment in coal mines, which effectively improves the equipment start-up rate and ensures coal mine production safety.

Surface quality prediction for FDM-printed parts using in-process material flow data

Efficient and sustainable manufacturing practices rely on the early detection and removal of faulty components in production processes. In the context of Fused Deposition Modeling, this means that identifying defective parts during the production cycle can help to minimize waste and optimize resource utilization. However, conventional quality control methods, which involve post-process inspection, can be time-consuming and inefficient, particularly if nonconforming parts are detected after the production cycle is complete. To address this issue, a real-time quality prediction system has been developed that utilizes in-process flow sensor data to detect and identify nonconforming parts as they are being produced. The system was tested on cuboid test specimens, which were deliberately modified to include defects on the surface of the part. By analyzing the sensor data in real-time, the system was able to identify the defective parts and provide corrective actions to minimize waste and optimize resource utilization. By implementing this approach, manufacturing processes can be streamlined and resource utilization can be optimized while minimizing the production cycle time. This approach represents a significant advance over traditional quality control methods, which rely on post-process inspection and human factors.

TECHNOLOGY-ASSISTED CARE PLANNING IN PRIMARY CARE OF A PROVIDER SHORTAGE AREA

Abstract We applied faireness algorithm to identify and mitigate bias in healthcare system. We also developed DeepPlan soft that can assist primary care providers, patients and caregivers with providing real-time risk prediction system, for example, discussing nursing home placement and designation of power of attorney. Our DeepPlan was built using cloud database system than can be exchangeable in secure way between patients and primary care providers. We linked DeepPlan use to workforce education and now we are ready for presenting education and patient clinical outcomes of using DeepPlan in primary care in a provider shortage area.

A real-time wave prediction in directional wave fields: Strategies for accurate continuous prediction in time

Previous research on real-time deterministic sea wave prediction has generally focused on evaluating the accuracy and efficiency of short-term wave fields within a specific prediction zone. However, for a real-time wave prediction system, it is necessary to provide a continuous description of the ocean wave surface based on short-term prediction segments. In this regard, we have developed algorithms for continuous wave prediction in directional wave fields based on the “practical” prediction zone. The practical prediction zone refers to the time interval available for generating a continuous wave forecast by excluding the reconstructed waves from the prediction zone proposed by Kim et al. (2023). We also introduce and discuss several important time factors, such as the update interval of the spatio-temporal wave dataset, the total computation time, and the length of the practical prediction zone. By gaining a deeper understanding of numerical modeling setups, we have established strategies to reduce computational costs, which are directly related to the accuracy of continuous wave prediction. In particular, the development of these strategies suggests guidance in specifying direction and frequency bandwidths for continuous wave prediction.

Real-time artificial intelligence predicts adverse outcomes in acute pancreatitis in the emergency department: Comparison with clinical decision rule.

Artificial intelligence (AI) prediction is increasingly used for decision making in health care, but its application for adverse outcomes in emergency department (ED) patients with acute pancreatitis (AP) is not well understood. This study aimed to clarify this aspect. Data from 8274 ED patients with AP in three hospitals from 2009 to 2018 were analyzed. Demographic data, comorbidities, laboratory results, and adverse outcomes were included. Six algorithms were evaluated, and the one with the highest area under the curve (AUC) was implemented into the hospital information system (HIS) for real-time prediction. Predictive accuracy was compared between the AI model and Bedside Index for Severity in Acute Pancreatitis (BISAP). The mean ± SD age was 56.1 ± 16.7 years, with 67.7% being male. The AI model was successfully implemented in the HIS, with Light Gradient Boosting Machine (LightGBM) showing the highest AUC for sepsis (AUC 0.961) and intensive care unit (ICU) admission (AUC 0.973), and eXtreme Gradient Boosting (XGBoost) showing the highest AUC for mortality (AUC 0.975). Compared to BISAP, the AI model had superior AUC for sepsis (BISAP 0.785), ICU admission (BISAP 0.778), and mortality (BISAP 0.817). The first real-time AI prediction model implemented in the HIS for predicting adverse outcomes in ED patients with AP shows favorable initial results. However, further external validation is needed to ensure its reliability and accuracy.

