Moderate Number Research Articles

Rheology of Bingham viscoplastic flow triggered by a rotating and radially stretching disk

Purpose This study aims to investigate the flow and heat transfer characteristics of a Bingham viscoplastic fluid subjected to the combined effects of axial rotation and radial stretching of a circular disk. Building upon existing models for Bingham fluids on stationary walls, we extend the formulation to incorporate the effects of a linearly stretching disk using von Kármán similarity transformations. Design/methodology/approach The resulting system of nonlinear ordinary differential equations is solved to characterize the flow and thermal fields. Three dimensionless parameters govern the momentum layer: a swirling number capturing the balance between rotation and stretching, a Bingham number characterizing the fluid’s yield stress and a modified Reynolds number incorporating the disk stretching. The Prandtl number controls the thermal response. Findings For purely stretching flows, a two-dimensional flow structure emerges. However, the introduction of rotation induces three-dimensional flow behavior. Unlike previous studies suggesting that moderate Bingham numbers are sufficient for non-Newtonian effects on purely revolving disks, the findings indicate that significantly higher yield stresses are required to observe non-Newtonian characteristics under radial stretching conditions. This difference can be attributed to the enhancing influence of wall movement on the fluid dynamics. At high Bingham numbers, a two-layer flow structure develops, comprising an unyielded plug region above the disk and a yielded shear layer adjacent to the wall. The von Kármán viscous pump mechanism drives the Bingham flow within this regime. Originality/value Physical quantities such as drag force due to wall shear stress, torque resulting from tangential shear stress and Nusselt number are extracted from the quantitative data.

Privacy-ensuring Open-weights Large Language Models Are Competitive with Closed-weights GPT-4o in Extracting Chest Radiography Findings from Free-Text Reports.

Background Large-scale secondary use of clinical databases requires automated tools for retrospective extraction of structured content from free-text radiology reports. Purpose To share data and insights on the application of privacy-preserving open-weights large language models (LLMs) for reporting content extraction with comparison to standard rule-based systems and the closed-weights LLMs from OpenAI. Materials and Methods In this retrospective exploratory study conducted between May 2024 and September 2024, zero-shot prompting of 17 open-weights LLMs was preformed. These LLMs with model weights released under open licenses were compared with rule-based annotation and with OpenAI's GPT-4o, GPT-4o-mini, GPT-4-turbo, and GPT-3.5-turbo on a manually annotated public English chest radiography dataset (Indiana University, 3927 patients and reports). An annotated nonpublic German chest radiography dataset (18 500 reports, 16 844 patients [10 340 male; mean age, 62.6 years ± 21.5 {SD}]) was used to compare local fine-tuning of all open-weights LLMs via low-rank adaptation and 4-bit quantization to bidirectional encoder representations from transformers (BERT) with different subsets of reports (from 10 to 14 580). Nonoverlapping 95% CIs of macro-averaged F1 scores were defined as relevant differences. Results For the English reports, the highest zero-shot macro-averaged F1 score was observed for GPT-4o (92.4% [95% CI: 87.9, 95.9]); GPT-4o outperformed the rule-based CheXpert [Stanford University] (73.1% [95% CI: 65.1, 79.7]) but was comparable in performance to several open-weights LLMs (top three: Mistral-Large [Mistral AI], 92.6% [95% CI: 88.2, 96.0]; Llama-3.1-70b [Meta AI], 92.2% [95% CI: 87.1, 95.8]; and Llama-3.1-405b [Meta AI]: 90.3% [95% CI: 84.6, 94.5]). For the German reports, Mistral-Large (91.6% [95% CI: 90.5, 92.7]) had the highest zero-shot macro-averaged F1 score among the six other open-weights LLMs and outperformed the rule-based annotation (74.8% [95% CI: 73.3, 76.1]). Using 1000 reports for fine-tuning, all LLMs (top three: Mistral-Large, 94.3% [95% CI: 93.5, 95.2]; OpenBioLLM-70b [Saama]: 93.9% [95% CI: 92.9, 94.8]; and Mixtral-8×22b [Mistral AI]: 93.8% [95% CI: 92.8, 94.7]) achieved significantly higher macro-averaged F1 score than did BERT (86.7% [95% CI: 85.0, 88.3]); however, the differences were not relevant when 2000 or more reports were used for fine-tuning. Conclusion LLMs have the potential to outperform rule-based systems for zero-shot "out-of-the-box" structuring of report databases, with privacy-ensuring open-weights LLMs being competitive with closed-weights GPT-4o. Additionally, the open-weights LLM outperformed BERT when moderate numbers of reports were used for fine-tuning. Published under a CC BY 4.0 license. Supplemental material is available for this article. See also the editorial by Gee and Yao in this issue.

