Clustering of the self-organizing map

Contextual contact tracing based on stochastic compartment modeling and spatial risk assessment

The current situation of COVID-19 highlights the paramount importance of infectious disease surveillance, which necessitates early monitoring for effective response. Policymakers are interested in data insights identifying high-risk areas as well as individuals to be quarantined, especially as the public gets back to their normal routine. We investigate both requirements by the implementation of disease outbreak modeling and exploring its induced dynamic spatial risk in form of risk assessment, along with its real-time integration back into the disease model. This paper implements a contact tracing-based stochastic compartment model as a baseline, to further modify the existing setup to include the spatial risk. This modification of each individual-level contact's intensity to be dependent on its spatial location has been termed as Contextual Contact Tracing. The results highlight that the inclusion of spatial context tends to send more individuals into quarantine which reduces the overall spread of infection. With a simulated example of an induced spatial high-risk, it is highlighted that the new spatio-SIR model can act as a tool to empower the analyst with a capability to explore disease dynamics from a spatial perspective. We conclude that the proposed spatio-SIR tool can be of great help for policymakers to know the consequences of their decision prior to their implementation.

Detecting Phase-Synchrony Connectivity Anomalies in EEG Signals. Application to Dyslexia Diagnosis

Objective Dyslexia diagnosis is a challenging task, since traditional diagnosis methods are not based on biological markers but on behavioural tests. Although dyslexia diagnosis has been addressed by these tests in clinical practice, it is difficult to extract information about the brain processes involved in the different tasks and, then, to go deeper into its biological basis. Thus, the use of biomarkers can contribute not only to the diagnosis but also to a better understanding of specific learning disorders such as dyslexia. In this work, we use Electroencephalography (EEG) signals to discover differences among controls and dyslexic subjects using signal processing and artificial intelligence techniques. Specifically, we measure phase synchronization among channels, to reveal the functional brain network activated during auditory processing. On the other hand, to explore synchronicity patterns risen by low-level auditory processing, we used specific stimuli consisting in band-limited white noise, modulated in amplitude at different frequencies. The differential information contained in the functional (i.e., synchronization) network has been processed by an anomaly detection system that addresses the problem of subjects variability by an outlier-detection method based on vector quantization. The results, obtained for 7 years-old children, show that the proposed method constitutes an useful tool for clinical use, with the area under ROC curve (AUC) values up to 0.95 in differential diagnosis tasks.

Early prediction of developing spontaneous activity in cultured neuronal networks

Synchronization and bursting activity are intrinsic electrophysiological properties of in vivo and in vitro neural networks. During early development, cortical cultures exhibit a wide repertoire of synchronous bursting dynamics whose characterization may help to understand the parameters governing the transition from immature to mature networks. Here we used machine learning techniques to characterize and predict the developing spontaneous activity in mouse cortical neurons on microelectrode arrays (MEAs) during the first three weeks in vitro. Network activity at three stages of early development was defined by 18 electrophysiological features of spikes, bursts, synchrony, and connectivity. The variability of neuronal network activity during early development was investigated by applying k-means and self-organizing map (SOM) clustering analysis to features of bursts and synchrony. These electrophysiological features were predicted at the third week in vitro with high accuracy from those at earlier times using three machine learning models: Multivariate Adaptive Regression Splines, Support Vector Machines, and Random Forest. Our results indicate that initial patterns of electrical activity during the first week in vitro may already predetermine the final development of the neuronal network activity. The methodological approach used here may be applied to explore the biological mechanisms underlying the complex dynamics of spontaneous activity in developing neuronal cultures.

Kohonen Neural Network Investigation of the Effects of the Visual, Proprioceptive and Vestibular Systems to Balance in Young Healthy Adult Subjects

Kohonen neural network (KNN) was used to investigate the effects of the visual, proprioceptive and vestibular systems using the sway information in the mediolateral (ML) and anterior-posterior (AP) directions, obtained from an inertial measurement unit, placed at the lower backs of 23 healthy adult subjects (10 males, 13 females, mean (standard deviation) age: 24.5 (4.0) years, height: 173.6 (6.8) centimeter, weight: 72.7 (9.9) kg). The measurements were based on the modified Clinical Test of Sensory Interaction and Balance (mCTSIB). KNN clustered the subjects’ time-domain sway measures by processing their sway’s root mean square position, velocity, and acceleration. Clustering effectiveness was established using external performance indicators such as purity, precision-recall, and F-measure. Differences in these measures, from the clustering of each mCTSIB condition with its condition, were used to extract information about the balance-related sensory systems, where smaller values indicated reduced sway differences. The results for the parameters of purity, precision, recall, and F-measure were higher in the AP direction as compared to the ML direction by 7.12%, 11.64%, 7.12%, and 9.50% respectively, with their differences statistically significant (p < 0.05) thus suggesting the related sensory systems affect majorly the AP direction sway as compared to the ML direction sway. Sway differences in the ML direction were lowest in the presence of the visual system. It was concluded that the effect of the visual system on the balance can be examined mostly by the ML sway while the proprioceptive and vestibular systems can be examined mostly by the AP direction sway.

