Class Of Patterns Research Articles

Characterizing cannabis use reduction and change in functioning during treatment: Initial steps on the path to new clinical endpoints.

Reduction-based cannabis use endpoints are needed to better evaluate treatments for cannabis use disorder (CUD). This exploratory, secondary analysis aimed to characterize cannabis frequency and quantity reduction patterns and corresponding changes in psychosocial functioning during treatment. We analyzed 16 weeks (4 prerandomization, 12 postrandomization) of data (n = 302) from both arms of a randomized clinical trial assessing pharmacotherapy for CUD. Cannabis consumption pattern classes were extracted with latent profile modeling using self-reported (a) past-week days used (i.e., frequency) and (b) past-week average grams used per using day (i.e., quantity). Changes in mean Marijuana Problem Scale (MPS) and Hospital Anxiety and Depression Scale (HADS) scores were examined among classes. Urine cannabinoid levels were examined in relation to self-reported consumption as a validity check. Two-, three-, four-, and five-class solutions each provided potentially useful conceptualizations of associations between frequency and quantity. Regardless of solution, reductions in MPS scores varied in magnitude across classes and closely tracked class-specific reductions in consumption (e.g., larger MPS reduction corresponded to larger frequency/quantity reductions). Changes in HADS scores were less pronounced and less consistent with consumption patterns. Urine cannabinoid levels closely matched class-specific self-reported consumption frequency. Findings illustrate that frequency and quantity can be used in tandem within mixture model frameworks to summarize heterogeneous cannabis use reduction patterns that may correspond to improved psychosocial functioning. Going forward, similar analytic strategies applied to alternative metrics of cannabis consumption may facilitate construction of useful reduction-based clinical endpoints. (PsycInfo Database Record (c) 2022 APA, all rights reserved).

A Review of Pattern Recognition Using Statistics

Abstract: Pattern recognition's primary objective is supervised or classification techniques. The statistical method has been the most intensively researched and implemented in practice among the various frameworks in which classification technique has typically been developed. In recent times, neural network techniques and methods based on statistical learning theory have gotten a lot of attention. The following issues must be carefully considered when designing a recognition system: pattern classes, sensing environment, pattern representation, feature extraction and selection, cluster analysis, classifier design, learning, training, test sample selection, and performance evaluation. The overall challenge of detecting intricate forms with modifiable direction, position, and size has not been addressed despite just Fifty years of intensive research. Data mining, web browsing, recovery of multimedia content, face recognition, and cursive handwriting recognition are just a couple of minor new and emerging applications that require strong and fast pattern recognition methods. The intention of this review paper is to summarize and analyze some of the adopted approaches in various phases of a recognition system, as well as to identify areas of research and applications that are at the forefront of innovation in this exciting and challenging field.

Vaccination with Mycoplasma pneumoniae membrane lipoproteins induces IL-17A driven neutrophilia that mediates Vaccine-Enhanced Disease

Bacterial lipoproteins are an often-underappreciated class of microbe-associated molecular patterns with potent immunomodulatory activity. We previously reported that vaccination of BALB/c mice with Mycoplasma pneumoniae (Mp) lipid-associated membrane proteins (LAMPs) resulted in lipoprotein-dependent vaccine enhanced disease after challenge with virulent Mp, though the immune responses underpinning this phenomenon remain poorly understood. Herein, we report that lipoprotein-induced VED in a mouse model is associated with elevated inflammatory cytokines TNF-α, IL-1β, IL-6, IL-17A, and KC in lung lavage fluid and with suppurative pneumonia marked by exuberant neutrophilia in the pulmonary parenchyma. Whole-lung-digest flow cytometry and RNAScope analysis identified multiple cellular sources for IL-17A, and the numbers of IL-17A producing cells were increased in LAMPs-vaccinated/Mp-challenged animals compared to controls. Specific IL-17A or neutrophil depletion reduced disease severity in our VED model—indicating that Mp lipoproteins induce VED in an IL-17A-dependent manner and through exuberant neutrophil recruitment. IL-17A neutralization reduced levels of TNF-α, IL-1β, IL-6, and KC, indicating that IL-17A preceded other inflammatory cytokines. Surprisingly, we found that IL-17A neutralization impaired bacterial clearance, while neutrophil depletion improved it—indicating that, while IL-17A appears to confer both maladaptive and protective responses, neutrophils play an entirely maladaptive role in VED. Given that lipoproteins are found in virtually all bacteria, the potential for lipoprotein-mediated maladaptive inflammatory responses should be taken into consideration when developing vaccines against bacterial pathogens.