CTIM-40. EARLY RADIOGRAPHIC RESPONSE AND ENGRAFTMENT OF PHASE I FIRST-IN-HUMAN BICISTRONIC CAR T CELLS CORRELATE WITH REAL-TIME AUTOLOGOUS GBM ORGANOID CYTOLYSIS

Abstract Glioblastoma (GBM) is a highly treatment-resistant cancer, in part due to prominent inter- and intratumoral heterogeneity. To date, monovalent CAR T-cell therapies have failed to overcome this heterogeneity. Bivalent CAR-T cells may overcome this challenge by expanding the population of target-positive tumor cells. NCT05168423 is first-in-human, phase 1 dose escalation study to assess safety and feasibility of intrathecal delivery of autologous bicistronic CAR-T cells, targeting both EGFR and IL13Rα2, in patients with EGFR-amplified recurrent GBM. For each patient enrolled in this study, high-fidelity glioblastoma organoids (GBOs) are derived from tumor tissue and utilized to test the autologous CAR-T product in real-time. GBOs are co-cultured with autologous CAR-T cells and individually monitored using modified impedance spectroscopy assays over six days for quantification of real-time tumor cytolysis. CAR-T efficacy in GBOs was correlated to matched clinical trial patient outcomes, including CAR-T engraftment. To date, three patients have been treated at the initial dose level (10,000,000 CART-EGFR-IL13Rα2 cells). All three patients demonstrated tumor shrinkage on MRI obtained within 24-48hrs post-infusion. MRI changes were accompanied by significant tumor-inflammation associated neurotoxicity that resolved with steroids and IL1R blockade. Serial CSF collection for qPCR following CAR-T infusion revealed average peaks of >100,000 copies of CAR-T cells/μg of gDNA, comparable to CART19-treated patients. Cytotoxicity values in GBOs strongly correlated with peak CSF engraftment levels of CAR-T cells in all three patients (r = 0.998, p = 0.037). In summary, we present striking radiographic and biological responses from the first cohort of recurrent GBM patients treated with a bivalent, intrathecally delivered CAR T cell product simultaneously targeting EGFR and IL13Rα2. Potent anti-tumor activity of CART-EGFR-IL13Rα2 cells was also observed by treating patient-derived GBOs with the patients’ corresponding autologous CAR T-cell products, providing a parallel, ex vivo system for real-time prediction of CAR T-cell engraftment.

Dynamic Delirium Prediction in the Intensive Care Unit using Machine Learning on Electronic Health Records.

Delirium is a syndrome of acute brain failure which is prevalent amongst older adults in the Intensive Care Unit (ICU). Incidence of delirium can significantly worsen prognosis and increase mortality, therefore necessitating its rapid and continual assessment in the ICU. Currently, the common approach for delirium assessment is manual and sporadic. Hence, there exists a critical need for a robust and automated system for predicting delirium in the ICU. In this work, we develop a machine learning (ML) system for real-time prediction of delirium using Electronic Health Record (EHR) data. Unlike prior approaches which provide one delirium prediction label per entire ICU stay, our approach provides predictions every 12 hours. We use the latest 12 hours of ICU data, along with patient demographic and medical history data, to predict delirium risk in the next 12-hour window. This enables delirium risk prediction as soon as 12 hours after ICU admission. We train and test four ML classification algorithms on longitudinal EHR data pertaining to 16,327 ICU stays of 13,395 patients covering a total of 56,297 12-hour windows in the ICU to predict the dynamic incidence of delirium. The best performing algorithm was Categorical Boosting which achieved an area under receiver operating characteristic curve (AUROC) of 0.87 (95% Confidence Interval; C.I, 0.86-0.87). The deployment of this ML system in ICUs can enable early identification of delirium, thereby reducing its deleterious impact on long-term adverse outcomes, such as ICU cost, length of stay and mortality.

Utilizing risk-controlling prediction calibration to reduce false alarm rates in epileptic seizure prediction.