Вычисление цен опционов в модели Хестона с помощью искусственных нейронных сетей

The article considers numerical methods for pricing options in the Heston model. The main attention is paid to solving the Cauchy problem for the partial differential equation for option prices using artificial neural networks. The possibility of approximating the solution using feedforward neural networks with one hidden layer and sigmoid activation function is substantiated. The advantage of the proposed method is the ability to explicitly calculate loss functions when choosing the logistic regression as an activation function. The loss function of neural networks tracks errors in the execution of equations for option prices and approximation of initial conditions. For comparison method of solving the boundary value problem, Monte-Carlo method and method based on recurrent formulas on a binary tree are used. Computational experiments show that even with a moderate number of neurons in the hidden mode, the neural network approximates the prices of European in-the-money options quite well.

Evaluation of agreement between common clustering strategies for DNA methylation-based subtyping of breast tumours.

Clustering algorithms have been widely applied to tumor DNA methylation datasets to define methylation-based cancer subtypes. This study aimed to evaluate the agreement between subtypes obtained from common clustering strategies. We used tumor DNA methylation data from 409 women with breast cancer from the Melbourne Collaborative Cohort Study (MCCS) and 781 breast tumors from The Cancer Genome Atlas (TCGA). Agreement was assessed using the adjusted Rand index for various combinations of number of CpGs, number of clusters and clustering algorithms (hierarchical, K-means, partitioning around medoids, and recursively partitioned mixture models). Inconsistent agreement patterns were observed for between-algorithm and within-algorithm comparisons, with generally poor to moderate agreement (ARI <0.7). Results were qualitatively similar in the MCCS and TCGA, showing better agreement for moderate number of CpGs and fewer clusters (K = 2). Restricting the analysis to CpGs that were differentially-methylated between tumor and normal tissue did not result in higher agreement. Our study highlights that common clustering strategies involving an arbitrary choice of algorithm, number of clusters and number of methylation sites are likely to identify different DNA methylation-based breast tumor subtypes.

A Leadfield-Free Optimization Framework for Transcranially Applied Electric Currents.

Transcranial Electrical Stimulation (TES), Temporal Interference Stimulation (TIS), Electroconvulsive Therapy (ECT) and Tumor Treating Fields (TTFields) are based on the application of electric current patterns to the brain. The optimal electrode positions, shapes and alignments for generating a desired current pattern in the brain vary between persons due to anatomical variability. The aim is to develop a flexible and efficient computational approach to determine individually optimal montages based on electric field simulations. We propose a leadfield-free optimization framework that allows the electrodes to be placed freely on the head surface. It is designed for the optimization of montages with a low to moderate number of spatially extended electrodes or electrode arrays. Spatial overlaps are systematically prevented during optimization, enabling arbitrary electrode shapes and configurations. The approach supports maximizing the field intensity in target region-of-interests (ROI) and optimizing for a desired focality-intensity tradeoff. We demonstrate montage optimization for standard two-electrode TES, focal center-surround TES, TIS, ECT and TTFields. Comparisons against reference simulations are used to validate the performance of the algorithm. The system requirements are kept moderate, allowing the optimization to run on regular notebooks and promoting its use in basic and clinical research. The new framework complements existing optimization methods that require small electrodes, a predetermined discretization of the electrode positions on the scalp and work best for multi-channel systems. It strongly extends the possibilities to optimize electrode montages towards application-specific aims and supports researchers in discovering innovative stimulation schemes. The framework is available in SimNIBS.

Effects of rolling on eating quality, starch structure, and water distribution in cooked indica rice dough.

