Cluster Counts Research Articles

Performance analysis of Kmeans with modified initial centroid selection algorithms and developed Kmeans9+ model

Partition the sample space to the user requirement is an easy and efficient cluster method in many applications. Kmeans is one of the simple and efficient partition algorithms used in many cluster solutions. We suggest Multivariant Silhouette method to predict the cluster count value for Kmeans algorithm. The work proposes three new algorithms for initial seed selection. The Kmeans algorithm is modified using statistical measure Mean, Median, Partition centre for the cluster centroid calculation. In conventional Kmeans algorithm, the samples are compared with all the partition centroids to decide the inclusion of a sample in a cluster. In Kmeans9+ algorithm, the samples are compared only with the centroids of current and eight nearest neighbouring cluster partitions. It reduces the needless comparisons of samples with cluster centroids and improve the efficiency of the algorithm. The performance of created algorithms is evaluated using data from UCI and KBPE SSLC. Cluster efficiency analysis is done using the cluster evaluation indices such as Silhouette and Dunn Index. Nine nearest neighbour uniform partition cluster model Kmeans9+ improve the performance of the K means algorithm and obtain the natural cluster results with minimum iterations.

Antibody Properties Associate with Clinical Phenotype in LGI1 Encephalitis.

Autoimmune encephalitis (AE) associated with autoantibodies against leucine-rich glioma-inactivated protein-1 (LGI1) can present with faciobrachial dystonic seizures (FBDS) and/or limbic encephalitis (LE). The reasons for this heterogeneity in phenotypes are unclear. We performed autoantibody (abs) characterization per patient, two patients suffering from LE and two from FBDS, using isolated antibodies specified with single amino acid epitope mapping. Electrophysiological slice recordings were conducted alongside spine density measurements, postsynaptic Alpha-amino-3-hydoxy-5-methyl-4-isoaxole-proprionate-receptors (AMPA-R) and N-methyl-D-aspartate-receptors receptor (NMDA-R) cluster counting. These results were correlated with the symptoms of each patient. While LGI1 abs from LE patients mainly interacted with the Leucine-rich repeat section of LGI1, abs from both FBDS patients also recognized the Epitempin section as well. Six-hour incubation of mouse hippocampal slices with LE patients autoantibodies but not from the FBDS patients resulted in a significant decline in long-term potentiation (p = 0.0015) or short-term plasticity at CA3-CA1 neurons and in decreased hippocampal synaptic density. Cluster differentiation showed no decrease in postsynaptic AMPA-R and NMDA-R. LGI1 autoantibodies selected by phenotype show an almost distinct epitope pattern, elicit disparate functional effects on hippocampal neurons, and cause divergent effects on spine density. This data illuminates potential pathomechanisms for disease heterogeneity in LGI1 AE.

An analysis of finding the best strategies of water security for water source areas using an integrated IT2FVIKOR with machine learning

Worldwide, water security is adversely affected by factors such as population growth, rural–urban migration, climate, hydrological conditions, over-abstraction of groundwater, and increased per-capita water use. Water security modeling is one of the key strategies to better manage water safety and develop appropriate policies to improve security. In view of the growing global demand for safe water, intelligent methods and algorithms must be developed. Therefore, this paper proposes an integrated interval type-2 Fuzzy VIseKriterijumska Optimizcija I Kompromisno Resenje (IT2FVIKOR) with unsupervised machine learning (ML). This includes IT2FVIKOR for ranking and selecting a set of alternatives. Unsupervised machine learning includes hierarchical clustering, self-organizing map, and autoencoder for clustering, silhouette analysis and elbow method to find the most optimal cluster count, and finally Adjusted Rank Index (ARI) to find the best comparison within two clusters. This proposed integrated method can be divided into a two-phase fuzzy-machine learning-based framework to select the best water security strategies and categorize the polluted area using the water datasets from the Terengganu River, one of Malaysia’s rivers. Phase 1 focuses on the IT2FVIKOR method to select five different strategies with five different criteria using five decision makers for finding the best water security strategies. Phase 2 continues the unsupervised machine learning where three different clustering algorithms, namely, hierarchical clustering, self-organizing map, and autoencoder, are used to cluster the polluted area in the Terengganu River. Silhouette analysis is applied along with the clustering algorithms to estimate the number of optimal clusters in a dataset. Then, the ARI is applied to find the best comparison within the original data with hierarchical clustering, self-organizing map, and autoencoder. Next, the elbow method is applied to double-confirm the best clusters for each clustering algorithm. Last, lists of polluted areas in each cluster are retrieved. Finally, this 2-phase fuzzy-Machine learning–based framework offers an alternative intelligent model to solve the water security problems and find the most polluted area.