Characterizing spatial point processes by percolation transitions

A set of discrete individual points located in an embedding continuum space can be seen as percolating or non-percolating, depending on the radius of the discs/spheres associated with each of them. This problem is relevant in theoretical ecology to analyze, e.g., the spatial percolation of a tree species in a tropical forest or a savanna. Here, we revisit the problem of aggregating random points in continuum systems (from 2 to 6-dimensional Euclidean spaces) to analyze the nature of the corresponding percolation transition in spatial point processes. This problem finds a natural description in terms of the canonical ensemble but not in the usual grand-canonical one, customarily employed to describe percolation transitions. This leads us to analyze the question of ensemble equivalence and study whether the resulting canonical continuum percolation transition shares its universal properties with standard percolation transitions, analyzing diverse homogeneous and heterogeneous spatial point processes. We, therefore, provide a powerful tool to characterize and classify a vast class of natural point patterns, revealing their fundamental properties based on percolation phase transitions.

Derived patterns of musculoskeletal symptoms and their relationships with ergonomic factors among electronic assembly workers: A latent class analysis

Introduction: Multi-site musculoskeletal symptoms (MSS) are considered to be more common and have more serious consequences than single-site MSS. This study aimed to determine whether derived patterns of MSS may be identified in electronic assembly workers and if extracted MSS classes are associated with personal and work-related factors. Method: A cross-sectional questionnaire study was performed with 700 participating electronic assembly workers. The questionnaire included individual factors, psychosocial and physical exposures, and MSS. The derived patterns of MSS and their relationships with ergonomic factors were analyzed using latent class analysis (LCA) and multinomial logistic regression models (MLRM). Results: The 1-year prevalence of MSS affecting only one body site or two or more body sites was 14.9% and 32.7%, respectively. The results of LCA showed three distinct classes of MSS patterns, which were labelled ‘MSS in most sites’ (5.0%), ‘MSS in neck and shoulder’ (27.0%), and ‘MSS in one or no site’ (68.0%). The results of MLRM showed that the ‘MSS in neck and shoulder’ was associated with job tenure (OR 5.579, 95% CI 2.488–12.511), excessive dynamic and static loads (OR 3.868, 95% CI 1.702–8.793 and OR 5.270, 95% CI 2.020–13.747, respectively); while the ‘MSS in most sites’ was associated with high job demands (OR 4.528, 95% CI 1.647–12.445) and excessive dynamic loads (OR 111.554, 95% CI 4.996–2490.793). Conclusions: The results showed unique patterns of MSS among electronic assembly workers that were associated with personal and work-related factors. Practical applications: The findings highlight that the high prevalence of multi-site MSS in this group should be a focus. It also provides further evidence that LCA considering the number and location of anatomical sites involving MSS can be used to determine distinct classes of MSS patterns, which is of great significance for the epidemiological study and management of MSS in the future.