Given the composition of rice and its lack of gluten proteins, rice flour fails to form a cohesive and elastic dough when mixed directly with water. Consequently, many rice products rely on rice sheets (RS) made by rolling cooked rice dough. Limited research exists on how the rolling process impacts the properties and structure of cooked indica rice dough. This study investigated the effect of the number of rolling passes on the eating quality, starch structure, and water distribution of cooked fermented indica RS formed by dough. When the number of rolling passes reached six, the RS (RP-6) that was obtained exhibited the lowest cooking loss, the highest hardness, adhesiveness, and chewiness, and optimal stretchability. It also demonstrated the lowest water loss after freezing. Dense microstructures were observed on both the surface and cross-section of RP-6. More ordered starch crystal structures and double helix structures were formed. The relative peak area of tightly bound water significantly increased in RP-6, indicating a stronger bonding status between the starch and water molecules. However, excessive rolling passes (more than six) led to a partial disruption of the internal RS structure, resulting in a decline in eating quality. The study demonstrated the importance of the rolling process in improving the performance of RS. It was found that a moderate number of rolling passes was conducive to producing excellent RS, providing a theoretical basis for the production of high-quality rice-based products such as rice noodles, dumplings, and cakes. © 2024 Society of Chemical Industry.

Expanding subtitles: An exploration into scaffolding and deepening educational content through subtitling

Educational videos have proliferated in recent years due to increased demands for online content in massive open online courses (MOOC) and blended classroom and pandemic contexts. While these videos are often subtitled to provide greater access, questions remain about whether the educational content of these videos can be deepened. This qualitative study initiates exploration into whether subtitles could and/or should be expanded, providing more in-depth information and thus bridging gaps in background knowledge and deepening comprehension of the video content. The study suggests expanding subtitles by creating clickable keywords and concepts within the subtitles. The findings indicate that higher-risk participants who accessed a moderate number of expansions (approximately one every 90 seconds) had better comprehension outcomes than lower-risk participants who either did not access expansions, or accessed a high number of expansions (approximately one every 50 seconds). Furthermore, the study finds that there was a positive attitude towards expanded subtitles, with 60% of participants indicating a desire to see more expanded subtitles. These findings suggest that future research taking a quantitative approach is necessary to determine the full educational value of this subtitle format.

Morphometric Analysis of Olanthichira Watershed using Geographic Information System in Malappuram District, Kerala, India

Morphometric analysis of watershed is important for understanding its geo-hydrological behaviour. The objective of this study is to determine morphometric parameters of Olanthichira watershed, Kuttipuram block, Malappuram district of state Kerala, India. The stream channels of the watershed were extracted from SRTM DEM and SOI topographical maps on 1:50,000 scale using Hydrology tools in Arc GIS software showing dendritic drainage pattern. Then morphometric analysis is carried out by determining parameters of linear, areal and relief aspects. Drainage density of 1.836 km/km2 is moderate suggests semi-permeable rocks, moderate relief, and a balanced hydrological response. The drainage texture of watershed is 4.359/km indicates a moderate to fine drainage texture means the watershed has a fairly advanced drainage system, with a moderate number of stream channels with highest channel maintenance constant of 0.544 experiencing low runoff and fewer structural disruptions. The moderate bifurcation ratio of 3.68 in a drainage network suggests that the network's development is influenced by some degree of structural control. The areal aspects of the watershed point out elongated shape which possess lower erosion and sediment transport capacities and less prone to sudden floods but might require interventions to prevent sediment deposition in channels. The results of morphometric analysis along with slope map, aspect map and vegetation (NDVI) map are helpful for decision makers to implement soil and water conservation measures for sustainable management of the resources in watershed.

Performance Comparison of RANSAC and Other Model Estimation Methods in Panoramic Image Mosaic

As computer vision technology advances, the significance of outlier processing algorithms increases across various applications. Despite this, there is a lack of comprehensive comparisons to assess the performance of these algorithms. This paper focuses on evaluating the principles, advantages, and disadvantages of three key models: RANSAC, least squares, and lmed. Through experimental testing in diverse scenarios, the study identifies the most suitable model for different contexts. Results indicate that RANSAC exhibits the highest robustness against outliers, while lmed performs effectively with a moderate number of outliers. These findings are crucial for selecting appropriate models tailored to specific applications, ultimately enhancing the quality of panoramic mosaic images. The comparative analysis presented in this paper aims to guide practitioners in choosing the best outlier processing techniques based on their specific requirements and application environments, thereby improving the robustness and accuracy of computer vision systems.

