Individual Network Research Articles

A randomized controlled trial of changes in resting-state functional connectivity associated with short-term motor learning of chopstick use with the non-dominant hand

IntroductionThis study identified the offline brain networks associated with motor learning of non-dominant hand chopstick use within-session. Methods40 healthy right-handed adults were randomly assigned to the practice and control groups (20 each). The performance, resting-state functional connectivity (RSFC), and their correlation were compared within and between groups. Both groups repeated 9 cycles of 30 s task and rest. During the task, the practice group performed the chopstick-use practice with their left hand, while the control group held chopsticks without acquiring any skills. During the rest, both groups fixated their gaze on a fixation point. The number of times candies were moved using chopsticks with the left hand in 30 s was used to evaluate the performance. RSFC was obtained by resting-state fMRI scanning and extracting Z-scores between the right primary motor cortex and all other brain regions. ResultsBoth the groups improved in the post-task performance; the practice group improved more. The RSFC of the two networks increased in the practice group. One network was the RSFC between the right M1 and the right cerebellar Crus I, positively correlated with performance in the post-task. Another was the RSFC between the right M1 and the left cerebellar Crus II, positively correlated with skills in the amount of change pre- and post-task. ConclusionOffline enhancement of RSFC in these networks was shown to contribute to early chopstick-use motor learning with the left hand. These results serve as a basis for future studies on compensatory networks in individuals with stroke.

Webs of science: mentor networks influence women's integration into STEM fields

Mentorship can be part of the solution to developing a more diverse global scientific workforce, but robust longitudinal evidence is limited. Developmental mentor network theory can advance our understanding of the impact of a wide range of mentors across social contexts by distinguishing between the content of mentorship support (eg career support) and the structural characteristics of an individual's mentor network (eg density of connections among mentors). We tested the influence of mentor network characteristics on longitudinal social integration into the Earth and environmental sciences, as indicated by science identity development (a key indicator of social integration) and graduate‐school applications in STEM (science, technology, engineering, and mathematics)‐related fields of study, based on a sample of 233 undergraduate women at nine universities in the US. Our findings indicated that belonging to close‐knit, larger, and skill‐focused mentorship networks creates a “sticky web” of social connections, providing information and resources that increase retention of college women in the Earth and environmental sciences.

Effect of Extremely Preterm Birth on Adolescent Brain Network Organization.

Introduction: Extremely preterm (EPT) birth, defined as birth at a gestational age (GA) < 28 weeks, can have a lasting impact on cognition throughout the lifespan. Previous investigations reveal differences in brain structure and connectivity between infants born preterm and full-term, but how does preterm birth impact the adolescent connectome? Methods: Here, we investigate how EPT birth can alter broadscale network organization later in life by comparing resting state fMRI connectome-based parcellations of the entire cortex in adolescents born EPT (N=22) to age-matched adolescents born full-term (FT, GA ≥ 37 weeks, N=28). We compare these parcellations to adult parcellations from previous studies and explore the relationship between an individual's network organization and behavior. Results: Primary (occipital and sensorimotor) and frontoparietal networks were observed in both groups. However, there existed notable differences in the limbic and insular networks. Surprisingly, the connectivity profile of the limbic network of EPT adolescents was more adultlike than the same network in FT adolescents. Finally, we found a relationship between an adolescent's overall cognition score and their limbic network maturity. Discussion: Overall, preterm birth may contribute to atypical development of broadscale network organization in adolescence and may partially explain the observed cognitive deficits.

Specific topology and topological connection sensitivity enhanced graph learning for lncRNA–disease association prediction

