Clustering of the self-organizing map

Identifying patterns in the multitrophic community and food-web structure of a low-turbidity temperate estuarine bay

Food web dynamics outline the ecosystem processes that regulate community structure. Challenges in the approaches used to capture topological descriptions of food webs arise due to the difficulties in collecting extensive empirical data with temporal and spatial variations in community structure and predator-prey interactions. Here, we use a Kohonen self-organizing map algorithm (as a measure of community pattern) and stable isotope-mixing models (as a measure of trophic interaction) to identify food web patterns across a low-turbidity water channel of a temperate estuarine-coastal continuum. We find a spatial difference in the patterns of community compositions between the estuarine and deep-bay channels and a seasonal difference in the plankton pattern but less in the macrobenthos and nekton communities. Dietary mixing models of co-occurring dominant taxa reveal site-specific but unchanging food web topologies and the prominent role of phytoplankton in the trophic base of pelagic and prevalent-detrital benthic pathways. Our approach provides realistic frameworks for linking key nodes from producers to predators in trophic networks.

An Intelligent System-on-a-Chip for a Real-Time Assessment of Fuel Consumption to Promote Eco-Driving

Pollution that originates from automobiles is a concern in the current world, not only because of global warming, but also due to the harmful effects on people’s health and lives. Despite regulations on exhaust gas emissions being applied, minimizing unsuitable driving habits that cause elevated fuel consumption and emissions would achieve further reductions. For that reason, this work proposes a self-organized map (SOM)-based intelligent system in order to provide drivers with eco-driving-intended driving style (DS) recommendations. The development of the DS advisor uses driving data from the Uyanik instrumented car. The system classifies drivers regarding the underlying causes of non-optimal DSs from the eco-driving viewpoint. When compared with other solutions, the main advantage of this approach is the personalization of the recommendations that are provided to motorists, comprising the handling of the pedals and the gearbox, with potential improvements in both fuel consumption and emissions ranging from the 9.5% to the 31.5%, or even higher for drivers that are strongly engaged with the system. It was successfully implemented using a field-programmable gate array (FPGA) device of the Xilinx ZynQ programmable system-on-a-chip (PSoC) family. This SOM-based system allows for real-time implementation, state-of-the-art timing performances, and low power consumption, which are suitable for developing advanced driving assistance systems (ADASs).

Pollution zone identification research during ozone pollution processes

Identifying an ozone pollution zone during the pollution processes is significant for ozone pollution management and environmental health risk assessment. However, few studies have focused on ozone pollution zone identification during pollution processes. A spatial-temporal clustering framework for identifying pollution zones during ozone pollution processes was initially proposed in this study, and an ozone pollution process in China in May 2017 was selected as a case. The results showed that the framework can help selecting one more accurate method to identify the pollution zone according to the pollution characteristics of air pollution process. In addition, different ozone pollution zone identification methods work well in different scenarios: The self-organizing map (SOM) method was suitable for identifying the zone with the duration of pollution between 24 and 48 h, the image fusion based on wavelet transform (IFbWT) method for the zone with the duration of pollution over 48 h and the Apriori method for the zone with obvious boundaries between high-value and low-value ozone concentrations. The proposed procedure can also be applied to identify the pollution zone of the pollution process of other pollutants.

Loss of G protein pathway suppressor 2 in human adipocytes triggers lipid remodeling by upregulating ATP binding cassette subfamily G member 1

OBJECTIVE

Adipogenesis is critical for adipose tissue remodeling during the development of obesity. While the role of transcription factors in the orchestration of adipogenic pathways is already established, the involvement of coregulators that transduce regulatory signals into epigenome alterations and transcriptional responses remains poorly understood. The aim of our study was to investigate which pathways are controlled by G protein pathway suppressor 2 (GPS2) during the differentiation of human adipocytes.

METHODS

We generated a unique loss-of-function model by RNAi depletion of GPS2 in human multipotent adipose-derived stem (hMADS) cells. We thoroughly characterized the coregulator depletion-dependent pathway alterations during adipocyte differentiation at the level of transcriptome (RNA-seq), epigenome (ChIP-seq H3K27ac), cistrome (ChIP-seq GPS2), and lipidome. We validated the in vivo relevance of the identified pathways in non-diabetic and diabetic obese patients.

RESULTS

The loss of GPS2 triggers the reprogramming of cellular processes related to adipocyte differentiation by increasing the responses to the adipogenic cocktail. In particular, GPS2 depletion increases the expression of BMP4, an important trigger for the commitment of fibroblast-like progenitors toward the adipogenic lineage and increases the expression of inflammatory and metabolic genes. GPS2-depleted human adipocytes are characterized by hypertrophy, triglyceride and phospholipid accumulation, and sphingomyelin depletion. These changes are likely a consequence of the increased expression of ATP-binding cassette subfamily G member 1 (ABCG1) that mediates sphingomyelin efflux from adipocytes and modulates lipoprotein lipase (LPL) activity. We identify ABCG1 as a direct transcriptional target, as GPS2 depletion leads to coordinated changes of transcription and H3K27 acetylation at promoters and enhancers that are occupied by GPS2 in wild-type adipocytes. We find that in omental adipose tissue of obese humans, GPS2 levels correlate with ABCG1 levels, type 2 diabetic status, and lipid metabolic status, supporting the in vivo relevance of the hMADS cell-derived in vitro data.

CONCLUSION

Our study reveals a dual regulatory role of GPS2 in epigenetically modulating the chromatin landscape and gene expression during human adipocyte differentiation and identifies a hitherto unknown GPS2-ABCG1 pathway potentially linked to adipocyte hypertrophy in humans.

Personalized mental stress detection with self-organizing map: From laboratory to the field

Stress has become a major health concern and there is a need to study and develop new digital means for real-time stress detection. Currently, the majority of stress detection research is using population based approaches that lack the capability to adapt to individual differences. They also use supervised learning methods, requiring extensive labeling of training data, and they are typically tested on data collected in a laboratory and thus do not generalize to field conditions. To address these issues, we present multiple personalized models based on an unsupervised algorithm, the Self-Organizing Map (SOM), and we propose an algorithmic pipeline to apply the method for both laboratory and field data. The performance is evaluated on a dataset of physiological measurements from a laboratory test and on a field dataset consisting of four weeks of physiological and smartphone usage data. In these tests, the performance on the field data was steady across the different personalization levels (accuracy around 60%) and a fully personalized model performed the best on the laboratory data, achieving accuracy of 92% which is comparable to state-of-the-art supervised classifiers. These results demonstrate the feasibility of SOM in personalized mental stress detection both in constrained and free-living environment.