Design and optimization of steam power systems in industrial parks based on the distributed steam turbine system

Steam power systems (SPSs) in industrial parks are the typical utility systems for heat and electricity supply. In SPSs, electricity is generated by steam turbines, and steam is generally produced and supplied at multiple levels to serve the heat demands of consumers with different temperature grades, so that energy is utilized in cascade. While a large number of steam levels enhances energy utilization efficiency, it also tends to cause a complex steam pipeline network in the industrial park. In practice, a moderate number of steam levels is always adopted in SPSs, leading to temperature mismatches between heat supply and demand for some consumers. This study proposes a distributed steam turbine system (DSTS) consisting of main steam turbines on the energy supply side and auxiliary steam turbines on the energy consumption side, aiming to balance the heat production costs, the distance-related costs, and the electricity generation of SPSs in industrial parks. A mixed-integer nonlinear programming model is established for the optimization of SPSs, with the objective of minimizing the total annual cost (TAC). The optimal number of steam levels and the optimal configuration of DSTS for an industrial park can be determined by solving the model. A case study demonstrates that the TAC of the SPS is reduced by 220.6 k$ (2.21%) through the arrangement of auxiliary steam turbines. The sub-optimal number of steam levels and a non-optimal operating condition slightly increase the TAC by 0.46% and 0.28%, respectively. The sensitivity analysis indicates that the optimal number of steam levels tends to decrease from 3 to 2 as electricity price declines.

Distributed Temporal Coding of Visual Memory Categories in Human Hippocampal Neurons.

The hippocampus is crucial for forming new episodic memories. While the encoding of spatial and temporal information (where and when) in the hippocampus is well understood, the encoding of objects (what) remains less clear due to the high dimensions of object space. Rather than encoding each individual object separately, the hippocampus may instead encode categories of objects to reduce this dimensionality. In this study, we developed and applied a combined experimental-modeling approach to investigate how the hippocampus encodes visual memory categories in humans. We recorded spikes from hippocampal CA3 and CA1 neurons in 24 epilepsy patients performing a visual delayed match-to-sample (DMS) task involving five image categories. An ensemble multi-temporal-resolution classification model was employed to decode these visual memory categories from the hippocampal spiking activity with moderate numbers of trials. This model enables the identification of the spatio-temporal characteristics of hippocampal encoding through its interpretable representations. Using this model, we estimated the optimal temporal resolutions for decoding each visual memory category for each neuron in the ensemble. Results indicate that visual memory categories can be decoded from hippocampal spike patterns despite the short data length, supporting the presence of category-specific coding in the human hippocampus. We found that hippocampal neuron ensembles encode visual memory categories in a distributed manner, akin to a population code, while individual neurons use a temporal code. Additionally, CA3 and CA1 neurons exhibit similar and redundant information regarding visual memory categories, likely due to the strong and diffuse feedforward synaptic connections from the CA3 region to the CA1 region.

Trajectory of Multiple Chronic Conditions and Associated Factors Among Noninstitutionalized Adults Aged 60 Years or Older in Southern Brazil.

