Developing a Prediction Model for Real-Time Incident Detection Leveraging User-Oriented Participatory Sensing Data

Density-Based Spatial Clustering of Application with Noise (DBSCAN) is a well-known, unsupervised machine learning tool for clustering. DBSCAN creates clusters based on dense regions while marking points that lie alone in low-density regions as outliers. In DBSCAN, the density is detected via core points which are quite sensitive to input parameters: ϵ is radius of the neighborhood and MinPts is minimum number of points constraint within ϵ radius. In contrast to crisp core point definition, intuitionistic fuzzy core point definition makes DBSCAN algorithm capable to detect different patterns of density by two different combinations of input parameters. In this study, a DBSCAN extension is proposed based on this idea: IFDBSCAN. The proposed extension is then illustrated by a computational experiment on a synthetic dataset. Results show that, IFDBSCAN can establish fine-tuned clusters relatively to classical DBSCAN and provide users more insight on the estimation of input parameters.

This study compares the efficiency of six early warning systems (EWSs) to determine whether the EWS used in most public hospitals in Queensland, Australia, The Queensland Adult Deterioration Detection System (Q-ADDS), is best suited for use in small regional and rural hospitals. In this retrospective case-control study, patients who experienced an in-hospital severe adverse event (index patients) for a 3.5-year period were demographically and diagnostically matched with patients who had uneventful hospital stays (control patients). The EWS efficiency was based on the area under the receiver operator characteristic curve (AUROC) and the number of false and true alerts generated by each EWS. The incidence of severe adverse events was 1.2% of in-hospital patients, and 2500 sets of vital signs were collected from 159 index and 172 control patients. The EWSs were only able to identify approximately half of the index patients. The AUROC was 0.666 to 0.801 and the EWS generated 2.4 to 7.6 false alerts to every true alert per 1000 admissions. The National Early Warning Score had the best ratio of false to true alerts (2.4:1) but was only able to identify 40.8% of deteriorating patients. The Q-ADDS identified 46.5% of the deteriorating patients and had a false to true alert ratio of 3.2:1. When compared with the National Early Warning Score, systems with higher AUROCs (0.744 and 0.801) also had higher proportion of false alerts. None of the alternative EWSs seem to provide marked benefits over Q-ADDS. At present, there is insufficient evidence to replace Q-ADDS with an alternative EWS. Because the EWSs were only able to identify half of the deteriorating patients, EWSs should be used in conjunction with good clinical judgment.

PurposeRecent investigations failed to reproduce the positive rotor-guided ablation outcomes shown by initial studies for treating persistent atrial fibrillation (persAF). Phase singularity (PS) is an important feature for AF driver detection, but algorithms for automated PS identification differ. We aim to investigate the performance of four different techniques for automated PS detection.Methods2048-channel virtual electrogram (VEGM) and electrocardiogram signals were collected for 30 s from 10 patients undergoing persAF ablation. QRST-subtraction was performed and VEGMs were processed using sinusoidal wavelet reconstruction. The phase was obtained using Hilbert transform. PSs were detected using four algorithms: (1) 2D image processing based and neighbor-indexing algorithm; (2) 3D neighbor-indexing algorithm; (3) 2D kernel convolutional algorithm estimating topological charge; (4) topological charge estimation on 3D mesh. PS annotations were compared using the structural similarity index (SSIM) and Pearson’s correlation coefficient (CORR). Optimized parameters to improve detection accuracy were found for all four algorithms using Fβ score and 10-fold cross-validation compared with manual annotation. Local clustering with density-based spatial clustering of applications with noise (DBSCAN) was proposed to improve algorithms 3 and 4.ResultsThe PS density maps created by each algorithm with default parameters were poorly correlated. Phase gradient threshold and search radius (or kernels) were shown to affect PS detections. The processing times for the algorithms were significantly different (p < 0.0001). The Fβ scores for algorithms 1, 2, 3, 3 + DBSCAN, 4 and 4 + DBSCAN were 0.547, 0.645, 0.742, 0.828, 0.656, and 0.831. Algorithm 4 + DBSCAN achieved the best classification performance with acceptable processing time (2.0 ± 0.3 s).ConclusionAF driver identification is dependent on the PS detection algorithms and their parameters, which could explain some of the inconsistencies in rotor-guided ablation outcomes in different studies. For 3D triangulated meshes, algorithm 4 + DBSCAN with optimal parameters was the best solution for real-time, automated PS detection due to accuracy and speed. Similarly, algorithm 3 + DBSCAN with optimal parameters is preferred for uniform 2D meshes. Such algorithms – and parameters – should be preferred in future clinical studies for identifying AF drivers and minimizing methodological heterogeneities. This would facilitate comparisons in rotor-guided ablation outcomes in future works.