A dynamic analysis of household debt using a self-organizing map

The Korean consumer credit panel offers a well-organized set of microdata representing various characteristics of individual borrowers. To overcome the difficulty of fragmented microdata details, we construct a cluster of Korean consumers' credit, to develop a self-organizing map that visualizes individuals' characteristics along two dimensions. The result of cluster analysis reveals that most borrowers belong to one large cluster representing diligent borrowers who honor their loan payments. Conversely, several small clusters that represent borrowers with high default probability are identified, and we also found that these borrowers' characteristics vary. No significant change is found in the structure of the cluster, even when the aggregate amount of consumer credit is increased. Moreover, the expansionary monetary policy did not change the quantitative structure of household debt in Korea.

SUPPLEMENTARY INFORMATION

The online version contains supplementary material available at 10.1007/s00181-021-02120-5.

Deep Learning: A Comprehensive Overview on Techniques, Taxonomy, Applications and Research Directions

Deep learning (DL), a branch of machine learning (ML) and artificial intelligence (AI) is nowadays considered as a core technology of today’s Fourth Industrial Revolution (4IR or Industry 4.0). Due to its learning capabilities from data, DL technology originated from artificial neural network (ANN), has become a hot topic in the context of computing, and is widely applied in various application areas like healthcare, visual recognition, text analytics, cybersecurity, and many more. However, building an appropriate DL model is a challenging task, due to the dynamic nature and variations in real-world problems and data. Moreover, the lack of core understanding turns DL methods into black-box machines that hamper development at the standard level. This article presents a structured and comprehensive view on DL techniques including a taxonomy considering various types of real-world tasks like supervised or unsupervised. In our taxonomy, we take into account deep networks for supervised or discriminative learning, unsupervised or generative learning as well as hybrid learning and relevant others. We also summarize real-world application areas where deep learning techniques can be used. Finally, we point out ten potential aspects for future generation DL modeling with research directions. Overall, this article aims to draw a big picture on DL modeling that can be used as a reference guide for both academia and industry professionals.

Approximate spectral clustering using both reference vectors and topology of the network generated by growing neural gas

Spectral clustering (SC) is one of the most popular clustering methods and often outperforms traditional clustering methods. SC uses the eigenvectors of a Laplacian matrix calculated from a similarity matrix of a dataset. SC has serious drawbacks: the significant increases in the time complexity derived from the computation of eigenvectors and the memory space complexity to store the similarity matrix. To address the issues, I develop a new approximate spectral clustering using the network generated by growing neural gas (GNG), called ASC with GNG in this study. ASC with GNG uses not only reference vectors for vector quantization but also the topology of the network for extraction of the topological relationship between data points in a dataset. ASC with GNG calculates the similarity matrix from both the reference vectors and the topology of the network generated by GNG. Using the network generated from a dataset by GNG, ASC with GNG achieves to reduce the computational and space complexities and improve clustering quality. In this study, I demonstrate that ASC with GNG effectively reduces the computational time. Moreover, this study shows that ASC with GNG provides equal to or better clustering performance than SC.

Comparison of multimodal findings on epileptogenic side in temporal lobe epilepsy using self-organizing maps

OBJECTIVE

To develop a decision-making tool to evaluate and compare the performance of neuroimaging markers with clinical findings and the significance of attributes for presurgical lateralization of mesial temporal lobe epilepsy (mTLE).

METHODS

Thirty-five unilateral mTLE patients who qualified as candidates for surgical resection were studied. Seizure semiology, ictal EEG, ictal epileptogenic zone, interictal-irritative zone, and MRI findings were used as clinical markers. Hippocampal T1 volumetry and FLAIR intensity, DTI estimated; mean diffusivity (MD) in the hippocampus and fractional anisotropy (FA) in posteroinferior cingulum and crus of fornix, and the output of logistic regression method on volumetrics of the hippocampus, amygdala, and thalamus were adopted as neuroimaging markers. The self-organizing map (SOM) method was applied to markers to provide predictive methods for mTLE lateralization.

RESULTS

The SOM clustered all clinical attributes correctly with 100% accuracy and sensitivity for both the left and right mTLE. Among the clinical markers, seizure semiology and interictal-irrelative zone are the most sensitive attribute for the left-mTLE group lateralization. The accuracy achieved by applying the SOM method to the neuroimaging attributes was 94%, while the sensitivity was achieved 90% for left and 100% for right mTLE. SOM evidence indicated that the hippocampal volume is the most sensitive attribute for the prediction of the laterality in left-mTLE groups.

CONCLUSION

The proposed SOM method showed that neuroimaging markers may not replace with clinical findings. Nevertheless, multimodal neuroimaging can play an effective role in preoperative lateralization to reduce the costs and risks of surgical resection.

Species-environment relationships of fish and map-based variables in small boreal streams: Linkages with climate change and bioassessment

Species-environment relationships were studied between the occurrence of 13 fish and lamprey species and 9 mainly map-based environmental variables of Finnish boreal small streams. A self-organizing map (SOM) analysis showed strong relationships between the fish species and environmental variables in a single model (explained variance 55.9%). Besides basic environmental variables such as altitude, catchment size, and mean temperature, land cover variables were also explored. A logistic regression analysis indicated that the occurrence probability of brown trout, Salmo trutta L., decreased with an increasing percentage of peatland ditch drainage in the upper catchment. Ninespine stickleback, Pungitius pungitius (L.), and three-spined stickleback, Gasterosteus aculeatus L., seemed to benefit from urban areas in the upper catchment. Discovered relationships between fish species occurrence and land-use attributes are encouraging for the development of fish-based bioassessment for small streams. The presented ordination of the fish species in the mean temperature gradient will help in predicting fish community responses to climate change.