The prevalence of chronic conditions is increasing worldwide. The objective of this study was to describe the trajectory of the occurrence of multiple chronic conditions during 6 years of follow-up and investigate their association with demographic, socioeconomic, and behavioral health characteristics of older adults in Southern Brazil. We used data from a longitudinal study (the Como Vai? study) of noninstitutionalized adults aged 60 or older living in the urban area of Pelotas, Rio Grande do Sul. We assessed the number of chronic conditions based on a list of 24 conditions in 3 interviews, conducted in 2014, 2016-2017, and 2019-2020. We used group-based semiparametric modeling to identify groups of participants based on the number of chronic conditions. For associations with participant characteristics, we performed multinomial logistic regression and considered a low, moderate, and high burden of chronic conditions. Of the 1,451 older adults in the cohort, 1,098 (75.7%) were included in analysis. Almost one-third (30.9%) had a low burden (2.3 conditions), more than half (52.0%) had a moderate burden (5.6 conditions), and 17.1% had a high burden (9.7 conditions). Men (relative risk [RR] = 6.10; 95% CI, 3.64-10.22), those aged 80 years or older (RR = 2.33; 95% CI, 1.15-4.72), those with no education (RR= 4.78; 95% CI, 2.19-10.45), and former smokers (RR = 1.53; 95% CI, 0.96-2.44) had a higher risk of being classified in the high-burden group than in the low-burden group. Most older adults belonged to the group with a moderate number of chronic conditions. Several sociodemographic characteristics were associated with belonging to the trajectory with a greater number of conditions.

Analyzing the performance of deep convolutional neural network models for weed identification in potato fields

Weeds pose a significant and fundamental challenge in agriculture, competing with crops for vital resources such as water, nutrients, and sunlight. This competition often leads to reduced crop yields and diminished quality of produce. Additionally, weeds can host pests and diseases that further harm crops, increasing the risk of infestation and reducing farm productivity. Accurate weed identification through deep learning offers a solution, enabling farmers to implement site-specific herbicide spraying, thus lowering herbicide usage and minimizing environmental impact. This study introduces a benchmark crop and weed classification dataset and evaluates seven state-of-the-art deep-learning models for weed identification. The dataset was obtained from potato fields in Punjab, India, over two consecutive growth seasons (2022 and 2023). Seven deep learning models, Convolution Neural Network (CNN)-11, CNN-14, Inceptionv3, AlexNet, VGG16, ResNet50, and the YOLOv8 were trained and tested on this dataset for potato and weed classification. Among these models, YOLOv8 emerges as the top performer, achieving flawless accuracy of 100% with 37.5 million parameters. The custom CNN-11 model, despite having the fewest parameters (2.2 million), achieves 52% accuracy, making it suitable for resource-constrained environments. ResNet50, with its residual networks, also demonstrates exceptional performance with 99% accuracy and a moderate number of parameters (23 million), which can be a significant consideration in environments with limited resources or when deploying models on edge devices. These findings guide researchers and practitioners in selecting optimal models to reduce herbicide usage, minimize environmental impact, and enhance precision agriculture practices. Ultimately, this study advances weed management strategies, supporting sustainable crop management and improving agricultural productivity.

Flow disturbance by intrusive instrumentation located on the surface of a compressor blade

In this paper, we present an experimental and numerical investigation into the flow disturbance by intrusive instrumentation in the form of spanwise pressure tubes mounted on the pressure and suction side of a compressor blade in a linear cascade wind tunnel at a moderate subsonic Mach number of 0.5 and a Reynolds number of 790,000. The pressure tubes were modelled as simplified bumps extending over the entire span with a width of 24% of the chord length and a height of 67% of the maximum blade thickness. The leading edge of the bumps is located at 30% of the streamwise surface length. We found that the total pressure losses generated by these pressure tubes are considerably larger when they are placed on the suction side of the blade. The suction-sided bump increases the total pressure loss coefficient by a factor of four, while the pressure-sided bump approximately doubles it. CFD simulations reveal that the suction-sided bump causes the flow to form a large, open recirculation zone, while for the pressure-sided bump, the flow reattaches near the trailing edge. In conclusion, we provide insights into the effects of pressure tubes on the losses generated by compressor blades. The mounted tubes significantly impact the flow and should be carefully considered during instrumentation design.