Transient Heat Therapy Has Few Effects on Grapevine Performance in the Okanagan Valley, British Columbia, Canada

<h3>Background and goals</h3> The goal of this study was to evaluate a commercially available machine (Agrotherm XT, Agrothermal Systems) for transient heat application to enhance vineyard canopy temperature during the growing season and to study the effects of transient heat application on grapevine performance, phenology, and berry quality. <h3>Methods and key findings</h3> Heat was applied up to six times during the growing seasons in 2019 and 2020 in three vineyards located in East and West Kelowna in the Okanagan Valley in British Columbia, Canada. Shoot and cluster counts, bud fruitfulness, grapevine phenology, leaf greenness, harvest, pruning weight, and crop load data were collected each year. Results showed that transient heat application is not an effective management practice. Results were not consistent among vineyard locations or growing seasons. In 2019, treated Merlot vines had fewer berries per cluster but higher juice pH, and treated Chardonnay 2 vines had reduced leaf greenness but higher pruning weight. In 2020, treated Merlot vines had reduced leaf greenness but higher pruning weights, and treated Chardonnay 2 vines had reduced leaf greenness but higher berry weights and soluble solids. In 2021 only, treated Chardonnay 2 vines had greater fruitfulness (clusters/shoot). <h3>Conclusions and significance</h3> Heat treatment had inconsistent effects on grapevine performance and berry quality and no changes in grapevine phenology were observed. Given the high purchase costs, transient heat therapy is not an effective management strategy to improve grapevine performance and berry quality in the north central part of the Okanagan Valley in British Columbia.

Yield estimation using machine learning from satellite imagery

Accurate and early yield estimation (from pea size) allows 1.- Make decisions at field level: green harvesting, irrigation management. 2.- Advance or organise the purchase of grapes from suppliers. 3.- Forecast the volume of wine produced in the campaign that has not yet begun. 4.- Define the quality of the vintage: regular and detailed monitoring of whether, or not, the heterogeneity of the leaf surface, photosynthetic activity or soil moisture observed in the vineyards is as expected at this time, compared with historical values. 5.- Precise control of each vine in production, knowing which vines are no longer productive or should be grubbed up. The Sentinel-2 satellite has generated a time series of images spanning more than six years, which is a great help in analysing the state of permanent crops such as vineyards, where grapes are produced every year. The weekly comparison of what is happening in the current season with what has happened in the previous six seasons is information that is in line with agricultural practices: Winegrowers make the mental exercise of comparing how the vines are developing today with how they developed in previous seasons, with the aim of repeating the years of good yields. In addition, several commercial satellites can now capture images of 50 centimetres pixel resolution or even better, making it possible to check the health of each vine every year. Since 2020, GMV and Pago de Carraovejas have been working together to develop a yield estimation service based on field information and satellite images that feed machine learning algorithms. This paper describes the path followed from the beginning and the steps taken, summarising as follows: 1. - Machine learning algorithm trained with cluster counting and satellite data. 2. - Adjustment of the number of vines in production in each vineyard using very high-resolution imagery. 3. - Machine learning algorithm trained on real production from past campaigns and historical Sentinel-2 time series. The results obtained by comparing the actual grape intake in the winery with the yield estimation range from 91% accuracy in 2020 to 95% accuracy in 2022.

Simulation study of particle identification using cluster counting technique for the BESIII drift chamber

The particle identification of charged hadrons, especially for the separation of K and π, is crucial for the flavour physics study. Ionization measurement with the cluster counting technique, which has a much less fluctuation than traditional dE/dx measurement, is expected to provide better particle identification for the BESIII experiment. Simulation studies, including a Garfield++ based waveform analysis and a performance study of K/π identification in the BESIII offline software system have been performed. The results show that K/π separation power and PID efficiency would be improved appreciably in the momentum range above 1.2 GeV/c using cluster counting technique even with a conservative resolution assumption.