Metabolic Footprint, towards Understanding Type 2 Diabetes beyond Glycemia

Type 2 diabetes (T2D) heterogeneity is a major determinant of complications risk and treatment response. Using cluster analysis, we aimed to stratify glycemia within metabolic multidimensionality and extract pathophysiological insights out of metabolic profiling. We performed a cluster analysis to stratify 974 subjects (PREVADIAB2 cohort) with normoglycemia, prediabetes, or non-treated diabetes. The algorithm was informed by age, anthropometry, and metabolic milieu (glucose, insulin, C-peptide, and free fatty acid (FFA) levels during the oral glucose tolerance test OGTT). For cluster profiling, we additionally used indexes of metabolism mechanisms (e.g., tissue-specific insulin resistance, insulin clearance, and insulin secretion), non-alcoholic fatty liver disease (NAFLD), and glomerular filtration rate (GFR). We found prominent heterogeneity within two optimal clusters, mainly representing normometabolism (Cluster-I) or insulin resistance and NAFLD (Cluster-II), at higher granularity. This was illustrated by sub-clusters showing similar NAFLD prevalence but differentiated by glycemia, FFA, and GFR (Cluster-II). Sub-clusters with similar glycemia and FFA showed dissimilar insulin clearance and secretion (Cluster-I). This work reveals that T2D heterogeneity can be captured by a thorough metabolic milieu and mechanisms profiling—metabolic footprint. It is expected that deeper phenotyping and increased pathophysiology knowledge will allow to identify subject’s multidimensional profile, predict their progression, and treat them towards precision medicine.

Interferons and viruses induce a novel primate-specific isoform dACE2 and not the SARS-CoV-2 receptor ACE2

Severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2), which causes COVID-19, utilizes angiotensin-converting enzyme 2 (ACE2) for entry into target cells. ACE2 has been proposed as an interferon-stimulated gene (ISG). Thus, interferon-induced variability in ACE2 expression levels could be important for susceptibility to COVID-19 or its outcomes. Here, we report the discovery of a novel, primate-specific isoform of ACE2, which we designate as deltaACE2 (dACE2). We demonstrate that dACE2, but not ACE2, is an ISG. In vitro, dACE2, which lacks 356 N-terminal amino acids, was non-functional in binding the SARS-CoV-2 spike protein and as a carboxypeptidase. Our results reconcile current knowledge on ACE2 expression and suggest that the ISG-type induction of dACE2 in IFN-high conditions created by treatments, inflammatory tumor microenvironment, or viral co-infections is unlikely to affect the cellular entry of SARS-CoV-2 and promote infection.

AHP versus DEA: a comparative analysis for the gradual improvement of unsustainable suppliers

Purpose

Data envelopment analysis (DEA) and analytical hierarchy process (AHP) are two widely applied methods to evaluate and rank suppliers in terms of sustainability. In this study, to investigate the extent to which potential differences in the outcomes of these two methods influence the benchmarking strategies, a comparative analysis based on a common set of data gathered from 19 logistics service providers is implemented.

Design/methodology/approach

As suppliers' sustainability cannot be improved in a single-step process due to several limitations, improvement needs to proceed gradually. Therefore, using the self-organising map method, the suppliers were classified into clusters within a novel framework for gradually improving their sustainability. Then, the two processes of gradual improvement based on the outcomes of DEA and AHP were compared.

Findings

The findings show that although the rankings of suppliers guided by the methods correlated to a high degree, the benchmarking strategies provided by the methods for gradually improving the sustainability of suppliers differed considerably. In particular, whereas AHP suggests a benchmarking policy better suited for unsustainable or less sustainable suppliers with limited access to resources, DEA proposes one for suppliers able to dramatically boost their sustainability with few quick, significant leaps in performance.

Originality/value

First, this study revealed a novel gradual improvement framework using the self-organising map method. Second, it clarified the extent to which the benchmarking policies are influenced by the type of evaluation method.

Spatial-Temporal Signals and Clinical Indices in Electrocardiographic Imaging (II): Electrogram Clustering and T-wave Alternans

During the last years, attention and controversy have been present for the first commercially available equipment being used in Electrocardiographic Imaging (ECGI), a new cardiac diagnostic tool which opens up a new field of diagnostic possibilities. Previous knowledge and criteria of cardiologists using intracardiac Electrograms (EGM) should be revisited from the newly available spatial-temporal potentials, and digital signal processing should be readapted to this new data structure. Aiming to contribute to the usefulness of ECGI recordings in the current knowledge and methods of cardiac electrophysiology, we previously presented two results: First, spatial consistency can be observed even for very basic cardiac signal processing stages (such as baseline wander and low-pass filtering); second, useful bipolar EGMs can be obtained by a digital processing operator searching for the maximum amplitude and including a time delay. In addition, this work aims to demonstrate the functionality of ECGI for cardiac electrophysiology from a twofold view, namely, through the analysis of the EGM waveforms, and by studying the ventricular repolarization properties. The former is scrutinized in terms of the clustering properties of the unipolar an bipolar EGM waveforms, in control and myocardial infarction subjects, and the latter is analyzed using the properties of T-wave alternans (TWA) in control and in Long-QT syndrome (LQTS) example subjects. Clustered regions of the EGMs were spatially consistent and congruent with the presence of infarcted tissue in unipolar EGMs, and bipolar EGMs with adequate signal processing operators hold this consistency and yielded a larger, yet moderate, number of spatial-temporal regions. TWA was not present in control compared with an LQTS subject in terms of the estimated alternans amplitude from the unipolar EGMs, however, higher spatial-temporal variation was present in LQTS torso and epicardium measurements, which was consistent through three different methods of alternans estimation. We conclude that spatial-temporal analysis of EGMs in ECGI will pave the way towards enhanced usefulness in the clinical practice, so that atomic signal processing approach should be conveniently revisited to be able to deal with the great amount of information that ECGI conveys for the clinician.

Classification and Prediction of Violence Against Chinese Medical Staff on the Sina Microblog Based on a Self-Organizing Map: Quantitative Study

Background

For the last decade, doctor-patient contradiction in China has remained prominent, and workplace violence toward medical staff still occurs frequently. However, little is known about the types and laws of propagation of violence against medical staff online.