Background: This study focuses on the latest knowledge regarding Micro, Small and Medium Enterprises (MSMEs) as a current central issue. These enterprises have shown their significance in providing employment opportunities and contributing to the country’s economy. However, MSMEs face various challenges that must be addressed to optimize their outcomes. Understanding the characteristics of this group was crucial in formulating effective strategies. Objective: This study proposed to cluster or combine micro, small, and medium enterprises (MSMEs) data in a particular area based on asset value and turnover. As a result, this study aimed to gain insights into the MSME landscape in the area and provided valuable information for decision-makers and stakeholders. Methods: This study utilized two methods, namely the DBSCAN (Density-Based Spatial Clustering of Applications with Noise) method and the K-Means method. These methods were chosen for their distinct capabilities. DBSCAN was selected for its ability to handle noisy data and identify clusters with diverse forms, while K-Means was chosen for its popularity and ability to group data based on proximity. The study used a dataset containing MSME information, including asset values and turnover, collected from various sources. Results: The outcomes encompassed identifying clusters of MSMEs based on their closeness in the feature space within a specific region. Optimizing the clustering outcomes involved modifying algorithm parameters like epsilon and minimum points for DBSCAN and the number of clusters for K-Means. Furthermore, this study attained a deeper understanding of the arrangement and characteristics of MSME clusters in the region through a comparative analysis of the two methodologies. Conclusion: This study offered perspectives on clustering MSMEs based on asset value and turnover in a specific region. Employing DBSCAN and K-Means methodologies allowed researchers to depict the MSME landscape and grasp the business attributes of these enterprises. These results could aid in decision-making and strategic planning concerning the advancement of the MSME sector in the mentioned area. Future study may investigate supplementary factors and variables to deepen comprehension of MSME clusters and promote regional growth and sustainability. Keywords: Asset Value, Clustering, DBSCAN, K-Means, Turnover

BackgroundIn the pediatric intensive care unit (PICU), quantifying illness severity can be guided by risk models to enable timely identification and appropriate intervention. Logistic regression models, including the pediatric index of mortality 2 (PIM-2) and pediatric risk of mortality III (PRISM-III), produce a mortality risk score using data that are routinely available at PICU admission. Artificial neural networks (ANNs) outperform regression models in some medical fields.ObjectiveIn light of this potential, we aim to examine ANN performance, compared to that of logistic regression, for mortality risk estimation in the PICU.MethodsThe analyzed data set included patients from North American PICUs whose discharge diagnostic codes indicated evidence of infection and included the data used for the PIM-2 and PRISM-III calculations and their corresponding scores. We stratified the data set into training and test sets, with approximately equal mortality rates, in an effort to replicate real-world data. Data preprocessing included imputing missing data through simple substitution and normalizing data into binary variables using PRISM-III thresholds. A 2-layer ANN model was built to predict pediatric mortality, along with a simple logistic regression model for comparison. Both models used the same features required by PIM-2 and PRISM-III. Alternative ANN models using single-layer or unnormalized data were also evaluated. Model performance was compared using the area under the receiver operating characteristic curve (AUROC) and the area under the precision recall curve (AUPRC) and their empirical 95% CIs.ResultsData from 102,945 patients (including 4068 deaths) were included in the analysis. The highest performing ANN (AUROC 0.871, 95% CI 0.862-0.880; AUPRC 0.372, 95% CI 0.345-0.396) that used normalized data performed better than PIM-2 (AUROC 0.805, 95% CI 0.801-0.816; AUPRC 0.234, 95% CI 0.213-0.255) and PRISM-III (AUROC 0.844, 95% CI 0.841-0.855; AUPRC 0.348, 95% CI 0.322-0.367). The performance of this ANN was also significantly better than that of the logistic regression model (AUROC 0.862, 95% CI 0.852-0.872; AUPRC 0.329, 95% CI 0.304-0.351). The performance of the ANN that used unnormalized data (AUROC 0.865, 95% CI 0.856-0.874) was slightly inferior to our highest performing ANN; the single-layer ANN architecture performed poorly and was not investigated further.ConclusionsA simple ANN model performed slightly better than the benchmark PIM-2 and PRISM-III scores and a traditional logistic regression model trained on the same data set. The small performance gains achieved by this two-layer ANN model may not offer clinically significant improvement; however, further research with other or more sophisticated model designs and better imputation of missing data may be warranted.