Clustering algorithms with prediction the optimal number of clusters

The clustering is a widely used technique for grouping of objects. The objects, which are similar to each other should be in the same cluster. One disadvantage of general clustering algorithms is that the user must specify the number of clusters in advance, as input parameter. This is a major drawback since it is possible that the user cannot specify the number of clusters correctly, and the algorithm thus creates a clustering that puts very different elements into the same cluster. The aim of this paper is to present our representation and evaluation technique to determine the optimal cluster count automatically. With this technique, the algorithms itself determine the number of clusters. In this paper first, the classical clustering algorithms are introduced, then the construction and improvement algorithms and then our representation and evaluation method are presented. Then the performance of the algorithms with test results are compared.

Particle identification with the cluster counting technique for the IDEA drift chamber

IDEA (Innovative Detector for an Electron–positron Accelerator) is a general-purpose detector concept, designed to study electron–positron collisions in a wide energy range in a very large circular leptonic collider. Its drift chamber is designed to provide an efficient tracking, a high precision momentum measurement and an excellent particle identification by exploiting the application of the cluster counting technique. To investigate the potential of the cluster counting techniques on physics events, a simulation of the ionization cluster generation is needed, therefore we developed an algorithm which can use the energy deposit information provided by the Geant4 toolkit to reproduce, in a fast and convenient way, the cluster number and cluster size distributions. The results obtained confirm that the cluster counting technique allows to reach a resolution two times better than the traditional dE/dx method. A beam test has been performed during November 2021 at CERN on the H8 beam line to validate the simulations results, to define the limiting effects for a fully efficient cluster counting and to count the number of electron clusters released by an ionizing track at a fixed βγ as a function of the track angle. The simulation and the beam test results will be described briefly in this issue.

1872. Clusters of SARS-CoV-2 infection across six schools for children with intellectual and developmental disabilities

Abstract Background In-person learning is important for children with intellectual and developmental disabilities (IDD) because of the additional health, vocational, and functional services for students at these schools. It may be difficult to reduce SARS-CoV-2 transmission in IDD schools because students require assistance with activities of daily living such as eating, during which social distancing and masking cannot occur. Surveillance testing and cluster tracking in schools for children with IDD, which may be considered high-risk environments for transmissions, could have benefits for mitigating transmission and keeping students in schools. The objective of this study was to identify SARS-CoV-2 clusters in IDD specific schools to compare viral transmission in delta and BA.1 variant waves. Methods A saliva-based PCR test was offered to students and staff for weekly SARS-CoV-2 screening at six Special School District (SSD) schools dedicated to children with IDD. Clusters, which are considered 2 or more positives cases in the same classroom having an epidemiological link, were then recorded. All weekly testing took place between November 23, 2020 and May 27, 2022. Clusters were recorded from November 15, 2021 to January 28, 2022. A Fisher's exact test was used to compare categorical variables. Results 545 (90%) and 113 (16%) students participated in weekly testing. 160 participants tested positive throughout the study, 23 (14%) during the delta variant wave and 115 (72%) during the BA.1 variant wave. There was no significant variation in age, race, ethnicity, gender, or vaccination status between positive cases recorded from alpha, delta, and BA.1 variant waves (Table 1). Notably, the vaccination rate of positive participants was lower than the vaccination rate of participants who did not test positive. 42 clusters were recorded, 3 (7%) during the delta variant wave and 39 (93%) during the BA.1 variant wave (Fig. 1). Table 1.Demographics and SARS-CoV-2 Vaccination Status of Positive Cases during Three Variant WavesFigure 1.Reported Counts of Positive Cases and Clusters per Week at SSD Schools Conclusion The highly transmissible BA.1 variant resulted in an increase in clusters observed in IDD specific schools. Mitigation strategies for less transmissible alpha and delta waves were not as effective in reducing transmission during the BA.1 wave. Disclosures Jason Newland, MD, AHRQ: Grant/Research Support|Merck: Grant/Research Support|NIH: Grant/Research Support|PEW Charitable Trust: Grant/Research Support|Pfizer: Grant/Research Support.