Effects of turbulence boundary conditions on Spalart-Allmaras RANS simulations for active flow control applications

We assess the suitability of Reynolds-Averaged Navier–Stokes (RANS) simulation using the Spalart-Allmaras (SA) turbulence model as a closure in analysing the performance of fluidic Active Flow Control (AFC) applications. In particular, we focus on the optimal set of actuation parameters found by Tousi et al. (Appl Math Model, 2021) and Tousi et al. (Aerospace Sci Technol 127:107679, 2022) for a SD7003 airfoil at a Reynolds number Re=6×104 and post-stall angle of attack α=14∘ fitted with a Synthetic Jet Actuator (SJA). The Large Eddy Simulation (LES) presented in that work is taken as the reference to identify the best choice of boundary conditions for the turbulence field ν~ at both domain inlet and jet orifice in two-dimensional SA-RANS computations. Although SA-RANS is far less accurate than LES, our findings show that it can still predict macroscopic aggregates such as lift and drag coefficients quite satisfactorily and at a much lower computational cost, provided that turbulence levels of the actuator jet are set to a realistic value. An adequate value of ν~ is instrumental in capturing the correct flow behaviour of the reattached boundary layers for close-to-optimal actuated cases. This validates the use of RANS-SA as a reliable and cost-effective simulation method for the preliminary optimisation of SJA parameters in AFC applications, provided that thorough sensitivity analysis on turbulence-related boundary conditions is performed. Given the strong sensitivity of flow detachment on the laminar or turbulent nature of boundary layers, our results suggest that such analyses are particularly indispensable for vastly separated flow scenarios in general, notably for bluff bodies at moderate transcritical Reynolds numbers.

The detection of distomolar teeth on panoramic radiographs using different artificial intelligence models

PurposesOne notable anomaly, presence of distomolars, arises beyond the typical sequence of the human dental system. In this study, convolutional neural networks (CNNs) based machine learning methods were employed to classify distomolar tooth existence using panoramic radiography (PR). MethodsPRs dataset, composed of 117 subjects with distomolar teeth and 146 subjects without distomolar teeth, was constructed. These images were assessed using AlexNet, DarkNet, DenseNet, EfficientNet, GoogLeNet, ResNet, MobileNet, NasNet-Mobile, VGG, and XceptionNet frameworks for distomolar teeth existence. Considering the moderate number dataset samples, transfer learning was also utilized to improve the performance of these CNN based networks along with 5-fold cross-validation. The final classification was obtained through the fusion of the classifiers results. ResultsPerformance of the experimental studies was assessed utilizing accuracy (Acc), sensitivity (sen), specificity (spe) and precision (pre) metrics. Best accuracy value of 96.2 % was obtained for the fusion of DarkNet, DenseNet, and ResNet, three best individual performing architectures, in distomolar teeth classification problem. Conclusion and Practical ImplicationsIn summary, this study has demonstrated the significant potential of CNNs in accurately detecting distomolar teeth in dental radiographs, a critical task for clinical diagnosis and treatment planning. The fusion of CNN architectures, particularly ResNet, Darknet, and DenseNet, has shown exceptional performance, pointing towards the future of artificial intelligence (AI) driven dental diagnostics. Our findings showed that these systems can help clinicians during radiologic examinations.

Relationship between grading of invasive ductal breast carcinoma at RSPAL Dr. Ramelan Surabaya and tumor infiltrating lymphomocytes (TILs) period January 2020–December 2022

Worldwide, including in Indonesia, breast carcinoma is the most frequent form of cancer among women. Out of 396,914 cancer patients in Indonesia, 68,858 (16.6%) were diagnosed with breast cancer. An essential component of the immune response to cancer, Tumor Infiltrating Lymphocytes (TILs) are receiving more and more attention in studies on breast carcinoma. Groups of immune cells, including T lymphocytes, are discovered in tumor tissue and are referred to as TILs. Prognostic information can be gleaned from TIL existence, composition, and activity, according to substantial evidence. Grading, shows the degree of tumor aggressiveness and is crucial for treatment planning, is an essential part of breast cancer. We still do not completely understand the significance of TILs in deciding grading of invasive ductal carcinoma, one of the most common kinds of breast carcinoma. We set out to document the correlation between TILs and invasive ductal breast cancer grading from 2020 to 2022. The study used a retrospective cross-sectional analysis as its methodology. RSPAL Dr. Ramelan Surabaya surveyed patients diagnosed with invasive ductal carcinoma utilizing a total sampling technique. The patient preparations for invasive ductal carcinoma were cut into paraffin blocks, stained, and then separated into three groups. Patients with grade 1 invasive ductal carcinoma were prepared in group 1, those with grade 2 invasive ductal carcinoma in group 2, and patients with grade 3 invasive ductal carcinoma in group 3. Subsequently, TILs were counted under a microscope for each group. There were 28% TILs in the mild group, 44% in the moderate group, and 28% in the high group of patients with grade 1 invasive ductal breast cancer. Patients with grade 2 invasive ductal breast cancer were categorized as having a low number of TILs 16%, a moderate number 40% or a high number 44%. Subjects with grade 3 invasive ductal breast cancer were categorized as having a low number of TILs 44%, a moderate number 24%, or a high number 32%. The study found a highly significant (p = 0.001) correlation between the TILs variable and the grade of invasive ductal breast cancer. This correlation was shown to be unidirectional.