Content-aware data distribution over cluster nodes

Proper data items distribution may seriously improve the performance of data processing in distributed environment. However, typical datastorage systems as well as distributed computational frameworks do not pay special attention to that aspect. In this paper author introduces two custom data items addressing methods for distributed datastorage on the example of Scalable Distributed Two-Layer Datastore. The basic idea of those methods is to preserve that data items stored on the same cluster node are similar to each other following concepts of data clustering. Still, most of the data clustering mechanisms have serious problem with data scalability which is a severe limitation in Big Data applications. The proposed methods allow to efficiently distribute data set over a set of buckets. As it was shown by the experimental results, all proposed methods generate good results efficiently in comparison to traditional clustering techniques like k-means, agglomerative and birch clustering. Distributed environment experiments shown that proper data distribution can seriously improve the effectiveness of Big Data processing.

Oral health, organic and inorganic saliva composition of men with Schizophrenia: Case-control study

BACKGROUND

Schizophrenia (SCZ) presents complex challenges related to diagnosis and clinical monitoring. The study of conditions associated with SCZ can be facilitated by using potential markers and patterns that provide information to support the diagnosis and oral health.

METHODS

The salivary composition of patients diagnosed with SCZ (n = 50) was evaluated and compared to the control (n = 50). Saliva samples from male patients were collected and clinical parameters were evaluated. The concentration of total proteins and amylase were determined and salivary macro- and microelements were quantified by ICP OES and ICP-MS. Exploratory data analysis based on artificial intelligence tools was used in the investigation.

RESULTS

There was a significant increase in the salivary concentrations of Al, Fe, Li, Mg, Na, and V, higher prevalence of caries (p < 0.001), periodontal disease (p < 0.001), and reduced salivary flow rate (p = 0.019) in SCZ patients. Also, samples were grouped into six clusters. As, Co, Cr, Cu, Mn, Mo, Ni, Se, and Sr were correlated with each other, while Fe, K, Li, Ti, and V showed the highest concentrations in the samples distributed in the clusters with the highest association between SZC patients and controls.

CONCLUSIONS

The results obtained indicate changes in salivary flow, organic composition, and levels of macro- and microelements in SCZ patients. Salivary concentrations of Fe, Mg, and Na may be related to oral conditions, higher prevalence of caries, and periodontal disease. The exploratory analysis showed different patterns in the salivary composition of SCZ patients impacted by associations between oral health conditions and the use of medications. Future studies are encouraged to confirm the results investigated in this study.

Assessment of the PETase conformational changes induced by poly(ethylene terephthalate) binding

Recently, a bacterium strain of Ideonella sakaiensis was identified with the uncommon ability to degrade the poly(ethylene terephthalate) (PET). The PETase from I. sakaiensis strain 201-F6 (IsPETase) catalyzes the hydrolysis of PET converting it to mono(2-hydroxyethyl) terephthalic acid (MHET), bis(2-hydroxyethyl)-TPA (BHET), and terephthalic acid (TPA). Despite the potential of this enzyme for mitigation or elimination of environmental contaminants, one of the limitations of the use of IsPETase for PET degradation is the fact that it acts only at moderate temperature due to its low thermal stability. Besides, molecular details of the main interactions of PET in the active site of IsPETase remain unclear. Herein, molecular docking and molecular dynamics (MD) simulations were applied to analyze structural changes of IsPETase induced by PET binding. Results from the essential dynamics revealed that the β1-β2 connecting loop is very flexible. This loop is located far from the active site of IsPETase and we suggest that it can be considered for mutagenesis to increase the thermal stability of IsPETase. The free energy landscape (FEL) demonstrates that the main change in the transition between the unbound to the bound state is associated with the β7-α5 connecting loop, where the catalytic residue Asp206 is located. Overall, the present study provides insights into the molecular binding mechanism of PET into the IsPETase structure and a computational strategy for mapping flexible regions of this enzyme, which can be useful for the engineering of more efficient enzymes for recycling plastic polymers using biological systems.

Identification of SARS-CoV-2-induced pathways reveals drug repurposing strategies

The global outbreak of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) necessitates the rapid development of new therapies against coronavirus disease 2019 (COVID-19) infection. Here, we present the identification of 200 approved drugs, appropriate for repurposing against COVID-19. We constructed a SARS-CoV-2-induced protein network, based on disease signatures defined by COVID-19 multiomics datasets, and cross-examined these pathways against approved drugs. This analysis identified 200 drugs predicted to target SARS-CoV-2-induced pathways, 40 of which are already in COVID-19 clinical trials, testifying to the validity of the approach. Using artificial neural network analysis, we classified these 200 drugs into nine distinct pathways, within two overarching mechanisms of action (MoAs): viral replication (126) and immune response (74). Two drugs (proguanil and sulfasalazine) implicated in viral replication were shown to inhibit replication in cell assays. This unbiased and validated analysis opens new avenues for the rapid repurposing of approved drugs into clinical trials.