Objective

By using a self-organizing map (SOM), we aimed to explore the microblog propagation law for violent incidents in China that involve medical staff, to classify the types of incidents and provide a basis for rapidly and accurately predicting trends in public opinion and developing corresponding measures to improve the relationship between doctors and patients.

Methods

For this study, we selected 60 cases of violent incidents in China involving medical staff that led to heated discussions on the Sina microblog from 2011 to 2018, searched the web data of the microblog using crawler software, recorded the amount of new tweets every 2 hours, and used the SOM neural network to cluster the number of tweets. Polynomial and exponential functions in MATLAB software were applied to predict and analyze the data.

Results

Trends in the propagation of online public opinion regarding the violent incidents were categorized into 8 types: bluff, waterfall, zigzag, steep, abrupt, wave, steep slope, and long slope. The communications exhibited different characteristics. The prediction effect of 4 types of incidents (ie, bluff, waterfall, zigzag, and steep slope) was good and accorded with actual spreading trends.

Conclusions

Our study found that the more serious the consequences of a violent incident, such as a serious injury or death, the more attention it drew on the microblog, the faster was its propagation speed, and the longer was its duration. In these cases, the propagation types were mostly steep slope, long slope, and zigzag. In addition, the more serious the consequences of a violent incident, the higher popularity it exhibited on the microblog. The popularity within a week was significantly higher for acts resulting from patients’ dissatisfaction with treatments than for acts resulting from nontherapeutic incidents.

Impact of the Partitioning Method on Multidimensional Adaptive-Chemistry Simulations

The large number of species included in the detailed kinetic mechanisms represents a serious challenge for numerical simulations of reactive flows, as it can lead to large CPU times, even for relatively simple systems. One possible solution to mitigate the computational cost of detailed numerical simulations, without sacrificing their accuracy, is to adopt a Sample-Partitioning Adaptive Reduced Chemistry (SPARC) approach. The first step of the aforementioned approach is the thermochemical space partitioning for the generation of locally reduced mechanisms, but this task is often challenging because of the high-dimensionality, as well as the high non-linearity associated to reacting systems. Moreover, the importance of this step in the overall approach is not negligible, as it has effects on the mechanisms’ level of chemical reduction and, consequently, on the accuracy and the computational speed-up of the adaptive simulation. In this work, two different clustering algorithms for the partitioning of the thermochemical space were evaluated by means of an adaptive CFD simulation of a 2D unsteady laminar flame of a nitrogen-diluted methane stream in air. The first one is a hybrid approach based on the coupling between the Self-Organizing Maps with K-Means (SKM), and the second one is the Local Principal Component Analysis (LPCA). Comparable results in terms of mechanism reduction (i.e., the mean number of species in the reduced mechanisms) and simulation accuracy were obtained for both the tested methods, but LPCA showed superior performances in terms of reduced mechanisms uniformity and speed-up of the adaptive simulation. Moreover, the local algorithm showed a lower sensitivity to the training dataset size in terms of the required CPU-time for convergence, thus also being optimal, with respect to SKM, for massive dataset clustering tasks.

Artificial Intelligence-Empowered Mobilization of Assessments in COVID-19-like Pandemics: A Case Study for Early Flattening of the Curve

The global outbreak of the Coronavirus Disease 2019 (COVID-19) pandemic has uncovered the fragility of healthcare and public health preparedness and planning against epidemics/pandemics. In addition to the medical practice for treatment and immunization, it is vital to have a thorough understanding of community spread phenomena as related research reports 17.9–30.8% confirmed cases to remain asymptomatic. Therefore, an effective assessment strategy is vital to maximize tested population in a short amount of time. This article proposes an Artificial Intelligence (AI)-driven mobilization strategy for mobile assessment agents for epidemics/pandemics. To this end, a self-organizing feature map (SOFM) is trained by using data acquired from past mobile crowdsensing (MCS) campaigns to model mobility patterns of individuals in multiple districts of a city so to maximize the assessed population with minimum agents in the shortest possible time. Through simulation results for a real street map on a mobile crowdsensing simulator and considering the worst case analysis, it is shown that on the 15th day following the first confirmed case in the city under the risk of community spread, AI-enabled mobilization of assessment centers can reduce the unassessed population size down to one fourth of the unassessed population under the case when assessment agents are randomly deployed over the entire city.

Development of Operational Strategies of Energy Storage System Using Classification of Customer Load Profiles under Time-of-Use Tariffs in South Korea

This study proposes a methodology to develop adaptive operational strategies of customer-installed Energy Storage Systems (ESS) based on the classification of customer load profiles. In addition, this study proposes a methodology to characterize and classify customer load profiles based on newly proposed Time-of-Use (TOU) indices. The TOU indices effectively distribute daily customer load profiles on multi-dimensional domains, indicating customer energy consumption patterns under the TOU tariff. The K-means and Self-Organizing Map (SOM) sophisticated clustering methods were applied for classification. Furthermore, this study demonstrates peak shaving and arbitrage operations of ESS with current supporting polices in South Korea. Actual load profiles accumulated from customers under the TOU rate were used to validate the proposed methodologies. The simulation results show that the TOU index-based clustering effectively classifies load patterns into ‘M-shaped’ and ‘square wave-shaped’ load patterns. In addition, the feasibility analysis results suggest different ESS operational strategies for different load patterns: the ‘M-shaped’ pattern fixes a 2-cycle operation per day due to battery life, while the ‘square wave-shaped’ pattern maximizes its operational cycle (a 3-cycle operation during the winter) for the highest profits.

Pattern Recognition and Anomaly Detection by Self-Organizing Maps in a Multi Month E-nose Survey at an Industrial Site

Currently people are aware of the risk related to pollution exposure. Thus odor annoyances are considered a warning about the possible presence of toxic volatile compounds. Malodor often generates immediate alarm among citizens, and electronic noses are convenient instruments to detect mixture of odorant compounds with high monitoring frequency. In this paper we present a study on pattern recognition on ambient air composition in proximity of a gas and oil pretreatment plant by elaboration of data from an electronic nose implementing 10 metal-oxide-semiconductor (MOS) sensors and positioned outdoor continuously during three months. A total of 80,017 e-nose vectors have been elaborated applying the self-organizing map (SOM) algorithm and then k-means clustering on SOM outputs on the whole data set evidencing an anomalous data cluster. Retaining data characterized by dynamic responses of the multisensory system, a SOM with 264 recurrent sensor responses to air mixture sampled at the site and four main air type profiles (clusters) have been identified. One of this sensor profiles has been related to the odor fugitive emissions of the plant, by using ancillary data from a total volatile organic compound (VOC) detector and wind speed and direction data. The overall and daily cluster frequencies have been evaluated, allowing us to identify the daily duration of presence at the monitoring site of air related to industrial emissions. The refined model allowed us to confirm the anomaly detection of the sensor responses.