254-OR: Glucose Response Patterns Based on 75g OGTTS during Pregnancy and Their Association with the Risk of Macrosomia: A Latent Class Analysis of TWO Cohort Studies

Background and Aims: People with different glucose response patterns during the 2-h oral glucose tolerance test (OGTT) have substantially different clinical characteristics and risk for long-term outcomes in the general population. We aimed to identify comparable glucose response patterns in pregnancy and their association with maternal and fetal outcomes. Materials and Methods: We used latent class trajectory modelling to identify glucose response patterns using 5-point 75g OGTT in 470 pregnant women at 27.3±2.2 weeks of gestation. We assessed these classes and investigated pregnancy outcomes in an independent cohort of 7073 pregnant women with 3-point OGTTs. Results: We identified five different glucose response patterns (classes) . Rate of gestational diabetes (GDM) was lowest in class 1 (< 7%) and highest in class 5 (100%) for both cohorts. Class 3 was characterized by transient hyperinsulinemia at 30 minutes but the prevalence of GDM was only 25-36%. Compared to class 1, women in class 3 had an increased gestational weight gain (GWG, β=0.85 SE: 0.31kg, p=0.0065, adjusted for age, gestational age, pre-gestational BMI, GDM treatment and AUCGlucose) . New-borns in class 3 had the highest risk of macrosomia (OR 1.47 95% CI: [1.12, 1.93] vs. class 1) after adjustment for maternal age, parity and smoking, but this was attenuated by additional adjustment for pre-pregnancy BMI and GWG. Conclusion: We found an easily identifiable group of pregnant women who have an increased risk of macrosomia without formally meeting GDM diagnosis. These women would most likely benefit from a therapy (e.g. nutritional counselling) aiming at preventing excessive glucose excursions and GWG. Note: LF and AH contributed equally to this work. Disclosure L.Fritsche: None. R.Wagner: Advisory Panel; Akcea Therapeutics, Daiichi Sankyo, Sanofi-Aventis Deutschland GmbH, Speaker's Bureau; Lilly, Novo Nordisk, Sanofi-Aventis Deutschland GmbH. A.Hulman: None. K.Prystupa: None. M.Heni: Advisory Panel; Boehringer Ingelheim International GmbH, Research Support; Boehringer Ingelheim International GmbH, Sanofi, Speaker's Bureau; Amryt Pharma Plc, Boehringer Ingelheim International GmbH, Novo Nordisk. A.L.Birkenfeld: None. A.Peter: None. A.Fritsche: Advisory Panel; Boehringer Ingelheim International GmbH, Novo Nordisk, Sanofi-Aventis Deutschland GmbH. A.Kun: None. A.Tabak: Consultant; 77 Elektronika Kft., Boehringer Ingelheim International GmbH, Speaker's Bureau; 77 Elektronika Kft., AstraZeneca, Sanofi_aventis Zrt. Funding The PREG study is supported in part by a grant from the Federal Ministry of Education and Research (BMBF) (01GI0925) to the German Center for Diabetes Research (DZD)

Automatic Classification of Bloodstains with Deep Learning Methods

The classification of detected bloodstains into predetermined categories is a crucial component of the so-called bloodstain pattern analysis. As in other forensic disciplines, deep learning methods may help to reduce human subjectivity within this process, may increase the classification accuracy, shorten the calculation time and thus, enable high-throughput analysis. In this work, an approach is presented in which a convolutional neural network (Inception v3) was trained from 965 drip stains (passive origin) and 1595 blood spatters (active origin). The trained CNN was evaluated with a test data set consisting of 366 images of drip stains and blood spatters. The success rate was 99.73% which suggests that neural networks could also be used to automatically classify other classes of bloodstain patterns to speed up the investigation process in the future.

Existence of localized radial patterns in a model for dryland vegetation

Abstract Localized radial patterns have been observed in the vegetation of semi-arid ecosystems, often as localized patches of vegetation or in the form of ‘fairy circles’. We consider stationary localized radial solutions to a reduced model for dryland vegetation on flat terrain. By considering certain prototypical pattern-forming systems, we prove the existence of three classes of localized radial patterns bifurcating from a Turing instability. We also present evidence for the existence of localized gap solutions close to a homogeneous instability. Additionally, we numerically solve the vegetation model and use continuation methods to study the bifurcation structure and radial stability of localized radial spots and gaps. We conclude by investigating the effect of varying certain parameter values on the existence and stability of these localized radial patterns.