Machine Learning-Based Algorithms to Knowledge Extraction from Time Series Data: A Review

To predict the future behavior of a system, we can exploit the information collected in the past, trying to identify recurring structures in what happened to predict what could happen, if the same structures repeat themselves in the future as well. A time series represents a time sequence of numerical values observed in the past at a measurable variable. The values are sampled at equidistant time intervals, according to an appropriate granular frequency, such as the day, week, or month, and measured according to physical units of measurement. In machine learning-based algorithms, the information underlying the knowledge is extracted from the data themselves, which are explored and analyzed in search of recurring patterns or to discover hidden causal associations or relationships. The prediction model extracts knowledge through an inductive process: the input is the data and, possibly, a first example of the expected output, the machine will then learn the algorithm to follow to obtain the same result. This paper reviews the most recent work that has used machine learning-based techniques to extract knowledge from time series data.

Aquatic Macrophytes Determine the Spatial Distribution of Invertebrates in a Shallow Reservoir

Aquatic macrophytes determine the physical structure of many microhabitats in water and strongly influence the distribution of various aquatic animals. In this study, we analyzed the main microhabitat characteristics that affected the spatial distribution of invertebrates in shallow wetlands of South Korea (Jangcheok Reservoir). Environmental variables, macrophyte biomass, and invertebrate groups were used to analyze invertebrate distribution using a self-organizing map (SOM). Thirteen invertebrate groups were mapped onto the SOM, and each group was compared with the distribution of environmental variables and macrophyte biomass. Based on a U-matrix, five clusters were categorized according to Euclidean distance on the SOM. Invertebrate groups were closely related to macrophyte biomass. In particular, Lymnaeidae, Physidae, Viviparidae, Ecnomidae, and Hydrophilidae were abundant in quadrats with a high cover of Paspalum distichum and Nelumbo nucifera. Bithyniidae and Coenagrionidae were strongly associated with Trapa japonica and Hydrocharis dubia, whereas Planorbidae, Corduliidae, and Hydrophilidae were abundant with a high cover of Typha orientalis. Similar habitat preferences were found in a survey of gastropod distribution on the surface of each macrophyte species. The results clearly indicated that invertebrate distribution clusters were related to the spatial distribution of aquatic macrophytes in a shallow wetland.

A full-parallel implementation of Self-Organizing Maps on hardware

Self-Organizing Maps (SOMs) are extensively used for data clustering and dimensionality reduction. However, if applications are to fully benefit from SOM based techniques, high-speed processing is demanding, given that data tends to be both highly dimensional and yet "big". Hence, a fully parallel architecture for the SOM is introduced to optimize the system's data processing time. Unlike most literature approaches, the architecture proposed here does not contain sequential steps - a common limiting factor for processing speed. The architecture was validated on FPGA and evaluated concerning hardware throughput and the use of resources. Comparisons to the state of the art show a speedup of 8.91× over a partially serial implementation, using less than 15% of hardware resources available. Thus, the method proposed here points to a hardware architecture that will not be obsolete quickly.

Self-organizing map applied to the choice of internal standards for the determination of Cd, Pb, Sn, and platinum group elements by inductively coupled plasma mass spectrometry

The behaviors of internal standards, according to different flow rates of the cell collision gas (He), were studied for the determination of Cd, Pb, Pd, Pt, Rh, and Sn in samples of fish and mollusks by inductively coupled plasma mass spectrometry (ICP-MS). The elements Bi, Ge, In, Sc, and Y were selected as internal standards, considering their masses and first ionization energies. Addition and recovery experiments were carried out at three concentration levels to evaluate the accuracy of the method applied for the analysis of two samples with different matrices. The results were evaluated using a self-organizing map (SOM). The best analyte/IS pairs were as follows: ¹¹⁴Cd⁺/⁷⁴Ge⁺, ¹⁹⁵Pt⁺/⁷⁴Ge⁺, and ²⁰⁸Pb⁺/⁷⁴Ge⁺. For ¹⁰³Rh⁺, ¹⁰⁶Pd⁺, and ¹²⁰Sn⁺, greater accuracy was achieved without use of an internal standard. Helium gas (2.8 mL min^-1) was used in the collision cell for the analytes, except for Sn, and recoveries ranged from 98 to 101% under optimal conditions. The use of SOM as an exploratory analysis tool was an effective approach for selection of the most appropriate internal standards.

Assessment of extrinsic and intrinsic influences on water quality variation in subtropical agricultural multipond systems