Variability of Kuroshio Surface Axis Northeast of Taiwan Island Derived from Satellite Altimeter Data

The spatial and temporal variability of the Kuroshio surface axis northeast of Taiwan Island is investigated using 24 years of surface geostrophic currents derived from satellite altimeter data from 1993 to 2016. The Kuroshio surface axis is derived by an extraction method with three selected parameters, including the length of the subsidiary line, the intervals between two adjacent points, and the distance between the two adjacent subsidiary lines. The empirical mode decomposition analysis on the 24-year Kuroshio axes reveals that the mean periods of intra-seasonal and inter-annual variability, which are the two dominant components, are about 3.2 months and 1.3 years, respectively. The self-organizing map analysis reveals that the variation of Kuroshio axis northeast of Taiwan Island has four best matching unit (BMU) patterns: straight-path (BMUS), meandering-path (BMUM) and two transition stages (BMUT1 and BMUT2). The straight-path pattern shows strong seasonality: more likely occurring in summer. The meandering-path pattern is less frequent than straight-path pattern. During a typical period from November 26, 2012 to January 27, 2013, which is chosen as an independent example, the analysis on the satellite altimeter and sea surface temperature data shows that the patterns of the Kuroshio axis change successively in order of BMUT1→BMUM→BMUT2→BMUS, i.e., the Kuroshio axis migrates from the meandering-path to the straight-path pattern. During the typical period the warm water intrusion and a mesoscale eddy occur at the second stage corresponding to BMUM and migrate northwestward gradually at the last two stages corresponding to BMUT2 and BMUS. The transient order appears only during this typical period but it is not common for the whole study period. The monthly mean relatively vorticity is calculated and analyzed to evaluate the impact of the eddies on the Kuroshio surface axis variability, the results show that the anticyclonic (cyclonic) eddies can promote the Kuroshio surface axis to present the meandering-path (straight-path) pattern because of the potential vorticity conservation. The impacts of the anticyclonic eddies and the cyclonic eddies on the variability of the Kuroshio surface axis are opposite. The long-term day-to-day detection contributes to improving understanding the variability of Kuroshio surface axis northeast of Taiwan Island.

Validation of two hybrid approaches for clustering age-related groups based on gait kinematics data

Age-associated changes in walking parameters are relevant to recognize functional capacity and physical performance. However, the sensible nuances of slightly different gait patterns are hardly noticeable by inexperienced observers. Due to the complexity of this evaluation, we aimed at verifying the efficiency of applied hybrid-adaptive algorithms to cluster groups with similar gait patterns. Based on self-organizing maps (SOM), k-means clustering (KM), and fuzzy c-means (FCM), we compared the hybrid algorithms to a conventional FCM approach to cluster accordingly age-related groups. Additionally, we performed a relevance analysis to identify the principal gait characteristics. Our experiments, based on inertial-sensors data, comprised a sample of 180 healthy subjects, divided into age-related groups. The outcomes suggest that our methods outperformed the FCM algorithm, demonstrating a high accuracy (88%) and consistent sensitivity also to distinguish groups that presented a significant difference (p < .05) only in one of the six observed gait features. The applied algorithms showed a compatible performance, but the SOM + KM required less computation cost and, therefore, was more efficient. Furthermore, the results indicate the overall importance of cadence, as a measurement of physical performance, especially when clustering subjects by their age. Such output provides valuable information to healthcare professionals, concerning the subject's physical performance related to his age, supporting and guiding the physical evaluation.

Application of a Self-Organizing Map of Isotopic and Chemical Data for the Identification of Groundwater Recharge Sources in Nasunogahara Alluvial Fan, Japan

Paddy rice fields on an alluvial fan not only use groundwater for irrigation but also play an important role as groundwater recharge sources. In this study, we investigated the spatial distribution of isotopic and hydrochemical compositions of groundwater in the Nasunogahara alluvial fan in Japan and applied a self-organizing map (SOM) to characterize the groundwater. The SOM assisted with the hydrochemical and isotopic interpretation of the groundwater in the fan, and clearly classified the groundwater into four groups reflecting the different origins. Two groundwater groups with lower isotopic ratios of water than the mean precipitation values in the fan were influenced by the infiltration of river water flowing from higher areas in the catchments and were differentiated from each other by their Na+ and Cl− concentrations. A groundwater group with higher isotopic ratios was influenced by the infiltration of paddy irrigation water that had experienced evaporative isotopic enrichment. Groundwater in the fourth group, which was distributed in the upstream area of the fan where dairy farms dominated, showed little influence of recharge waters from paddy rice fields. The findings of this study will contribute to proper management of the groundwater resources in the fan.

A Weather-Pattern-Based Evaluation of the Performance of CMIP5 Models over Mexico

The fifth phase of the Coupled Model Inter-Comparison Project (CMIP5) is frequently used to force regional climate models for dynamic downscaling and projections, which decision makers in turn use for future plans in different sectors. It is, therefore, highly important to assess their performance in order to use them as reliable tools. A weather-type approach for the evaluation of the performance of CMIP5 models is employed in this study, with the objective of providing insight into model errors under a set of distinct synoptic conditions and circulation types associated with the rainy season over Mexico and Central America. The Self-Organizing Maps algorithm is used to identify the main weather regimes (constructed from sea level pressure, specific humidity, and low-level winds at a daily time-scale), which are then evaluated against reanalysis. The results show that model performance depends on the weather type in all of the variables except for sea level pressure, which confirms the usefulness of this approach. The models simulate better the humidity patterns that show weak deviations from the climatological norm. In addition, the wind pattern representing the Caribbean Low Level Jet is well reproduced by all the models. The results show the capacity of this methodology for determining the extent to which climate models represent the main circulation patterns that characterize the climate and local weather in Mexico.