Implementation of ward's agglomerative hierarchical clustering model to detect pulmonary tuberculosis endemic areas in Aceh Utara regency

This study aims to detect pulmonary Tuberculosis (TB) endemic areas in Aceh Utara Regency based on altitude, population density, and the number of TB cases. The Agglomerative Hierarchical Clustering (AHC) algorithm was used to cluster 27 subdistricts, each measured by these three factors. The clustering results divided the subdistricts into three main clusters with distinct characteristics. Cluster 1 consists of subdistricts with low altitude, high population density, and relatively high numbers of TB cases, identifying this area as having the highest risk of TB endemicity. Cluster 2 includes areas with moderate population density and TB case numbers, while Cluster 3 consists of subdistricts at higher altitudes with fewer TB cases. The clustering results were evaluated using three key metrics: Silhouette Score, Davies-Bouldin Index, and Dunn Index, which indicated that the clustering model performed well, although some subdistricts were positioned near the cluster boundaries. This research provides valuable information for health authorities to prioritize interventions and allocate resources to areas most in need of TB management.

Nonlinear Perturbations and Weak Shock Waves in Isentropic Atmospheres

Acoustic perturbations to stellar envelopes can lead to the formation of weak shock waves via nonlinear wave steepening. Close to the stellar surface, the weak shock wave increases in strength and can potentially lead to the expulsion of part of the stellar envelope. While accurate analytic solutions to the fluid equations exist in the limits of low-amplitude waves or strong shocks, connecting these phases generally requires simulations. We address this problem using the fact that the plane-parallel Euler equations, in the presence of a constant gravitational field, admit exact Riemann invariants when the flow is isentropic. We obtain exact solutions for acoustic perturbations and show that after they steepen into shock waves, Whitham’s approximation can be used to solve for the shock’s dynamics in the weak to moderately strong regimes, using a simple ordinary differential equation. Numerical simulations show that our analytic shock approximation is accurate up to moderate (∼few–15) Mach numbers, where the accuracy increases with the adiabatic index.

Dynamics of an oblate spheroidal particle in a square duct filled with viscoelastic fluids

Herein, we used the fictitious domain method to numerically investigate the lateral migration and rotation of an oblate spheroidal particle in a square duct filled with Oldroyd-B fluids. We adopted Reynolds numbers ranging from 25 to 100 and Weissenberg numbers from 0.01 to 0.50. At low to moderate Weissenberg numbers (Wi ≤ 0.50), viscous forces remain dominant in particle motion. Additionally, we considered the effects of initial lateral position, orientation, and blocking ratio on particle dynamics. The results indicate that for flow in square channels with finite fluid inertia, as Wi increases, the elastic effects gradually strengthen, causing the equilibrium position of the particles to shift from near the centerline of the channel toward the diagonal. Notably, under significant fluid elasticity conditions, additional equilibrium positions emerge in the corners of the channel. When released with their x0–y0 plane (containing the two longest axes of the oblate spheroid) parallel to the x–y plane (duct cross section) of the flow field, particles exhibited three distinct motion modes: tumbling, rolling, and kayaking. Tumbling was influenced by fluid inertia and corner attraction, which exhibited transitions to rolling or kayaking. The study also emphasized that the initial orientation of the particles impacted their sustained tumbling under low inertial flows. In addition, the blockage ratio (the ratio of the equivalent diameter of the particle to the duct height) mainly affected the equilibrium positions, and particles with a blockage ratio β ≤ 0.125 were readily attracted to the corners.