Understanding wetland water quality dynamics and associated influencing factors is important to assess the numerous ecosystem services they provide. We present a combined self-organizing map (SOM) and linear mixed-effects model (LMEM) to relate water quality variation of multipond systems (MPSs, a common type of non-floodplain wetlands in agricultural regions of southern China) to their extrinsic and intrinsic influences for the first time. Across the 6 test MPSs with environmental gradients, ammonium nitrogen (NH₄⁺-N), total nitrogen (TN), and total phosphate (TP) almost always exceeded the surface water quality standard (2.0, 2.0, and 0.4 mg/L, respectively) in the up- and midstream ponds, while chlorophyll-a (Chl-a) exhibited hypertrophic state (≥28 μg/L) in the midstream ponds during the wet season. Synergistic influences explained 69±12% and 73±10% of the water quality variations in the wet and dry season, respectively. The adverse, extrinsic influences were generally 1.4, 6.9, 3.2, and 4.3 times of the beneficial, intrinsic influences for NH₄⁺-N, nitrate nitrogen (NO₃^--N), TP, and potassium permanganate index (COD_Mn), respectively, although the influencing direction and degree of forest and water area proportion were spatiotemporally unstable. While COD_Mn was primarily linked with rural residential areas in the midstream, higher TN and TP concentrations in the up- and midstream were associated with agricultural land, and NH₄⁺-N reflected a small but non-negligible source of free-range poultry feeding. Pond surface sediments exhibited consistent, adverse effects with amplifications during rainfall, while macrophyte biomass can reflect the biological uptake of COD_Mn and Chl-a, especially in the mid- and downstream during the wet season. Our study advances nonpoint source pollution (NPSP) research for small water bodies, explores nutrient "source-sink" dynamics, and provides a timely guide for rural planning and pond management. The modelling procedures and analytical results can inform refined assessment of similar NFWs elsewhere, where restoration efforts are required.

Rapid assessment tool for oil spill planning and contingencies

The Rapid Oil Spill Hazard Assessment is presented as a demonstration of concept for a tool providing a framework for managers and planners to assess potential impact areas of oil spills. The tool consists of precomputed oil spill scenarios derived from the analysis of twenty years of modeled current data using Self-Organizing Maps to identify 16 representative patterns. These patterns were used to provide boundary conditions for hydrodynamic and wave models to generate higher resolution current fields, used to drive a Lagrangian oil particle transport model creating the most probable oil spill dispersion patterns. To demonstrate the concept, the tool is applied to the Perdido region in the western Gulf of Mexico. A total of 896 oil spill simulations were performed, considering surface and bottom spills, and were stored in a database for easy access to map arrival probabilities and times to be used in risk and vulnerability analysis.

It Often Howls More than It Chugs: Wind versus Ship Noise Under Water in Australia’s Maritime Regions

Marine soundscapes consist of cumulative contributions by diverse sources of sound grouped into: physical (e.g., wind), biological (e.g., fish), and anthropogenic (e.g., shipping)—each with unique spatial, temporal, and frequency characteristics. In terms of anthropophony, shipping has been found to be the greatest (ubiquitous and continuous) contributor of low-frequency underwater noise in several northern hemisphere soundscapes. Our aim was to develop a model for ship noise in Australian waters, which could be used by industry and government to manage marine zones, their usage, stressors, and potential impacts. We also modelled wind noise under water to provide context to the contribution of ship noise. The models were validated with underwater recordings from 25 sites. As expected, there was good congruence when shipping or wind were the dominant sources. However, there was less agreement when other anthropogenic or biological sources were present (i.e., primarily marine seismic surveying and whales). Off Australia, pristine marine soundscapes (based on the dominance of natural, biological and physical sound) remain, in particular, near offshore reefs and islands. Strong wind noise dominates along the southern Australian coast. Underwater shipping noise dominates only in certain areas, along the eastern seaboard and on the northwest shelf, close to shipping lanes.

Artificial Neural Network, Quantile and Semi-Log Regression Modelling of Mass Appraisal in Housing

We used a large sample of 188,652 properties, which represented 4.88% of the total housing stock in Catalonia from 1994 to 2013, to make a comparison between different real estate valuation methods based on artificial neural networks (ANNs), quantile regressions (QRs) and semi-log regressions (SLRs). A literature gap in regard to the comparison between ANN and QR modelling of hedonic prices in housing was identified, with this article being the first paper to include this comparison. Therefore, this study aimed to answer (1) whether QR valuation modelling of hedonic prices in the housing market is an alternative to ANNs, (2) whether it is confirmed that ANNs produce better results than SLRs when assessing housing in Catalonia, and (3) which of the three mass appraisal models should be used by Spanish banks to assess real estate. The results suggested that the ANNs and SLRs obtained similar and better performances than the QRs and that the SLRs performed better when the datasets were smaller. Therefore, (1) QRs were not found to be an alternative to ANNs, (2) it could not be confirmed whether ANNs performed better than SLRs when assessing properties in Catalonia and (3) whereas small and medium banks should use SLRs, large banks should use either SLRs or ANNs in real estate mass appraisal.

A Neural Network Model for Learning 3D Object Representations Through Haptic Exploration

Humans initially learn about objects through the sense of touch, in a process called “haptic exploration.” In this paper, we present a neural network model of this learning process. The model implements two key assumptions. The first is that haptic exploration can be thought of as a type of navigation, where the exploring hand plays the role of an autonomous agent, and the explored object is this agent's “local environment.” In this scheme, the agent's movements are registered in the coordinate system of the hand, through slip sensors on the palm and fingers. Our second assumption is that the learning process rests heavily on a simple model of sequence learning, where frequently-encountered sequences of hand movements are encoded declaratively, as “chunks.” The geometry of the object being explored places constraints on possible movement sequences: our proposal is that representations of possible, or frequently-attested sequences implicitly encode the shape of the explored object, along with its haptic affordances. We evaluate our model in two ways. We assess how much information about the hand's actual location is conveyed by its internal representations of movement sequences. We also assess how effective the model's representations are in a reinforcement learning task, where the agent must learn how to reach a given location on an explored object. Both metrics validate the basic claims of the model. We also show that the model learns better if objects are asymmetrical, or contain tactile landmarks, or if the navigating hand is articulated, which further constrains the movement sequences supported by the explored object.