Supervised and Semi-Supervised Self-Organizing Maps for Regression and Classification Focusing on Hyperspectral Data

Machine learning approaches are valuable methods in hyperspectral remote sensing, especially for the classification of land cover or for the regression of physical parameters. While the recording of hyperspectral data has become affordable with innovative technologies, the acquisition of reference data (ground truth) has remained expensive and time-consuming. There is a need for methodological approaches that can handle datasets with significantly more hyperspectral input data than reference data. We introduce the Supervised Self-organizing Maps (SuSi) framework, which can perform unsupervised, supervised and semi-supervised classification as well as regression on high-dimensional data. The methodology of the SuSi framework is presented and compared to other frameworks. Its different parts are evaluated on two hyperspectral datasets. The results of the evaluations can be summarized in four major findings: (1) The supervised and semi-Supervised Self-organizing Maps (SOM) outperform random forest in the regression of soil moisture. (2) In the classification of land cover, the supervised and semi-supervised SOM reveal great potential. (3) The unsupervised SOM is a valuable tool to understand the data. (4) The SuSi framework is versatile, flexible, and easy to use. The SuSi framework is provided as an open-source Python package on GitHub.

A Survey to Discover Current Food Choice Behaviors

Food choices are complex functions of several elements that could change over time. Nowadays consumers appear careful about sustainable food consumption: the behavior of “food citizenship”, as the practice to support a sustainable food system during the consumption actions, arises. This study aims to recognize the existence of food choice behaviors in the contemporary scenario and to investigate the relation between the food choice factors and the behaviors recognized. Following a quantitative research method, a sample of 380 participants, recruited from a traditional Italian food and wine event, completed a questionnaire in order to detect their attitude about food. Four current food choice behaviors were recognized: The Individualist, The Foodie, The Environmentalist and The Health enthusiast. The relation between food choice factors and food choice behaviors was explained. Several stakeholders could benefit from the study results, in order to better understand how to adapt products and marketing strategies to satisfy the emerging customer’s needs and awareness. Even if a person can identify themselves within a single food choice behavior, they become aware of other choice models expanding their personal point of view. Finally, new research scenarios arose for the researchers.

A Text Classification Algorithm Based on Improved Multidimensional–Multiresolution Topological Pattern Recognition

Traditional pattern recognition is based on “optimal partition” and the goal is to find an optimal classification interface based on the distribution of each category in high-dimensional space, thus has its inherent shortcomings and deficiencies. While topology pattern recognition can effectively compensate for the shortcomings of traditional pattern recognition, topological pattern recognition is based on “cognition” and the goal is to find the appropriate cover according to the “complex set cover” in high-dimensional space to achieve cognitive effect. Topological pattern recognition can effectively consummate the characteristics of high error rate, low recognition rate and repetitive training in the existing recognition system with low training sample number. At present, topology pattern recognition has been applied in many areas of social life. However, one problem that can’t be ignored is that topological pattern recognition requires a long training time and low fault tolerance rate. Therefore, this paper proposes an improved multidimensional–multiresolution topological pattern recognition, and applies it to text classification and recognition. The results show that the improved multidimensional–multiresolution topological pattern recognition method can effectively reduce the training time of text classification and improve the classification efficiency.

A Review of Computational Methods for Clustering Genes with Similar Biological Functions

Clustering techniques can group genes based on similarity in biological functions. However, the drawback of using clustering techniques is the inability to identify an optimal number of potential clusters beforehand. Several existing optimization techniques can address the issue. Besides, clustering validation can predict the possible number of potential clusters and hence increase the chances of identifying biologically informative genes. This paper reviews and provides examples of existing methods for clustering genes, optimization of the objective function, and clustering validation. Clustering techniques can be categorized into partitioning, hierarchical, grid-based, and density-based techniques. We also highlight the advantages and the disadvantages of each category. To optimize the objective function, here we introduce the swarm intelligence technique and compare the performances of other methods. Moreover, we discuss the differences of measurements between internal and external criteria to validate a cluster quality. We also investigate the performance of several clustering techniques by applying them on a leukemia dataset. The results show that grid-based clustering techniques provide better classification accuracy; however, partitioning clustering techniques are superior in identifying prognostic markers of leukemia. Therefore, this review suggests combining clustering techniques such as CLIQUE and k-means to yield high-quality gene clusters.

Managing taste and odour metabolite production in drinking water reservoirs: The importance of ammonium as a key nutrient trigger

Taste and odour (T&O) compounds (most commonly 2-MIB and Geosmin) in drinking water are becoming an increasingly global problem for water management. Here, the trigger(s) for 2-MIB and Geosmin production were investigated in Plas Uchaf reservoir (North Wales, UK) with detailed water sample analysis between 2015 and 2016. Historical abstraction data from this reservoir and 4 reservoirs in Somerset (England, UK) were compared statistically using Self-Organising Map (SOM) analysis. In-reservoir measurements (2015-2016) revealed an 85% reduction in ammonium from the primary external loading source led to lower 2-MIB and Geosmin concentrations, with peak concentrations of 2-MIB declining from 60 to 21  ng l^-1 and Geosmin declining from 140 to 18  ng l^-1. No other measured water chemistry parameter showed a significant difference between years. The SOM results support the in-reservoir findings, revealing 2-MIB and Geosmin to be associated with high ammonium relative to nitrate for all 5 reservoirs. We conclude that ammonium is key for stimulating cyanobacterial productivity and production of T&O compounds. Whilst it is well understood that adequate availability of phosphorus is required for rapid growth in cyanobacteria, and hence should still be considered in management decisions, we suggest that monitoring sources and concentrations of ammonium is key for managing T&O outbreaks in drinking water reservoirs.

Comparison of Water Sampling between Environmental DNA Metabarcoding and Conventional Microscopic Identification: A Case Study in Gwangyang Bay, South Korea

Our study focuses on methodological comparison of plankton community composition in relation to ecological monitoring and assessment with data sampling. Recently, along with the advancement of monitoring techniques, metabarcoding has been widely used in the context of environmental DNA (eDNA). We examine the applicability of eDNA metabarcoding for effective monitoring and assessment of community composition, compared with conventional observation using microscopic identification in a coastal ecosystem, Gwangynag Bay in South Korea. Our analysis is based primarily on two surveys at a total of 15 study sites in early and late summer (June and September) of the year 2018. The results of our study demonstrate the similarity and dissimilarity of biological communities in composition, richness and diversity between eDNA metabarcoding and conventional microscopic identification. It is found that, overall, eDNA metabarcoding appears to provide a wider variety of species composition, while conventional microscopic identification depicts more distinct plankton communities in sites. Finally, we suggest that eDNA metabarcoding is a practically useful method and can be potentially considered as a valuable alternative for biological monitoring and diversity assessments.