Cosmological constraint precision of photometric and spectroscopic multi-probe surveys of China Space Station Telescope (CSST)

ABSTRACT As a Stage IV space-based telescope, the China Space Station Telescope (CSST) can perform photometric and spectroscopic surveys simultaneously to explore the Universe efficiently in extreme precision. In this work, we investigate several powerful CSST cosmological probes, including cosmic shear, galaxy–galaxy lensing, photometric and spectroscopic galaxy clustering, and number counts of galaxy clusters, and study the capability of these probes by forecasting the results of joint constraints on the cosmological parameters. By referring to real observational results, we generate mock data and estimate the measured errors based on CSST observational and instrumental designs. To study systematic effects on the results, we also consider a number of systematics in CSST photometric and spectroscopic surveys, such as the intrinsic alignment, shear calibration uncertainties, photometric redshift uncertainties, galaxy bias, non-linear effects, instrumental effects, etc. The Fisher matrix method is used to derive the constraint results on the cosmological and systematic parameters from individual or joint surveys. We find that the joint constraints achieved by including all these CSST cosmological probes can significantly improve the results from current observations by one order of magnitude at least, which gives Ωm and σ8 &amp;lt;1 per cent accuracy and w0 and wa &amp;lt;5 and 20 per cent accuracy, respectively. This indicates that CSST photometric and spectroscopic multi-probe surveys could provide powerful tools with which to explore the Universe and greatly improve the studies of relevant cosmological problems.

A parallel algorithm for approximating the silhouette using a ball tree

Clustering is widely used in many scientific fields. The contribution of enumerating the value of the silhouette is twofold: firstly it can help choosing a suitable cluster count and secondly it can be used to evaluate the quality of a clustering. Enumerating the silhouette exactly is an extremely time-consuming task, especially in big data applications; it is therefore common to approximate its value. This article presents an efficient shared-memory parallel algorithm for approximating the silhouette, which uses a ball tree. The process of initialising the ball tree and enumerating the silhouette are fully parallelised using the OpenMP API. The results of our experiments show that the proposed parallel algorithm substantially increases the speed of computing the silhouette whilst retaining necessary precision for real-world applications.

SummerTime: Variable-length Time Series Summarization with Application to Physical Activity Analysis

SummerTime seeks to summarize global time-series signals and provides a fixed-length, robust representation of the variable-length time series. Many machine learning methods depend on data instances with a fixed number of features. As a result, those methods cannot be directly applied to variable-length time series data. Existing methods such as sliding windows can lose minority local information. Summarization conducted by the SummerTime method will be a fixed-length feature vector which can be used in place of the time series dataset for use with classical machine learning methods. We use Gaussian Mixture models (GMM) over small same-length disjoint windows in the time series to group local data into clusters. The time series’ rate of membership for each cluster will be a feature in the summarization. By making use of variational methods, GMM converges to a more robust mixture, meaning the clusters are more resistant to noise and overfitting. Further, the model is naturally capable of converging to an appropriate cluster count. We validate our method on a challenging real-world dataset, an imbalanced physical activity dataset with a variable-length time series structure. We compare our results to state-of-the-art studies and show high-quality improvement by classifying with only the summarization.

The Effects of Hydrolyzed Collagen and Collagen Peptide in the Treatment of Superficial Chondral Lesions: An Experimental Study.

Objective To evaluate the effects of hydrolyzed collagen and collagen peptide in the treatment of superficial chondral lesions in rats. Method This research employed 18 Rattus norvegicus . A single intraarticular infiltration of sodium iodoacetate (2 mg solution) through the patellar ligament induced joint damage in previously anesthetized animals. We divided the animals into three groups: a control group, a collagen peptide group, and a hydrolyzed collagen group. Treatment consisted of oral administration of collagen peptide or hydrolyzed collagen for 30 days. Afterwards, we euthanized the animals and studied the joint chondral changes. We evaluated the results according to the chondrocyte clusters count and a histological evaluation, as per Pritzker et al. Results There was no statistical significance in injury stages between the control, hydrolyzed collagen, and collagen peptide groups ( p = 0.11). Regarding scores, there was a statistical significance between the groups treated with hydrolyzed collagen and collagen peptide ( p < 0.05), but not in comparison with the control group. Conclusion The proposed treatments of the induced chondral lesion with the oral administration of hydrolyzed collagen or collagen peptides were effective, resulting in lesion stabilization or regression, and warranting further experimental research to understand and improve the primary outcome of this study.