Identifying Spatiotemporal Patterns in Land Use and Cover Samples from Satellite Image Time Series

The use of satellite image time series analysis and machine learning methods brings new opportunities and challenges for land use and cover changes (LUCC) mapping over large areas. One of these challenges is the need for samples that properly represent the high variability of land used and cover classes over large areas to train supervised machine learning methods and to produce accurate LUCC maps. This paper addresses this challenge and presents a method to identify spatiotemporal patterns in land use and cover samples to infer subclasses through the phenological and spectral information provided by satellite image time series. The proposed method uses self-organizing maps (SOMs) to reduce the data dimensionality creating primary clusters. From these primary clusters, it uses hierarchical clustering to create subclusters that recognize intra-class variability intrinsic to different regions and periods, mainly in large areas and multiple years. To show how the method works, we use MODIS image time series associated to samples of cropland and pasture classes over the Cerrado biome in Brazil. The results prove that the proposed method is suitable for identifying spatiotemporal patterns in land use and cover samples that can be used to infer subclasses, mainly for crop-types.

Exploring the Effectiveness of Clustering Algorithms for Capturing Water Consumption Behavior at Household Level

As water scarcity becomes more prevalent, the analysis of urban water consumption patterns at the consumer level and the estimation of the corresponding water demand for water utility are expected to be among the top priorities of water companies in the near future. This study proposes a comprehensive methodology for water managers to achieve an efficient operation of urban water networks, by successfully detecting residential water consumption patterns corresponding to different household needs and behaviors. The methodology uses Self Organizing Maps as the main clustering algorithm in combination with K-means and Hierarchical Agglomerative Clustering. The objective is to create clusters in a literature dataset that includes water consumption from 21 customers located in Milford, Ohio, USA, for a 7-month period. Originally, water consumption data was recorded for every water use incident in the household, while for this analysis, the information is converted to half-hourly water consumption. Individual customers with similar consumption behavior are clustered and water-consumption curves are calculated for each cluster; these curves can be used by the water utility to obtain estimates of the spatio-temporal distribution of demand, thus giving insight into peak demands at different locations. Statistical indices of agreement are used to confirm a good agreement between the estimated and observed water use, when clustering is employed. The resulting curves show a clear improvement in capturing water consumption behavior at household level, when compared to corresponding curves obtained without clustering. This analysis offers water utilities an innovative solution that relies on real time data and uses data science principles for optimizing water supply and network operation and provides tools for the efficient use of water resources.

Current trends in flow cytometry automated data analysis software

Automated flow cytometry (FC) data analysis tools for cell population identification and characterization are increasingly being used in academic, biotechnology, pharmaceutical, and clinical laboratories. The development of these computational methods is designed to overcome reproducibility and process bottleneck issues in manual gating, however, the take-up of these tools remains (anecdotally) low. Here, we performed a comprehensive literature survey of state-of-the-art computational tools typically published by research, clinical, and biomanufacturing laboratories for automated FC data analysis and identified popular tools based on literature citation counts. Dimensionality reduction methods ranked highly, such as generic t-distributed stochastic neighbor embedding (t-SNE) and its initial Matlab-based implementation for cytometry data viSNE. Software with graphical user interfaces also ranked highly, including PhenoGraph, SPADE1, FlowSOM, and Citrus, with unsupervised learning methods outnumbering supervised learning methods, and algorithm type popularity spread across K-Means, hierarchical, density-based, model-based, and other classes of clustering algorithms. Additionally, to illustrate the actual use typically within clinical spaces alongside frequent citations, a survey issued by UK NEQAS Leucocyte Immunophenotyping to identify software usage trends among clinical laboratories was completed. The survey revealed 53% of laboratories have not yet taken up automated cell population identification methods, though among those that have, Infinicyt software is the most frequently identified. Survey respondents considered data output quality to be the most important factor when using automated FC data analysis software, followed by software speed and level of technical support. This review found differences in software usage between biomedical institutions, with tools for discovery, data exploration, and visualization more popular in academia, whereas automated tools for specialized targeted analysis that apply supervised learning methods were more used in clinical settings.

Optimizing Wave Overtopping Energy Converters by ANN Modelling: Evaluating the Overtopping Rate Forecasting as the First Step

Artificial neural networks (ANN) are extremely powerful analytical, parallel processing elements that can successfully approximate any complex non-linear process, and which form a key piece in Artificial Intelligence models. Its field of application, being very wide, is especially suitable for the field of prediction. In this article, its application for the prediction of the overtopping rate is presented, as part of a strategy for the sustainable optimization of coastal or harbor defense structures and their conversion into Waves Energy Converters (WEC). This would allow, among others benefits, reducing their initial high capital expenditure. For the construction of the predictive model, classical multivariate statistical techniques such as Principal Component Analysis (PCA), or unsupervised clustering methods like Self Organized Maps (SOM), are used, demonstrating that this close alliance is always methodologically beneficial. The specific application carried out, based on the data provided by the CLASH and EurOtop 2018 databases, involves the creation of a useful application to predict overtopping rates in both sloping breakwaters and seawalls, with good results both in terms of prediction error, such as correlation of the estimated variable.