Identification of clinically relevant dysglycemia phenotypes based on continuous glucose monitoring data from youth with type 1 diabetes and elevated hemoglobin A1c

BACKGROUND/OBJECTIVE

To identify and characterize subgroups of adolescents with type 1 diabetes (T1D) and elevated hemoglobin A1c (HbA1c) who share patterns in their continuous glucose monitoring (CGM) data as "dysglycemia phenotypes."

METHODS

Data were analyzed from the Flexible Lifestyles Empowering Change randomized trial. Adolescents with T1D (13-16 years, duration >1 year) and HbA1c 8% to 13% (64-119 mmol/mol) wore blinded CGM at baseline for 7 days. Participants were clustered based on eight CGM metrics measuring hypoglycemia, hyperglycemia, and glycemic variability. Clusters were characterized by their baseline features and 18 months changes in HbA1c using adjusted mixed effects models. For comparison, participants were stratified by baseline HbA1c (≤/>9.0% [75 mmol/mol]).

RESULTS

The study sample included 234 adolescents (49.8% female, baseline age 14.8 ± 1.1 years, baseline T1D duration 6.4 ± 3.7 years, baseline HbA1c 9.6% ± 1.2%, [81 ± 13 mmol/mol]). Three Dysglycemia Clusters were identified with significant differences across all CGM metrics (P < .001). Dysglycemia Cluster 3 (n = 40, 17.1%) showed severe hypoglycemia and glycemic variability with moderate hyperglycemia and had a lower baseline HbA1c than Clusters 1 and 2 (P < .001). This cluster showed increases in HbA1c over 18 months (p-for-interaction = 0.006). No other baseline characteristics were associated with Dysglycemia Clusters. High HbA1c was associated with lower pump use, greater insulin doses, more frequent blood glucose monitoring, lower motivation, and lower adherence to diabetes self-management (all P < .05).

CONCLUSIONS

There are subgroups of adolescents with T1D for which glycemic control is challenged by different aspects of dysglycemia. Enhanced understanding of demographic, behavioral, and clinical characteristics that contribute to CGM-derived dysglycemia phenotypes may reveal strategies to improve treatment.

Urban Vitality Area Identification and Pattern Analysis from the Perspective of Time and Space Fusion

Urban vitality provides an important basis for evaluating urban development and spatial balance. In the era of big data, the quantitative analysis of urban vitality has become a research hotspot in the field of urban sustainability and planning research. However, time variation characteristics are often neglected, which leads to one-sidedness in the pattern analysis of urban vitality. In this paper, a method for extracting vitality areas and integrating spatiotemporal features clustering is proposed. The method is used to divide urban space into multiple vitality areas scientifically. The spatial and temporal distribution patterns of urban vitality areas are found, and the driving factors of various vitality patterns are analyzed by combining points of interest (POI)-based land use characteristics. To illustrate this method, this paper takes Nanjing city as an example. One week's worth of mobile phone data indicated that Nanjing has 10 and 8 vitality areas on weekdays and weekends, respectively. The spatial and temporal distribution patterns of the vitality areas and their correlation with land use were analyzed, which proved that POI density and entropy have strong correlations with urban vitality.

Combined classification and source apportionment analysis for trace elements in western Philippine Sea sediments

Trace elements have been widely used for classification (of variables and of samples) and source apportionment studies, but the comparison and combination of the two is uncommon in previous works. In this paper, the grouping of trace elements, clustering of samples, and source identification were merged for an integrated understanding of the origin and distribution of trace elements in western Philippine Sea sediments. The grouping and clustering studies were implemented by a nonlinear clustering method called a self-organizing map (SOM), and the source identification was accomplished by a nontraditional factor analysis method called positive matrix factorization (PMF). Through visualization and clustering techniques, the SOM simultaneously classified a database of 26 trace elements into four groups of trace elements and five clusters of samples. Each sample cluster occupies a certain geographic area and is characterized by high concentrations of trace elements that are classified within one or two groups. Five potential sources were identified by PMF, representing the land mass of Taiwan Island, anthropogenic emissions from Taiwan, nutrient exportation from the South China Sea, mineral attachment in the deep ocean, and biogenetic components and riverine inputs from the Luzon Islands. The spatial distributions of the sample clusters are comparable to the ranges of high contributions from the five sources distinguished by PMF. This conclusion was further supported by displaying the PMF outputs on the SOM plane. Furthermore, a corresponding relationship was observed between every factor profile and every trace element group. Our work tests the consistency of the classification (of the trace elements and of the samples) and source identification and improves the application of multiperspective methodology in environmental studies.

Systems Pharmacology for Investigation of the Mechanisms of Action of Traditional Chinese Medicine in Drug Discovery

As a traditional medical intervention in Asia and a complementary and alternative medicine in western countries, traditional Chinese medicine (TCM) has attracted global attention in the life science field. TCM provides extensive natural resources for medicinal compounds, and these resources are generally regarded as effective and safe for use in drug discovery. However, owing to the complexity of compounds and their related multiple targets of TCM, it remains difficult to dissect the mechanisms of action of herbal medicines at a holistic level. To solve the issue, in the review, we proposed a novel approach of systems pharmacology to identify the bioactive compounds, predict their related targets, and illustrate the molecular mechanisms of action of TCM. With a predominant focus on the mechanisms of actions of TCM, we also highlighted the application of the systems pharmacology approach for the prediction of drug combination and dynamic analysis, the synergistic effects of TCMs, formula dissection, and theory analysis. In summary, the systems pharmacology method contributes to understand the complex interactions among biological systems, drugs, and complex diseases from a network perspective. Consequently, systems pharmacology provides a novel approach to promote drug discovery in a precise manner and a systems level, thus facilitating the modernization of TCM.

Combining domain filling with a self-organizing map to analyze multi-species hydrocarbon signatures on a regional scale

For the period of the Barnett Coordinated Campaign, October 16-31, 2013, hourly concentrations for 46 volatile organic compounds (VOCs) were recorded at 14 air monitoring stations within the Barnett Shale of North Texas. These measurements are used to identify and analyze multi-species hydrocarbon signatures on a regional scale through the novel combination of two techniques: domain filling with Lagrangian trajectories and the machine learning unsupervised classification algorithm called a self-organizing map (SOM). This combination of techniques is shown to accurately identify concentration enhancements in the lightest measured alkane species at and downwind of the locations of active-permit oil and gas facilities, despite the model having no a priori knowledge of these source locations. Site comparisons further identify the SOM's ability to distinguish between signatures with differing influences from oil- and gas-related processes and from urban processes. A random forest (a machine learning supervised classification) analysis is conducted to further probe the sensitivities of the SOM classification in response to changes in any hydrocarbon species' concentration values. The random forest analysis of four representative classes finds that the SOM classification is appropriately more sensitive to changes in certain urban-related species for urban-related classes, and to changes in oil- and gas-related species for oil- and gas-related classes.