The density based notion for clustering approach is used widely due to its easy implementation and ability to detect arbitrary shaped clusters in the presence of noisy data points without requiring prior knowledge of the number of clusters to be identified. Density-based spatial clustering of applications with noise (DBSCAN) is the first algorithm proposed in the literature that uses density based notion for cluster detection. Since most of the real data set, today contains feature space of adjacent nested clusters, clearly DBSCAN is not suitable to detect variable adjacent density clusters due to the use of global density parameter neighborhood radius Nrad and minimum number of points in neighborhood Npts. So the efficiency of DBSCAN depends on these initial parameter settings, for DBSCAN to work properly, the neighborhood radius must be less than the distance between two clusters otherwise algorithm merges two clusters and detects them as a single cluster. Through this paper: 1) We have proposed improved version of DBSCAN algorithm to detect clusters of varying density adjacent clusters by using the concept of neighborhood difference and using the notion of density based approach without introducing much additional computational complexity to original DBSCAN algorithm. 2) We validated our experimental results using one of our authors recently proposed space density indexing (SDI) internal cluster measure to demonstrate the quality of proposed clustering method. Also our experimental results suggested that proposed method is effective in detecting variable density adjacent nested clusters.

A method of two-stage clustering learning based on improved DBSCAN and density peak algorithm

An efficient and scalable density-based clustering algorithm for datasets with complex structures

This study aims to objectively analyze the feasibility of prospective recipients of the Smart Indonesia Card Scholarship (KIP-K) by integrating the Density-Based Spatial Clustering of Applications with Noise (DBSCAN) algorithm and the Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) method. The research dataset consists of 287 data on prospective scholarship recipients with 11 main attributes that reflect the socio-economic and academic conditions of students. The research process includes data collection, pre-processing, transformation of categorical attributes into numerical values using a linear weighting scheme, cluster analysis using DBSCAN, and candidate ranking using TOPSIS. DBSCAN is used to identify cluster patterns and detect anomalies in the data of potential recipients, while TOPSIS is used to rank candidates based on proximity to the ideal solution. The results of the grouping produced 10 clusters and one noise cluster that showed a variety of socio-economic characteristics of prospective scholarship recipients. The results of the ranking show that some of the candidates with the highest TOPSIS scores come from clusters with higher levels of economic vulnerability. In addition, some of the high-scoring candidates also came from the noise cluster, indicating that even though they did not belong to a particular group, they still met the eligibility criteria based on a multi-criteria evaluation. These findings show that the combination of DBSCAN and TOPSIS has the potential to support the process of analyzing the eligibility of scholarship recipients in a more systematic and data-driven manner.

Image segmentation is the process to divide a digital image into a number of regions for further usage. Color images can be segmented by applying density-based clustering methods, e.g. Density-Based Spatial Clustering of Applications with Noise (DBSCAN), which is used to identify the arbitrary shaped clusters. The drawback of DBSCAN is the high computational complexity whilst the size of image input is normally very large. Self-Organizing Map (SOM) is a dimensionality reduction method which can be applied to reduce the dimensions of image processing tasks. This paper proposes a hybrid method of SOM and DBSCAN (SOM-DBSCAN) for image segmentation. To evaluating the usability of the proposed SOM-DBSCAN method, four images are used to benchmark image segmentation.