Spatial patterning of chlorophyll a and water-quality measurements for determining environmental thresholds for local eutrophication in the Nakdong River basin

Management of water-quality in a river ecosystem needs to be focused on susceptible regions to eutrophication based on proper measurements. The stress-response relationships between nutrients and primary productivity of phytoplankton allow the derivation of ecologically acceptable thresholds of stressors under field conditions. However, spatio-temporal variations in heterogeneous environmental conditions have hindered the development of locally applicable criteria. To address these issues, we utilized a combination of a geographically specialized artificial neural network (Geo-SOM, geo-self-organizing map) and linear mixed-effect models (LMMs). The model was applied to a 24-month dataset of 54 stations that spanned a wide spatial gradient in the Nakdong River basin. The Geo-SOM classified 1286 observations in the basin into 13 clusters that were regionally and seasonally distinct. Inclusion of the random effects of Geo-SOM clustering improved the performance of each LMM, which suggests that there were significant spatio-temporal variations in the Chla-stressor relationships. These variations arise owing to differences in background seasonality and the effects of local pollutant variables and land-use patterns. Among the 16 environmental variables, the major stressors for Chla were total phosphate (TP) as a nutrient and biological oxygen demand (BOD) as a non-nutrient according to the results of both Geo-SOM and LMM analyses. Based on LMMs with the random effect of the Geo-SOM clusters on the intercept and the slope, we can propose recommended thresholds for TP (18.5 μg L^-1) and BOD (1.6 mg L^-1) in the Nakdong River. The combined method of LMM and Geo-SOM will be useful in guiding appropriate local water-quality-management strategies and in the global development of large-scale nutrient criteria.

Enrichment, sources and ecological risk mapping of heavy metals in agricultural soils of dhaka district employing SOM, PMF and GIS methods

Rapid urbanization and industrial growth have triggered heavy metal contamination in agricultural soil in Dhaka, which is a serious concern for ecological risk and public health issues. In this study, fifty-four soil samples from agricultural lands of Dhaka had been analyzed for assessing accumulation, spatial enrichment, ecological risk and sources apportionment of heavy metals using a combined approach of self-organizing map (SOM), positive matrix factorization (PMF), geographical information system (GIS), and enrichment factor (EF). The results of the enrichment factor, geoaccumulation index and contamination factor index showed that more than 90% of the soil samples were polluted by higher levels of Cr and Cd. The mean pollution load index (PLI) results demonstrated that about 73% of soil samples were moderately polluted by heavy metals. Based on SOM and PMF analysis, four potential sources of heavy metals were found in this study area: (i) agrochemical and sewage irrigation (Cd-As); (ii) combined effect of agriculture, industrial and natural sources (Mn, Co, Ni and Zn); (iii) atmospheric deposition and industrial emission (As-Pb); (iv) chemical and leather tanning industries (Cr). The ecological risk index demonstrated that in terms of Cd content, about 75% of soil samples were moderate to high risk, and 20% were moderate to considerable ecological risk, which was the serious environmental, ecological, and public health concern. The spatial projection of ecological risk values showed that the southern part of Dhaka (Keraniganj Upazila) is a high ecological risk in terms of heavy metal pollution. These risk maps in agricultural soils may play a vital role in reducing pollution sources; so that zonal pollution control, as well as ecological protection, may be achieved in this resource-based agricultural land.

OCEAN: A Non-Conventional Parameter Free Clustering Algorithm Using Relative Densities of Categories

In this paper, we propose a fully autonomous density-based clustering algorithm named ‘Ocean’, which is inspired by the oceanic landscape and phenomena that occur in it. Ocean is an improvement over conventional algorithms regarding both distance metric and the clustering mechanism. Ocean defines the distance between two categories as the difference in the relative densities of categories. Unlike existing approaches, Ocean neither assigns the same distance to all pairs of categories, nor assigns arbitrary weights to matches and mismatches between categories that can lead to clustering errors. Ocean uses density ratios of adjacent regions in multidimensional space to detect the edges of the clusters. Ocean is robust against clusters of identical patterns. Unlike conventional approaches, Ocean neither makes any assumption regarding the data distribution within clusters, nor requires tuning of free parameters. Empirical evaluations demonstrate improved performance of Ocean over existing approaches.

Beliefs in Conflict: The Management of Teno Atlantic Salmon in the Sámi Homeland in Finland

The subarctic Teno River is one of the most significant spawning rivers for Atlantic salmon in Europe. In 2009, research indicated that the Teno salmon stock was in a weak state, and concern about the future of Atlantic salmon in the Teno River arose on both sides of the river, in Finland and Norway. In 2017, the governments ratified the new Teno fishing agreement (Teno Fishing Act 2017). The agreement aimed to reduce the fishing volume by 30%, and the new regulations concerned all users, including the indigenous Sámi, other locals, tourists, and fishing entrepreneurs. This triggered concern and anger in the Sámi community and among other locals generally. The dispute raised a question concerning the management of Teno salmon. We conducted a Q inquiry with 43 statements, covering aspects of interest, knowledge, management, and policy needs related to Teno salmon. We hypothesised that the key reason for the management tensions lay in how scientific and traditional knowledge fitted administrative knowledge requirements. By using self-organising maps (SOMs), four webs of beliefs emerged from the data: traditional Sámi fishing; salmon protection; equal economic opportunity; and evidence-based decision-making. We also further analysed the statements according to how they reproduced diverging and similar beliefs. We discuss the identity-related struggle, rights, and stakes and the underlying issue of confidence and respect.