Predicting disease-related candidate long noncoding RNAs (lncRNAs) is beneficial for exploring disease pathogenesis due to the close relations between lncRNAs and the occurrence and development of human diseases. It is a long-term and challenging task to adequately extract specific and local topologies in individual lncRNA network and individual disease network, and integrate the information of the connection relationships. We propose a new graph learning-based prediction method to encode specific and local topologies from each individual network, neighbor topologies with different connection relationships, and pairwise attributes. We first construct a lncRNA network composed of all the lncRNA nodes and their similarities, and a single disease network that contains all the disease nodes and disease similarities. Then, a network-aware graph convolutional autoencoder is constructed to encode the specific and local topologies of each network. Secondly, a heterogeneous network is established to embed all lncRNA, disease, and miRNA nodes and their various connections. Afterwards, a connection-sensitive graph neural network is designed to deeply integrate the neighbor node attributes and connection characteristics in the heterogeneous network and learn neighbor topological representations. We also construct both connection-level and topology representation-level attention mechanisms to extract informative connections and topological representations. Finally, we build a multi-layer convolutional neural networks with weighted residuals to adaptively complement the detailed features to pairwise attribute encoding. Comprehensive experiments and comparison results demonstrated that NCPred outperforms seven state-of-the-art prediction methods. The ablation studies demonstrated the importance of local topology learning, neighbor topology learning, and pairwise attribute encoding. Case studies on prostate, lung, and breast cancers further revealed NCPred’s capacity to screen potential candidate disease-related lncRNAs.

Species-level drivers of avian centrality within seed-dispersal networks across different levels of organisation.

Bird-plant seed-dispersal networks are structural components of ecosystems. The role of bird species in seed-dispersal networks (from less [peripheral] to more connected [central]), determines the interaction patterns and their ecosystem services. These roles may be driven by morphological and functional traits as well as evolutionary, geographical and environmental properties acting at different spatial extents. It is still unknown if such drivers are equally important in determining species centrality at different network levels, from individual local networks to the global meta-network representing interactions across all local networks. Using 308 networks covering five continents and 11 biogeographical regions, we show that at the global meta-network level species' range size was the most important driver of species centrality, with more central species having larger range sizes, which would facilitate the interaction with a higher number of plants and thus the maintenance of seed-dispersal interactions. At the local network level, body mass was the only driver with a significant effect, implying that local factors related to resource availability are more important at this level of network organisation than those related to broad spatial factors such as range sizes. This could also be related to the mismatch between species-level traits, which do not consider intraspecific variation, and the local networks that can depend on such variation. Taken together, our results show that the drivers determining species centrality are relative to the levels of network organisation, suggesting that prediction of species functional roles in seed-dispersal interactions requires combined local and global approaches.

Comparison of CNN-based deep learning architectures for rice diseases classification

Although convolutional neural network (CNN) paradigms have expanded to transfer learning and ensemble models from original individual CNN architectures, few studies have focused on the performance comparison of the applicability of these techniques in detecting and localizing rice diseases. Moreover, most CNN-based rice disease detection studies only considered a small number of diseases in their experiments. Both these shortcomings were addressed in this study. In this study, a rice disease classification comparison of six CNN-based deep-learning architectures (DenseNet121, Inceptionv3, MobileNetV2, resNext101, Resnet152V, and Seresnext101) was conducted using a database of nine of the most epidemic rice diseases in Bangladesh. In addition, we applied a transfer learning approach to DenseNet121, MobileNetV2, Resnet152V, Seresnext101, and an ensemble model called DEX (Densenet121, EfficientNetB7, and Xception) to compare the six individual CNN networks, transfer learning, and ensemble techniques. The results suggest that the ensemble framework provides the best accuracy of 98%, and transfer learning can increase the accuracy by 17% from the results obtained by Seresnext101 in detecting and localizing rice leaf diseases. The high accuracy in detecting and categorisation rice leaf diseases using CNN suggests that the deep CNN model is promising in the plant disease detection domain and can significantly impact the detection of diseases in real-time agricultural systems. This research is significant for farmers in rice-growing countries, as like many other plant diseases, rice diseases require timely and early identification of infected diseases and this research develops a rice leaf detection system based on CNN that is expected to help farmers to make fast decisions to protect their agricultural yields and quality.

Abnormal cerebral metabolism and metabolic connectivity in individuals with heroin dependence: an integrated resting-state PET/fMRI study in large-scale networks.