BackgroundFetal alcohol syndrome (FAS) is a lifelong developmental disability that occurs among individuals with prenatal alcohol exposure (PAE). With improved prediction models, FAS can be diagnosed or treated early, if not completely prevented.ObjectiveIn this study, we sought to compare different machine learning algorithms and their FAS predictive performance among women who consumed alcohol during pregnancy. We also aimed to identify which variables (eg, timing of exposure to alcohol during pregnancy and type of alcohol consumed) were most influential in generating an accurate model.MethodsData from the collaborative initiative on fetal alcohol spectrum disorders from 2007 to 2017 were used to gather information about 595 women who consumed alcohol during pregnancy at 5 hospital sites around the United States. To obtain information about PAE, questionnaires or in-person interviews, as well as reviews of medical, legal, or social service records were used to gather information about alcohol consumption. Four different machine learning algorithms (logistic regression, XGBoost, light gradient-boosting machine, and CatBoost) were trained to predict the prevalence of FAS at birth, and model performance was measured by analyzing the area under the receiver operating characteristics curve (AUROC). Of the total cases, 80% were randomly selected for training, while 20% remained as test data sets for predicting FAS. Feature importance was also analyzed using Shapley values for the best-performing algorithm.ResultsOverall, there were 20 cases of FAS within a total population of 595 individuals with PAE. Most of the drinking occurred in the first trimester only (n=491) or throughout all 3 trimesters (n=95); however, there were also reports of drinking in the first and second trimesters only (n=8), and 1 case of drinking in the third trimester only (n=1). The CatBoost method delivered the best performance in terms of AUROC (0.92) and area under the precision-recall curve (AUPRC 0.51), followed by the logistic regression method (AUROC 0.90; AUPRC 0.59), the light gradient-boosting machine (AUROC 0.89; AUPRC 0.52), and XGBoost (AUROC 0.86; AURPC 0.45). Shapley values in the CatBoost model revealed that 12 variables were considered important in FAS prediction, with drinking throughout all 3 trimesters of pregnancy, maternal age, race, and type of alcoholic beverage consumed (eg, beer, wine, or liquor) scoring highly in overall feature importance. For most predictive measures, the best performance was obtained by the CatBoost algorithm, with an AUROC of 0.92, precision of 0.50, specificity of 0.29, F1 score of 0.29, and accuracy of 0.96.ConclusionsMachine learning algorithms were able to identify FAS risk with a prediction performance higher than that of previous models among pregnant drinkers. For small training sets, which are common with FAS, boosting mechanisms like CatBoost may help alleviate certain problems associated with data imbalances and difficulties in optimization or generalization.

BackgroundTreatment of sepsis requires prompt initiation of appropriate empiric antibiotics. Coverage for Gram-negative bacilli (GNB) is decided based on risk factors for resistant organisms, as documented in both structured and unstructured data in the electronic health record. We hypothesized that adding large language models (LLMs) that can decode clinical narratives would improve performance of deep learning models in predicting antimicrobial resistance in GNB sepsis.MethodsThis retrospective cohort analysis included all adult patients who met CDC sepsis criteria admitted to 10 BJC hospitals from 1/2018 to 12/2023. Sepsis episodes were classified as community-onset (< 48hrs from hospitalization) or hospital-onset. GNB were stratified as ceftriaxone-susceptible (SS), ceftriaxone-resistant but cefepime-susceptible (RS), and ceftriaxone- and cefepime-resistant (RR). Culture negative or non-GNB sepsis were labeled SS. Deep learning models (XGBoost, DeepAMR) were developed using all available structured data (e.g., demographics, vitals, labs). LLMs (BioGPT, Medical-LLaMa3-8B) were developed on full or partial history and physical exam notes. Models were assessed using area under the receiver operating characteristic curve (AUROC) and area under precision-recall curve (AUPRC).ResultsA total of 29,517 patients were included, for 23,949 community-onset and 25,101 hospital-onset sepsis episodes with RS+RR GNB in 9% and 17%, respectively (Table 1). Relevant features of structured data models included history of prior infections and resistant organisms. DeepAMR was the best performing deep learning model with AUROC 0.85/AUPRC 0.40 and AUROC 0.81/AUPRC 0.50 for community- and hospital-onset sepsis, respectively (Table 2). The best performing LLM was Medical-LLaMa3-8B on full notes (Figure 1), with AUROC 0.74/AUPRC 0.21 and AUROC 0.67/AUPRC 0.30 for community- and hospital-onset sepsis, respectively. Combining deep learning and LLMs did not significantly improve the prediction of resistant infections compared to deep learning alone.ConclusionStructured medical data provided the highest prediction of resistant GNB infections in sepsis. More analysis is needed to understand why the unstructured data did not improve model predictive performance.DisclosuresMaria Cristina Vazquez Guillamet, MD, Bionano: Stocks/Bonds (Public Company)|Charisma: Stocks/Bonds (Public Company)|Ocugen: Stocks/Bonds (Public Company)