Robust identification of Parkinson's disease subtypes using radiomics and hybrid machine learning

OBJECTIVES

It is important to subdivide Parkinson's disease (PD) into subtypes, enabling potentially earlier disease recognition and tailored treatment strategies. We aimed to identify reproducible PD subtypes robust to variations in the number of patients and features.

METHODS

We applied multiple feature-reduction and cluster-analysis methods to cross-sectional and timeless data, extracted from longitudinal datasets (years 0, 1, 2 & 4; Parkinson's Progressive Marker Initiative; 885 PD/163 healthy-control visits; 35 datasets with combinations of non-imaging, conventional-imaging, and radiomics features from DAT-SPECT images). Hybrid machine-learning systems were constructed invoking 16 feature-reduction algorithms, 8 clustering algorithms, and 16 classifiers (C-index clustering evaluation used on each trajectory). We subsequently performed: i) identification of optimal subtypes, ii) multiple independent tests to assess reproducibility, iii) further confirmation by a statistical approach, iv) test of reproducibility to the size of the samples.

RESULTS

When using no radiomics features, the clusters were not robust to variations in features, whereas, utilizing radiomics information enabled consistent generation of clusters through ensemble analysis of trajectories. We arrived at 3 distinct subtypes, confirmed using the training and testing process of k-means, as well as Hotelling's T2 test. The 3 identified PD subtypes were 1) mild; 2) intermediate; and 3) severe, especially in terms of dopaminergic deficit (imaging), with some escalating motor and non-motor manifestations.

CONCLUSION

Appropriate hybrid systems and independent statistical tests enable robust identification of 3 distinct PD subtypes. This was assisted by utilizing radiomics features from SPECT images (segmented using MRI). The PD subtypes provided were robust to the number of the subjects, and features.

Overcome chromatism of metasurface via Greedy Algorithm empowered by self-organizing map neural network

Chromatism generally exists in most metasurfaces. Because of this, the deflected angle of metasurface reflectors usually varies with frequency. This inevitably hinders wide applications of metasurfaces to broadband signal scenarios. Therefore, it is of great significance to overcome chromatism of metasurfaces. With this aim, we firstly analyze necessary conditions for achromatic metasurface deflectors (AMD) and deduce the ideal dispersions of meta-atoms. Then, we establish a Self-Organizing Map (SOM) Neural Network as a prepositive model to obtain a diversified searching map, which is then applied to Greedy Algorithm to search meta-atoms with the required dispersions. Using these meta-atoms, an AMD was designed and simulated, with a thickness about 1/15 the central wavelength. A prototype was fabricated and measured. Both the simulation and measurement show that the proposed AMD can achieve an almost constant deflected angle of 22° under normal incidence within 9.5-10.5GHz. This method may find wide applications in designing functional metasurfaces for satellite communications, mobile wireless communications and others.

Analysis of the Use of Geomorphic Elements Mapping to Characterize Subaqueous Bedforms Using Multibeam Bathymetric Data in River System

Riverbed micro-topographical features, such as crest and trough, flat bed, and scour pit, indicate the evolution of fluvial geomorphology, and have an influence on the stability of underwater structures and overall scour pits. Previous studies on bedform feature extraction have focused mainly on the rhythmic bed surface morphology and have extracted crest and trough, while flat bed and scour pit have been ignored. In this study, to extend the feature description of riverbeds, geomorphic elements mapping was used by employing three geomorphic element classification methods: Wood’s criteria, a self-organization map (SOM) technique, and geomorphons. The results showed that geomorphic element mapping can be controlled by adjusting the slope tolerance and curvature tolerance of Wood’s criteria, using the map unit number and combination of the SOM technique and the flatness of geomorphons. Relatively flat bed can be presented using “plane”, “flat planar”, and “flat” elements, while scour pit can be presented using a “pit” element. A comparison of the difference between parameter settings for landforms and bedforms showed that SOM using 8 or 10 map units is applicable for land and underwater surface and is thus preferentially recommended for use. Furthermore, the use of geomorphons is recommended as the optimal method for characterizing bedform features because it provides a simple element map in the absence of area loss.

Interferons and viruses induce a novel truncated ACE2 isoform and not the full-length SARS-CoV-2 receptor

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, transcriptionally independent truncated isoform of ACE2, which we designate as deltaACE2 (dACE2). We demonstrate that dACE2, but not ACE2, is an ISG. In The Cancer Genome Atlas, the expression of dACE2 was enriched in squamous tumors of the respiratory, gastrointestinal and urogenital tracts. In vitro, dACE2, which lacks 356 amino-terminal amino acids, was non-functional in binding the SARS-CoV-2 spike protein and as a carboxypeptidase. Our results suggest that the ISG-type induction of dACE2 in IFN-high conditions created by treatments, an inflammatory tumor microenvironment or viral co-infections is unlikely to increase the cellular entry of SARS-CoV-2 and promote infection.