Cluster counts

Despite the success of the Lambda cold dark matter (ΛCDM) cosmological model, current estimations of the amplitude of matter fluctuations (σ8) show an appreciable difference between its value inferred from the cosmic microwave background (CMB) angular power spectrum (Cℓ) and those obtained from cluster counts. Neutrinos or a modification of the growth of structures had been previously investigated as the possible origin of this discrepancy. In this work we examine whether further extensions to the ΛCDM model could alleviate the tension. To this end, we derived constraints on the parameters subject to the discrepancy, using CMB Cℓ combined with cluster counts from the Sunyaev–Zel’dovich (SZ) sample with a free dark energy equation of state parameter, while allowing the cluster mass calibration parameter (1 − b) to vary. This latter is degenerate with σ8, which translates the discrepancy within the ΛCDM framework into one between (1 − b)∼0.6, corresponding to constraints on σ8 obtained from CMB, and (1 − b)∼0.8, the value adopted for the SZ sample calibration. We find that a constant w, when left free to vary along with large priors on the matter density ([0.1, 1.0]) and the Hubble parameters ([30, 200]), can reduce the discrepancy to less than 2σ for values far below its fiducial w = −1. However, such low values of w are not allowed when we add other probes like the baryonic acoustic oscillation (BAO) feature angular diameter distance measured in galaxy clustering surveys. We also found, when we allow to vary in addition to w a modification of the growth rate through the growth index γ, that the tension is alleviated, with the (1 − b) likelihood now centred around the Planck calibration value of ∼0.8. However, here again, combining CMB and cluster counts with geometrical distance probes restores the discrepancy, with the (1 − b) preferred value reverting back to the ΛCDM value of ∼0.6. The same situation is observed when introducing, along with w and γ, further extensions to ΛCDM (e.g., massive neutrinos), although these extensions reduce the tension to 2σ, even when combined with BAO datasets. We also explore other common extensions by comparing two cases: allowing a dynamical w following a CPL parametrisation in addition to a constant growth index, and when the growth index is expanded through a second parameter γ1 along with a constant w. In the former we reach the same conclusions as with the case of a constant w and γ, where the discrepancy was alleviated only if we do not constrain w by BAO, while in the latter case, we observe that introducing γ1 drives (1 − b) towards lower values that would instead increase the discrepancy on σ8. We conclude that none of these common extensions to ΛCDM is able to fix the discrepancy and a misdetermination of the calibration factor is the most preferred explanation. Finally, we investigate the effect on our posteriors from limiting the Hubble constant priors to the usual common adopted range of [30, 100].

The SiD Digital ECal Based on Monolithic Active Pixel Sensors

The SiD detector concept capitalizes on high granularity in its tracker and calorimeter to achieve the momentum resolution and particle flow calorimetry physics goals in a compact design. The collaboration has had a long interest in the potential for improved granularity in both the tracker and ECal with an application of monolithic active pixel sensors (MAPS) and a study of MAPS in the SiD ECal was described in the ILC TDR. Work is progressing on the MAPS application in an upgraded SiD design with a prototyping design effort for a common SiD tracker/ECal design based on stitched reticules to achieve 10 × 10 cm2 sensors with 25 × 100 micron2 pixels. Application of large area MAPS in these systems would limit delicate and expensive bump-bonding, provide possibilities for better timing, and should be significantly cheaper than the TDR concept due to being a more conventional CMOS foundry process. The small pixels significantly improve shower separation. Recent simulation studies confirm previous performance projections, indicating electromagnetic energy resolution based on digital hit cluster counting provides better performance than the SiD TDR analog design based on 13 mm2 pixels. Furthermore, the two shower separation is excellent down to the millimeter scale. Geant4 simulation results demonstrate these expectations.

AMICO galaxy clusters in KiDS-DR3: Constraints on cosmological parameters and on the normalisation of the mass-richness relation from clustering

Aims. We analysed the clustering of a photometric sample of galaxy clusters selected from the Third Data Release of the Kilo-Degree Survey, focusing on the redshift-space two-point correlation function (2PCF). We compared our measurements to theoretical predictions of the standard Λ cold dark matter (ΛCDM) cosmological model. Methods. We measured the 2PCF of the sample in the cluster-centric radial range r ∈ [5, 80] h−1 Mpc, considering 4934 galaxy clusters with richness λ* ≥ 15 in the redshift range z ∈ [0.1, 0.6]. A Markov chain Monte Carlo analysis has been performed to constrain the cosmological parameters Ωm, σ8, and S8 ≡ σ8(Ωm/0.3)0.5, assuming Gaussian priors on the mass-richness relation given by the posteriors obtained from a joint analysis of cluster counts and weak lensing. In addition, we constrained the normalisation of the mass-richness relation, α, with fixed cosmological parameters. Results. We obtained Ωm = 0.28+0.05−0.04, σ8 = 0.82+0.14−0.12, and S8 = 0.80+0.08−0.08. The constraint on S8 is consistent within 1σ with the results from WMAP and Planck. Furthermore, by fixing the cosmological parameters to those provided by Planck, we obtained α = 0.12+0.06−0.06, which is fully consistent with the result obtained from the joint analysis of cluster counts and weak lensing performed for this sample.