Lane-level road information is especially crucial now that high-precision navigation maps are in more demand. Road information may be obtained rapidly and affordably by mining floating vehicle data (FCD). A method is proposed to extract the number of lanes on urban roads by combining deep learning and low-frequency FCD. Initially, the FCD is cleaned using the Density-Based Spatial Clustering of Applications with Noise (DBSCAN) clustering technique. Then, the FCD is split into three categories based on the typical urban road types: one-way one-lane, one-way two-lane, and one-way three-lane, and the deep learning sample data is created using segmentation, rotation, and gridding. Lastly, the number of urban road lanes is obtained by training and predicting the sample data using the LeNet-5 model. The number of urban road lanes was effectively identified from the low-frequency FCD with a detection accuracy of 92.7% through the cleaning and classification of Wuhan FCD. Urban roads can be efficiently covered by the FCD on a regular basis, and lane information can be efficiently collected using deep learning techniques. This method can be used to generate and update lane number information for high-precision navigation maps.

BackgroundAdolescent suicides are more likely to form clusters than those of other age groups. However, the definition of a cluster in the space–time dimension has not been established, neither are the factors contributing to it well known. Therefore, this study aimed to identify space–time clusters in adolescent suicides in Korea and to examine the differences between clustered and non-clustered cases using novel statistical methods.MethodsFrom 2016 to 2020, the dates and locations, including specific addresses from which the latitude and longitude of all student suicides (aged 9–18 years) in Korea were obtained through student suicide reports. Sociodemographic characteristics of the adolescents who died by suicide were collected, and the individual characteristics of each student who died by suicide were reported by teachers using the Strengths and Difficulties Questionnaire (SDQ). Density-Based Spatial Clustering of Applications with Noise (DBSCAN) analysis was used to assess the clustering of suicides.ResultsWe identified 23 clusters through the data analysis of 652 adolescent suicides using DBSCAN. By comparing the size of each cluster, we identified 63 (9.7%) spatiotemporally clustered suicides among adolescents, and the temporal range of these clusters was 7–59 days. The suicide cluster group had a lower economic status than the non-clustered group. There were no significant differences in other characteristics between the two groups.ConclusionThis study has defined the space–time cluster of suicides using a novel statistical method. Our findings suggest that when an adolescent suicide occurs, close monitoring and intervention for approximately 2 months are needed to prevent subsequent suicides. Future research using DBSCAN needs to involve a larger sample of adolescents from various countries to further corroborate these findings.

Unsupervised machine learning and especially unshaped clustering approaches like Density-Based Spatial Clustering of Applications with Noise (DBSCAN) remain at the center of researchers’ attention. Many DBSCAN improvements have been recently proposed, but none of them involve the data heterogeneity that we may find in real-life problems. To tackle this issue, in this paper, a novel DBSCAN-based approach is proposed and applied to real data of Saudi Arabia (KSA), in order to cluster the COVID-19 calls based on hospitals distribution for the resolution of the ambulances dispatching problem and COVID-19 emergency calls covering. The designed approach is called Heterogeneous DBSCAN (HDBSCAN). It considers different data types: Statically distributed hospitals over Saudi Arabia and dynamic COVID-19 incoming calls. The experiments show that the proposed HDBSCAN greatly speeds up the original DBSCAN and provides an efficient classification for the application problem.KeywordsHeterogeneous dataReal-time clusteringHeterogeneous DBSCANDBSCANEmergency callsSaudi Arabia COVID-19