Time efficiency analysis for undersampled quantitative MRI acquisitions.

To realize Quantitative MRI (QMRI) with clinically acceptable scan time, acceleration factors achieved by conventional parallel imaging techniques are often inadequate. Further acceleration is possible using model-based reconstruction. We propose a theoretical metric called TEUSQA: Time Efficiency for UnderSampled QMRI Acquisitions to inform sequence design and sample pattern optimisation. TEUSQA is designed for a particular class of reconstruction techniques that directly estimate tissue parameters, possibly using prior information to regularize the estimation. TEUSQA can be used to evaluate undersampling patterns for multi-contrast QMRI sequences targeting any tissue parameter. To verify the time efficiency predicted by TEUSQA, we performed Monte Carlo simulations and an accelerated parameter mapping with two sequences (Inversion prepared fast spin echo for T1 and T2 mapping and 3D GRASE for T2 and B0 inhomogeneity mapping). Using TEUSQA, we assessed several ways to generate undersampling patterns in silico, providing insight into the relation between sample distribution and time efficiency for different acceleration factors. The time efficiency predicted by TEUSQA was within 15% of that observed in the Monte Carlo simulations and the prospective acquisition experiment. The assessment of undersampling patterns showed that a class of good patterns could be obtained by low-discrepancy sampling. We believe that TEUSQA offers a valuable instrument for developers of novel QMRI sequences pushing the boundaries of acceleration to achieve clinically feasible protocols. Finally, we applied a time-efficient undersampling pattern selected using TEUSQA for a 32-fold accelerated scan to map T1 & T2 mapping of a healthy volunteer.

Belief Combination of Classifiers for Incomplete Data

Data with missing values, or incomplete information, brings some challenges to the development of classification, as the incompleteness may significantly affect the performance of classifiers. In this paper, we handle missing values in both training and test sets with uncertainty and imprecision reasoning by proposing a new belief combination of classifier (BCC) method based on the evidence theory. The proposed BCC method aims to improve the classification performance of incomplete data by characterizing the uncertainty and imprecision brought by incompleteness. In BCC, different attributes are regarded as independent sources, and the collection of each attribute is considered as a subset. Then, multiple classifiers are trained with each subset independently and allow each observed attribute to provide a sub-classification result for the query pattern. Finally, these sub-classification results with different weights (discounting factors) are used to provide supplementary information to jointly determine the final classes of query patterns. The weights consist of two aspects: global and local. The global weight calculated by an optimization function is employed to represent the reliability of each classifier, and the local weight obtained by mining attribute distribution characteristics is used to quantify the importance of observed attributes to the pattern classification. Abundant comparative experiments including seven methods on twelve datasets are executed, demonstrating the out-performance of BCC over all baseline methods in terms of accuracy, precision, recall, F1 measure, with pertinent computational costs.

Pattern Inversion as a Pattern Recognition Method for Machine Learning

Artificial neural networks use a lot of coefficients that take a great deal of computing power for their adjustment, especially if deep learning networks are employed. However, there exist coefficients-free extremely fast indexing-based technologies that work, for instance, in Google search engines, in genome sequencing, etc. The paper discusses the use of indexing-based methods for pattern recognition. It is shown that for pattern recognition applications such indexing methods replace with inverse patterns the fully inverted files, which are typically employed in search engines. Not only such inversion provides automatic feature extraction, which is a distinguishing mark of deep learning, but, unlike deep learning, pattern inversion supports almost instantaneous learning, which is a consequence of absence of coefficients. The paper discusses a pattern inversion formalism that makes use on a novel pattern transform and its application for unsupervised instant learning. Examples demonstrate a view-angle independent recognition of three-dimensional objects, such as cars, against arbitrary background, prediction of remaining useful life of aircraft engines, and other applications. In conclusion, it is noted that, in neurophysiology, the function of the neocortical mini-column has been widely debated since 1957. This paper hypothesizes that, mathematically, the cortical mini-column can be described as an inverse pattern, which physically serves as a connection multiplier expanding associations of inputs with relevant pattern classes.