Epilepsy is a neurological disease characterized by sudden, unprovoked seizures. The unexpected nature of epileptic seizures is a major component of the disease burden. Predicting seizure onset and alarming patients may allow timely intervention, which would improve clinical outcomes and patient quality of life. Currently, algorithms aiming to predict seizures suffer from a high false alarm rate, rendering them unsuitable for clinical use. We adopted here a risk-controlling prediction calibration method called Learn then Test to reduce false alarm rates of seizure prediction. This method calibrates the output of a "black-box" model to meet a specified false alarm rate requirement. The method was initially validated on synthetic data and subsequently tested on publicly available electroencephalogram (EEG) records from 15 patients with epilepsy by calibrating the outputs of a deep learning model. Validation showed that the calibration method rigorously controlled the false alarm rate at a user-desired level after our adaptation. Real data testing showed an average of 92% reduction in the false alarm rate, at the cost of missing four of nine seizures of six patients. Better-performing prediction models combined with the proposed method may facilitate the clinical use of real-time seizure prediction systems.

Real-time artificial intelligence system for bacteremia prediction in adult febrile emergency department patients

BackgroundArtificial intelligence (AI) holds significant potential to be a valuable tool in healthcare. However, its application for predicting bacteremia among adult febrile patients in the emergency department (ED) remains unclear. Therefore, we conducted a study to provide clarity on this issue. MethodsAdult febrile ED patients with blood cultures at Chi Mei Medical Center were divided into derivation (January 2017 to June 2019) and validation groups (July 2019 to December 2020). The derivation group was utilized to develop AI models using twenty-one feature variables and five algorithms to predict bacteremia. The performance of these models was compared with qSOFA score. The AI model with the highest area under the receiver operating characteristics curve (AUC) was chosen to implement the AI prediction system and tested on the validation group. ResultsThe study included 5,647 febrile patients. In the derivation group, there were 3,369 patients with a mean age of 61.4 years, and 50.7% were female, including 508 (13.8%) with bacteremia. The model with the best AUC was built using the random forest algorithm (0.761), followed by logistic regression (0.755). All five models demonstrated better AUC than the qSOFA score (0.560). The random forest model was adopted to build a real-time AI prediction system integrated into the hospital information system, and the AUC achieved 0.709 in the validation group. ConclusionThe AI model shows promise to predict bacteremia in adult febrile ED patients; however, further external validation in different hospitals and populations is necessary to verify its effectiveness.

A dynamic method to predict driving risk on sharp curves using multi-source data

Traffic accidents are likely to occur on sharp curves under poor driving conditions, and the severity level of such accidents is high. Therefore, predicting the risk associated with driving on curved roadways in real time can effectively improve driving safety. This paper aims to develop a dynamic real-time method that fuses multiple data sources to predict risk when driving on sharp curves in the context of the connected vehicle environment. Six curves with three small radii (60 m, 100 m, 150 m) and two driving directions (left and right) were designed for a driving simulation experiment. Driver maneuvering data, vehicle kinematic data, and physiological data of 55 drivers were collected for this study. The data were combined and spatially and dynamically segmented. The mean value of the critical lateral acceleration of the vehicle was set as the risk assessment index. K-means clustering was used to classify the driving risk into three levels: low, medium, and high. Then, the risk level was predicted using the maneuvering data, vehicle kinematic data, and physiological data as well as road alignment characteristics as input features for the proposed model that employs the long short-term memory (LSTM) network algorithm. Models with different combinations of observation window (lookback) and interval window (delay) were compared to derive the best window combination. The algorithms selected for comparison against the LSTM algorithm are random forest, XGBoost, and LightGBM. The results show that the proposed LSTM-based method can effectively predict dangerous driving behavior on sharp curves. The optimal window combination derived using the LSTM algorithm is lookback = 20 m and delay = 20 m. The prediction performance of the proposed model is significantly better than that of the other three compared algorithms, with F1-scores of 84.8% and 86.0% for the medium and high risk categories, respectively. In addition, the proposed LSTM-based model that fuses multiple data sources is proven to outperform the model that uses only vehicle kinematics data. The dynamic prediction method proposed in this paper can contribute to the development of a real-time prediction and warning system for driving risks at vehicle terminals in the intelligent connected vehicle environment.