QSAR Classification-Based Virtual Screening Followed by Molecular Docking Identification of Potential COX-2 Inhibitors in a Natural Product Library

Developments of natural inhibitors to prevent the function of cyclooxygenase-2 (COX-2) protein, responsible for a variety of inflammations and cancers, are a major challenge in the scientific community. In this study, robust QSAR classification models for predicting COX-2 inhibitor were developed, by which the self-organizing feature map neural network and random forest (RF) were adopted to improve the prediction of classification model ability. The F-score-based criterion combined with RF was used for feature selection, and good performance for COX-2 inhibitor prediction in overall accuracy was demonstrated. We used this model as a virtual screening tool for identifying the potential COX-2 inhibitor from a natural product library and found potential hit compounds. This compound further screened by applying molecular docking simulation identified five potential hits such as osthole, kavain, vanillyl acetone, myristicin, and psoralen, having a comparable binding affinity to COX-2 protein. However, in cell experiment, three hit compounds revealed COX-2 inhibitory activity in mRNA and protein level such as osthole, kavain, and psoralen.

pH-Induced Misfolding Mechanism of Prion Protein: Insights from Microsecond-Accelerated Molecular Dynamics Simulations

The conformational transition of prion protein (PrP) from a native form PrP^C to a pathological isoform PrP^Sc is the main cause of a number of prion diseases in human and animals. Thus, understanding the molecular basis of conformational transition of PrP will be valuable for unveiling the etiology of PrP-related diseases. Here, to explore the potential misfolding mechanism of PrP under the acidic condition, which is known to promote PrP misfolding and trigger its aggregation, the conventional and accelerated molecular dynamics (MD) simulations combined with the Markov state model (MSM) analysis were performed. The conventional MD simulations reveal that, at an acidic pH, the globular domain of PrP is partially unfolded, particularly for the α2 C-terminus. Structural analysis of the key macrostates obtained by MSM indicates that the α2 C-terminus and the β2-α2 loop may serve as important sites for the pH-induced PrP misfolding. Meanwhile, the α1 may also participate in the pH-induced structural conversion by moving away from the α2-α3 subdomain. Notably, dynamical network analysis of the key metastable states indicates that the protonated H187 weakens the interactions between the α2 C-terminus, α1-β2 loop, and α2-α3 loop, leading these domains, especially the α2 C-terminus, to become unstable and to begin to misfold. Therefore, the α2 C-terminus plays a key role in the PrP misfolding process and serves as a potential site for drug targeting. Overall, our findings can deepen the understanding of the pathogenesis related to PrP and provide useful guidance for the future drug discovery.

The Impact of Global Warming on Wind Energy Resources and Ramp Events in Japan

This study investigated the impact of global warming on Japanese wind energy resources and their short-term variations using the large ensemble d4PDF dataset, which consists of dynamically downscaled historical and +4K future climate projections. The capacity factor under the future and present climate was estimated from an idealized power curve based on hourly near-surface wind speeds. The +4K warming future climate projections showed significant changes in wind energy resources that varied both regionally and seasonally. The wind energy potential was projected to slightly increase (decrease) from winter to spring over northern (southern) Japan and decrease from summer to autumn over most of Japan. The projected annual production decreased by about ~5% over Japan in response to climate change. The frequency of wind ramp events also decreased in the latter seasons. The relationship to synoptic weather was investigated using self-organizing maps, whereby weather patterns (WPs) over the region in the present and future +4K climate were classified for a two-dimensional lattice. Future probabilistic projections of WPs under the global warming scenario showed both increases and decreases in the frequency of different WPs, with corresponding advantages and disadvantages for wind power generation with regard to future changes in capacity factors in Japan. The importance of these frequency changes on the total change was further assessed by separating the dynamical and thermodynamic contributions.

Position Estimation Based on Grid Cells and Self-Growing Self-Organizing Map

As the basis of animals' natal homing behavior, path integration can continuously provide current position information relative to the initial position. Some neurons in freely moving animals' brains can encode current positions and surrounding environments by special firing patterns. Research studies show that neurons such as grid cells (GCs) in the hippocampus of animals' brains are related to the path integration. They might encode the coordinate of the animal's current position in the same way as the residue number system (RNS) which is based on the Chinese remainder theorem (CRT). Hence, in order to provide vehicles a bionic position estimation method, we propose a model to decode the GCs' encoding information based on the improved traditional self-organizing map (SOM), and this model makes full use of GCs' firing characteristics. The details of the model are discussed in this paper. Besides, the model is realized by computer simulation, and its performance is analyzed under different conditions. Simulation results indicate that the proposed position estimation model is effective and stable.

Artificial intelligence for understanding concussion: Retrospective cluster analysis on the balance and vestibular diagnostic data of concussion patients

OBJECTIVES

We propose a bottom-up, machine-learning approach, for the objective vestibular and balance diagnostic data of concussion patients, to provide insight into the differences in patients' phenotypes, independent of existing diagnoses (unsupervised learning).

METHODS

Diagnostic data from a battery of validated balance and vestibular assessments were extracted from the database of the Swiss Concussion Center. The desired number of clusters within the patient database was estimated using Calinski-Harabasz criteria. Complex (self-organizing map, SOM) and standard (k-means) clustering tools were used, and the formed clusters were compared.

RESULTS

A total of 96 patients (81.3% male, age (median [IQR]): 25.0[10.8]) who were expected to suffer from sports-related concussion or post-concussive syndrome (52[140] days between diagnostic testing and the concussive episode) were included. The cluster evaluation indicated dividing the data into two groups. Only the SOM gave a stable clustering outcome, dividing the patients in group-1 (n = 38) and group-2 (n = 58). A large significant difference was found for the caloric summary score for the maximal speed of the slow phase, where group-1 scored 30.7% lower than group-2 (27.6[18.2] vs. 51.0[31.0]). Group-1 also scored significantly lower on the sensory organisation test composite score (69.0[22.3] vs. 79.0[10.5]) and higher on the visual acuity (-0.03[0.33] vs. -0.14[0.12]) and dynamic visual acuity (0.38[0.84] vs. 0.20[0.20]) tests. The importance of caloric, SOT and DVA, was supported by the PCA outcomes. Group-1 tended to report headaches, blurred vision and balance problems more frequently than group-2 (>10% difference).