Patterns of Alcohol Use Among Italian Young Adults Before and During a COVID-19 Lockdown: A Latent Class Analysis Study

Two not mutually exclusive theories explain the effects of the COVID-19 pandemic on alcohol use. The Availability hypothesis contends that reduced opportunities to drink due to the closure of outlets and consumption sites should lead to decreases in alcohol use, whereas the Stress and Coping hypothesis argues that those exposed to stressful situations may increase drinking. The primary aim of this study was to examine changes—separately by gender—in the prevalence of drinking patterns among Italian young adults (18–34 years) before and during a COVID-19 lockdown. Study design was a repeated cross-sectional study, whereby data collected in 2015 and 2020 from nationally representative samples were analyzed. Latent class analysis identified five, fully invariant for women and partially invariant for men, drinking pattern classes among both cohorts: current non-drinkers (CND), weekend risky (WRD) and weekend non-risky drinkers (WnRD), daily non-risky (DnRD) and daily risky drinkers (DRD). In support of the Availability hypothesis, increases in abstaining and moderate drinking women and men were observed from 2015 to 2020. Concomitantly, among men only there were also increases in the prevalence of patterns characterized by risky drinking, coping drinking motives and related harm (Stress and Coping hypothesis). The pandemic and the three-tier lockdown imposed by the Italian government likely reduced overall alcohol use in the general population who drink moderately. However, there was a substantial increase in the prevalence of a small but significant group of men who drank daily and heavily to cope. Outreach and prevention efforts should target primarily this group, but also consider the opportunities that the exceptional circumstances of a quarantine offer to any individuals to reshape their lifestyle and health-related behaviors.

Clustering and graph mining techniques for classification of complex structural variations in cancer genomes

For many years, a major question in cancer genomics has been the identification of those variations that can have a functional role in cancer, and distinguish from the majority of genomic changes that have no functional consequences. This is particularly challenging when considering complex chromosomal rearrangements, often composed of multiple DNA breaks, resulting in difficulties in classifying and interpreting them functionally. Despite recent efforts towards classifying structural variants (SVs), more robust statistical frames are needed to better classify these variants and isolate those that derive from specific molecular mechanisms. We present a new statistical approach to analyze SVs patterns from 2392 tumor samples from the Pan-Cancer Analysis of Whole Genomes (PCAWG) Consortium and identify significant recurrence, which can inform relevant mechanisms involved in the biology of tumors. The method is based on recursive KDE clustering of 152,926 SVs, randomization methods, graph mining techniques and statistical measures. The proposed methodology was able not only to identify complex patterns across different cancer types but also to prove them as not random occurrences. Furthermore, a new class of pattern that was not previously described has been identified.

Relationship between demographic and cephalometric measures and electromyographic activityu of the facial musculature. A preliminary study in children and adolescents

Surface electromyography (EMGS) is used to evaluate the electromyographic activity of the orofacial musculature at rest, or when performing certain activities such as swallowing saliva or water, or tasks of maximum voluntary contraction. Numerous studies have evaluated the relationship between facial morphology and muscle activity, with contradictory results. The aim of this study was to determine whether there are significant differences in the electromyographic activity of the orofacial musculature as a function of the variables: age, height, weight, facial pattern and skeletal class. For this purpose, 81 patients with dental malocclusion were selected (35 males and 46 females, aged 6 to 17 years). Bilateral EMGS activity of the temporalis, masseter, orbicularis oris of the upper and lower lip of the mouth and suprahyoid muscles was measured in the resting position, maximum voluntary contraction and swallowing of saliva 2 ml of water. The results obtained show that the mandibular plane and the ANB angle present significant differences in all the variables of EMGS activity; however, the facial pattern only shows significant differences in the mean deviation of muscle activity. Skeletal class shows no significant effect on any of the EMGS variables recorded. The variables Age, Weight and Height show significant results on all EMGS activity variables. The variable Sex only shows significant results in the mean deviation of muscle activity. This work supports previous studies that found no significant results based on facial pattern and skeletal class. It would be advisable to perform a more exhaustive analysis of the muscles involved and their relationship between the different tasks evaluated, and to increase the sample size of patients with skeletal Class III in order to reach more definitive conclusions.