A high‐speed unsupervised hardware architecture for rapid diagnosis of COVID‐19

SummaryIn the diagnosis of COVID‐19, investigation, analysis, and automatic counting of blood cell clusters are the most essential steps. Currently employed methods for cell segmentation, identification, and counting are time‐consuming and sometimes performed manually from sampled blood smears, which is hard and needs the support of an expert laboratory technician. The conventional method for the blood‐count‐test is by automatic hematology analyzer which is quite expensive and slow. Moreover, most of the unsupervised learning techniques currently available presume the medical practitioner to have a prior knowledge regarding the number and action of possible segments within the image before applying recognition. This assumption fails most often as the severity of the disease gets increased like the advanced stages of COVID‐19, lung cancer etc. In this manuscript, a simplified automatic histopathological image analysis technique and its hardware architecture suited for blind segmentation, cell counting, and retrieving the cell parameters like radii, area, and perimeter has been identified not only to speed up but also to ease the process of diagnosis as well as prognosis of COVID‐19. This is achieved by combining three algorithms: the K‐means algorithm, a novel statistical analysis technique‐HIST (histogram separation technique), and an islanding method an improved version of CCA algorithm/blob detection technique. The proposed method is applied to 15 chronic respiratory disease cases of COVID‐19 taken from high profile hospital databases. The output in terms of quantitative parameters like PSNR, SSIM, and qualitative analysis clearly reveals the usefulness of this technique in quick cytological evaluation. The proposed high‐speed and low‐cost architecture gives promising results in terms of performance of 190 MHz clock frequency, which is two times faster than its software implementation.

PClusBA: A Novel Partition Clustering-Based Biometric Authentication Mechanism

In any communication paradigm, authentication plays an important role. Common authentication methods such as passwords are vulnerable to brute force attacks and cryptanalysis. Authentication mechanisms based on biometrics prove to be a better alternative owing to their permanence, uniqueness, and non-reproducibility. Fingerprint biometrics is more suitable in the context of network security applications due to its uniqueness and ease of processing. In general, each fingerprint has 30–40 unique points. The lesser the number of unique points extracted from the fingerprint, the more flexible it is to represent a Unique ID (UID). Clustering can be used to prune the unique point count. Since partition clustering gives us the choice of selecting the cluster count apriori, we have used partition clustering in the proposed biometric authentication mechanism named PClusBA. The cluster count is set to eight to obtain a UID of 192 bits in the proposed format. The length of the UID conforms to the NIST standards for enhancing security in commercial applications. The algorithm was implemented and tested using images from DB1 FVC 2002 database and the results were found to be unique for each image.

Machine learning approach identified clusters for patients with low cardiac output syndrome and outcomes after cardiac surgery.

Low cardiac output syndrome (LCOS) is the most serious physiological abnormality with high mortality for patients after cardiac surgery. This study aimed to explore the multidimensional data of clinical features and outcomes to provide individualized care for patients with LCOS. The electronic medical information of the intensive care units (ICUs) was extracted from a tertiary hospital in South China. We included patients who were diagnosed with LCOS in the ICU database. We used the consensus clustering approach based on patient characteristics, laboratory data, and vital signs to identify LCOS subgroups. The consensus clustering method involves subsampling from a set of items, such as microarrays, and determines to cluster of specified cluster counts (k). The primary clinical outcome was in-hospital mortality and was compared between the clusters. A total of 1,205 patients were included and divided into three clusters. Cluster 1 (n = 443) was defined as the low-risk group [in-hospital mortality =10.1%, odds ratio (OR) = 1]. Cluster 2 (n = 396) was defined as the medium-risk group [in-hospital mortality =25.0%, OR = 2.96 (95% CI = 1.97-4.46)]. Cluster 3 (n = 366) was defined as the high-risk group [in-hospital mortality =39.2%, OR = 5.75 (95% CI = 3.9-8.5)]. Patients with LCOS after cardiac surgery could be divided into three clusters and had different outcomes.

Modeling clustered count data with discrete weibull regression model

Modeling clustered count data with discrete weibull regression model