CONCLUSION

The SOM divided the data into one group with prominent vestibular disorders and another with no clear vestibular or balance problems, suggesting that artificial intelligence might help improve the diagnostic process.

Change Detection in Remote Sensing Images Based on Image Mapping and a Deep Capsule Network

Homogeneous image change detection research has been well developed, and many methods have been proposed. However, change detection between heterogeneous images is challenging since heterogeneous images are in different domains. Therefore, direct heterogeneous image comparison in the way that we do it is difficult. In this paper, a method for heterogeneous synthetic aperture radar (SAR) image and optical image change detection is proposed, which is based on a pixel-level mapping method and a capsule network with a deep structure. The mapping method proposed transforms an image from one feature space to another feature space. Then, the images can be compared directly in a similarly transformed space. In the mapping process, some image blocks in unchanged areas are selected, and these blocks are only a small part of the image. Then, the weighted parameters are acquired by calculating the Euclidean distances between the pixel to be transformed and the pixels in these blocks. The Euclidean distance calculated according to the weighted coordinates is taken as the pixel gray value in another feature space. The other image is transformed in a similar manner. In the transformed feature space, these images are compared, and the fusion of the two different images is achieved. The two experimental images are input to a capsule network, which has a deep structure. The image fusion result is taken as the training labels. The training samples are selected according to the ratio of the center pixel label and its neighboring pixels’ labels. The capsule network can improve the detection result and suppress noise. Experiments on remote sensing datasets show the final detection results, and the proposed method obtains a satisfactory performance.

Multi Similarity Metric Fusion in Graph-Based Semi-Supervised Learning

In semi-supervised label propagation (LP), the data manifold is approximated by a graph, which is considered as a similarity metric. Graph estimation is a crucial task, as it affects the further processes applied on the graph (e.g., LP, classification). As our knowledge of data is limited, a single approximation cannot easily find the appropriate graph, so in line with this, multiple graphs are constructed. Recently, multi-metric fusion techniques have been used to construct more accurate graphs which better represent the data manifold and, hence, improve the performance of LP. However, most of these algorithms disregard use of the information of label space in the LP process. In this article, we propose a new multi-metric graph-fusion method, based on the Flexible Manifold Embedding algorithm. Our proposed method represents a unified framework that merges two phases: graph fusion and LP. Based on one available view, different simple graphs were efficiently generated and used as input to our proposed fusion approach. Moreover, our method incorporated the label space information as a new form of graph, namely the Correlation Graph, with other similarity graphs. Furthermore, it updated the correlation graph to find a better representation of the data manifold. Our experimental results on four face datasets in face recognition demonstrated the superiority of the proposed method compared to other state-of-the-art algorithms.

Spatiotemporal Data Clustering: A Survey of Methods

Large quantities of spatiotemporal (ST) data can be easily collected from various domains such as transportation, social media analysis, crime analysis, and human mobility analysis. The development of ST data analysis methods can uncover potentially interesting and useful information. Due to the complexity of ST data and the diversity of objectives, a number of ST analysis methods exist, including but not limited to clustering, prediction, and change detection. As one of the most important methods, clustering has been widely used in many applications. It is a process of grouping data with similar spatial attributes, temporal attributes, or both, from which many significant events and regular phenomena can be discovered. In this paper, some representative ST clustering methods are reviewed, most of which are extended from spatial clustering. These methods are broadly divided into hypothesis testing-based methods and partitional clustering methods that have been applied differently in previous research. Research trends and the challenges of ST clustering are also discussed.

An Efficient MRI Brain Tumor Segmentation by the Fusion of Active Contour Model and Self-Organizing-Map

Accurate detection of tumors in brain MR images is very important for the medical image analysis and interpretation. Tumors which are detected and treated in the early stage gives better long-term survival than those detected lately. This paper proposes a combined method of Self-Organizing –Map (SOM) and Active Contour Model (ACM) for the effective segmentation of the brain tumor from MR images. ACMs are energy-based image segmentation methods and they treat the segmentation as an optimization problem. The optimization function is formulated in terms of appropriate parameters and is designed such that the minimum value of its correspondence to a contour which is a near approximation of the real object boundary. The traditional ACMs depend on pixel intensity as well as very susceptible to parameter tuning and it turns out to be a challenge for these ACMs to deal the image objects of distinct intensities. Conversely, Neural Networks (NNs) are very effective in dealing inhomogeneities but usually results in noise due to the misclassification of pixels. Additionally, NNs deal the segmentation problems without objective function. Hence we proposed a framework for the brain tumor segmentation which integrates SOM with ACM and is termed as SOMACM. This works by exactly integrating the global information derived from the weights or prototypes of the trained SOM neurons to aid choosing whether to shrink or enlarge the present contour during the optimization process and is performed in an iterative way. The proposed method can deal with the images of complex intensity distributions, even in the presence of noise. Exploratory outcomes demonstrate the high accuracy in the segmentation results of SOMACM on different tumor images, compared to the ACM as well as the general SOM segmentation methods. Furthermore, the proposed framework is not highly sensitive to parameter tuning.

Associations Across Caregiver and Care Recipient Symptoms: Self-Organizing Map and Meta-analysis

Purpose of the Study

The main objective of this study was to reveal generalizable associations across caregiver burden (CGB), caregiver depression (CGD), care recipient cognitive ability (CRCA), and care recipient behavioral and psychological symptoms of dementia (BPSD).

Design and Methods

Studies published between 2004 and 2014 and reporting CGB and/or CGD together with CRCA and/or BPSD were included. Only 95 out of 1,955 studies provided enough data for data clustering with the Self-Organizing Map (SOM) and 27 of them for meta-analyses based on correlation coefficients.

Results

Caregiver and care recipient symptoms were not tightly associated with each other, except for the CGB-BPSD interaction at the individual level. SOM emphasized the cluster comprising studies reporting low CGB, low CGD, high CRCA, and few BPSD. Meta-analyses indicated high heterogeneity between the original studies.

Implications

Relationships between caregiver and care recipient symptoms should be treated as situation-specific phenomena, at least when the symptoms are moderate at most. Dementia caregiving per se should not be understood as a source of stress and mental health problems. More systematic and coherent use of measures is necessary to enable a comprehensive analysis of caregiving.