EXTENDED EUCLIDEAN ALGORITHM OF AUNU BINARY POLYNOMIALS OF CARDINALITY ELEVEN

Binary polynomials representation of Aunu permutation patterns has been used to perform arithmetic operations on words and sub-words of the polynomials using addition, multiplication, and division modulo two. The polynomials were also found to form some mathematical structures such as group, ring, and field. This paper presents the extension of our earlier work as it reports the Aunu binary polynomials of cardinality eleven and how to find their greatest common divisor (gcd) using the extended Euclidean algorithm. The polynomials are pairly permuted and the results found showed that one polynomial is a factor of the other polynomial or one polynomial is relatively prime to the other and some gave different results. This important feature is of combinatorial significance and can be investigated further to formulate some theoretic axioms for this class of Aunu permutation pattern. Binary polynomials have important applications in coding theory, circuit design, and the construction of cryptographic primitives.

Anchoring in truncation: A typological analysis

Analyses of truncation patterns (e.g. Rob, from Robert) have traditionally focused on the templatic shape of such forms, while less attention has been paid to the question of which parts of the base word may survive in the truncatum. This feature of truncation, usually referred to as ‘anchoring,’ is investigated in the present paper. On the empirical basis of the generalizations emerging from an extensive database of truncation patterns in the world’s languages, a formal typology of anchoring in the framework of Optimality Theory is constructed and its defining ranking conditions (its ‘typological properties’) are extracted. The typological properties reveal the grammatical forces shaping the various classes of truncation patterns. They show that anchoring constraints must indeed form an integral part of any model of morphological truncation since—in interaction with templatic size-restrictor constraints—they determine whether truncation occurs at all and whether output forms vary in size or are of a fixed templatic shape. A thorough analysis of anchoring thus also provides evidence for templatic shapes emerging from constraint interaction and against an approach in terms of fixed, language-specific templates.

Longitudinal Patterns of Grandchild Care in South Korea.

This study examined the longitudinal patterns of grandchild care to observe the influence of factors related to social participation, financial support to grandparents, demographic characteristics, and family structure on classifying the grandchild care. The rate of grandparent care for grandchildren was increasing, and the amount of time commitment for grandchild care was large in South Korea. Understanding how grandchild care unfolds over time and who is likely to provide ongoing grandchild care helps to advance the knowledge about grandparents providing grandchild care. The total sample consisted of 333 South Korean grandparents derived from the 3 waves of nationally representative data. This study utilized growth mixture modeling to identify latent classes of longitudinal patterns of grandchild care, and ran a multinomial logistic regression to examine the relationships between factors related to grandparents, adult children, and family structure and the identified latent classes. Grandchild care was classified into one of three categories: low-level decrease, high-level decrease, and low-to-high increase. Grandparents in the group of low-to-high increase were more likely to have higher financial dependence on adult children and have lower social participation than grandparents in other groups. Findings indicate that there are distinct subgroups among grandparents who care for their grandchildren. Additionally, those in the three classifications varied according to financial support received from adult children, social participation, and personal and family structure. Our findings inform policymakers to provide older adults a means to maintain their self-sufficiency. The community needs to provide programs and resources for working parents on behalf of grandchild care.

Multimorbidity Patterns and the Disablement Process among Public Long-Term Care Insurance Claimants in the City of Yiwu (Zhejiang Province, China).

This study aimed to identify multimorbidity patterns and explore the disablement process by utilizing the model raised by Verbrugge and Jette as a theoretical framework. This cross-sectional study used public Long-term Care Insurance (LTCI) claimants’ assessment data of Yiwu city in Zhejiang Province, China, for 2604 individuals aged 60 years and older, from September through December 2018. Latent Class Analysis (LCA) was conducted using 10 common chronic conditions. Structural Equation Modeling was used to examine the disablement process. The latent classes of multimorbidity patterns were the “coronary atherosclerotic heart disease” class (19.0%), the “lower limb fractures” class (26.4%), and the “other diseases” class (54.6%). The structural model results show that coronary atherosclerotic heart disease had a significant influence on incontinence, but it was not statistically significant in predicting vision impairment and mobility impairment. Lower limb fractures had significant effects on vision impairment, incontinence, and mobility impairment. Vision impairment, incontinence, and mobility impairment had significant effects on physical activities of daily living (ADLs). Our findings suggest that different impairments exist from specific patterns of multimorbidity to physical ADL disability, which may provide insights for researchers and policy makers to develop tailored care and provide support for physically disabled older people.

Assessment of glomerular morphological patterns by deep learning algorithms

BackgroundCompilation of different morphological lesion signatures is characteristic of renal pathology. Previous studies have documented the potential value of artificial intelligence (AI) in recognizing relatively clear-cut glomerular structures and patterns, such as segmental or global sclerosis or mesangial hypercellularity. This study aimed to test the capacity of deep learning algorithms to recognize complex glomerular structural changes that reflect common diagnostic dilemmas in nephropathology.MethodsFor this purpose, we defined nine classes of glomerular morphological patterns and trained twelve convolutional neuronal network (CNN) models on these. The two-step training process was done on a first dataset defined by an expert nephropathologist (12,253 images) and a second consensus dataset (11,142 images) defined by three experts in the field.ResultsThe efficacy of CNN training was evaluated using another set with 180 consensus images, showing convincingly good classification results (kappa-values 0.838–0.938).Furthermore, we elucidated the image areas decisive for CNN-based decision making by class activation maps. Finally, we demonstrated that the algorithm could decipher glomerular disease patterns coinciding in a single glomerulus (e.g. necrosis along with mesangial and endocapillary hypercellularity).ConclusionsIn summary, our model, focusing on glomerular lesions detectable by conventional microscopy, is the first sui generis to deploy deep learning as a reliable and promising tool in recognition of even discrete and/or overlapping morphological changes. Our results provide a stimulus for ongoing projects that integrate further input levels next to morphology (such as immunohistochemistry, electron microscopy, and clinical information) to develop a novel tool applicable for routine diagnostic nephropathology.

Symmetry-breaking mechanism for the formation of cluster chimera patterns.

The emergence of order in collective dynamics is a fascinating phenomenon that characterizes many natural systems consisting of coupled entities. Synchronization is such an example where individuals, usually represented by either linear or nonlinear oscillators, can spontaneously act coherently with each other when the interactions' configuration fulfills certain conditions. However, synchronization is not always perfect, and the coexistence of coherent and incoherent oscillators, broadly known in the literature as chimera states, is also possible. Although several attempts have been made to explain how chimera states are created, their emergence, stability, and robustness remain a long-debated question. We propose an approach that aims to establish a robust mechanism through which cluster synchronization and chimera patterns originate. We first introduce a stability-breaking method where clusters of synchronized oscillators can emerge. At variance with the standard approach where synchronization arises as a collective behavior of coupled oscillators, in our model, the system initially sets on a homogeneous fixed-point regime, and, only due to a global instability principle, collective oscillations emerge. Following a combination of the network modularity and the model's parameters, one or more clusters of oscillators become incoherent within yielding a particular class of patterns that we here name cluster chimera states.