Increasing evidence suggests that heroin addiction may be related to the dysfunction among the triple brain network (default mode network [DMN], salience network [SN] and executive control network [ECN]). However, the characteristics of glucose metabolism and metabolic connectivity among core regions of the triple brain network remain unknown. Therefore, we hypothesized that individuals with heroin dependence would show abnormal glucose metabolism and accompanied abnormal metabolic connectivity within the triple brain network. Individuals with heroin dependence and healthy controls matched for age and sex underwent integrated positron emission tomography/magnetic resonance imaging (PET/MRI). Differences in glucose metabolism and metabolic connectivity among the DMN, SN and ECN were analyzed based on 18F-fluorodeoxyglucose PET and resting-state fMRI data. We included 36 individuals with heroin dependence and 30 matched healthy controls in our study. The heroin dependence group showed a significant reduction of glucose metabolism in the bilateral anterior insula (AI) and inferior parietal lobule (IPL), and a significantly decreased metabolic connectivity between the right AI and the left dorsolateral prefrontal cortex (DLPFC). The daily dose of methadone was negatively correlated with glucose metabolism of the right AI and right IPL. The results revealed the glucose metabolism alterations and metabolic connectivity only within the triple brain network in individuals with heroin dependence; additional brain networks should be investigated in future studies. Although methadone is an opioid with a similar neurophysiological mechanism as heroin, the specific chronic effects of methadone on cerebral metabolism and metabolic connectivity should also be investigated in future studies. Our findings suggest that long-term opioid use might, to some extent, be associated with reduced synergistic ability between the SN and ECN, which may be associated with the dysfunction of cognitive control. In particular, the right AI, which showed hypometabolism and related reduction in SN-ECN metabolic connectivity, should receive increasing attention in future studies.

A Deep Learning-Informed Design Scheme for Shear Friction at Concrete-to-Concrete Interface: Recommendations for Inclusion in AASHTO LRFD Guidelines

Recent advancements in construction technology have led to high-strength concrete and steel. However, these developments have depreciated the accuracy of the design equations in current provisions, which were based on normal-grade materials. To fill such a research gap, this study presents a novel deep learning-based computation scheme that can replace the current design provisions by virtue of its superior accuracy and reliability. The proposed approach exploits Neural Additive Models (NAMs) in which geometric and material properties associated with a normalweight concrete-to-concrete shear interface are inputted to individual neural network blocks. The outputs of the individual blocks are linearly combined to produce the prediction for interfacial shear strength. This model provides a way to identify and quantify the individual contributions of the input parameters, thus enhancing the interpretability of the model predictions for shear strength at the normalweight concrete-to-concrete interface. The deep learning-informed design (LID) scheme improves the prediction accuracy of the shear strength equation in the existing AASHTO LRFD Bridge Design Specifications by over 32%.

Personalized dynamic network models of the human brain as a future tool for planning and optimizing epilepsy therapy.

Epilepsy is a common neurological disorder, with one third of patients not responding to currently available antiepileptic drugs. The proportion of pharmacoresistant epilepsies has remained unchanged for many decades. To cure epilepsy and control seizures requires a paradigm shift in the development of new approaches to epilepsy diagnosis and treatment. Contemporary medicine has benefited from the exponential growth of computational modeling, and the application of network dynamics theory to understanding and treating human brain disorders. In epilepsy, the introduction of these approaches has led to personalized epileptic network modeling that can explore the patient's seizure genesis and predict the functional impact of resection on its individual network's propensity to seize. The application of the dynamic systems approach to neurostimulation therapy of epilepsy allows designing stimulation strategies that consider the patient's seizure dynamics and long-term fluctuations in the stability of their epileptic networks. In this article, we review, in a nontechnical fashion suitable for a broad neuroscientific audience, recent progress in personalized dynamic brain network modeling that is shaping the future approach to the diagnosis and treatment of epilepsy.

Altered topology of individual brain structural covariance networks in major depressive disorder.

The neurobiological pathogenesis of major depression disorder (MDD) remains largely controversial. Previous literatures with limited sample size utilizing group-level structural covariance networks (SCN) commonly generated mixed findings regarding the topology of brain networks. We analyzed T1 images from a high-powered multisite sample including 1173 patients with MDD and 1019 healthy controls (HCs). We used regional gray matter volume to construct individual SCN by utilizing a novel approach based on the interregional effect size difference. We further investigated MDD-related structural connectivity alterations using topological metrics. Compared to HCs, the MDD patients showed a shift toward randomization characterized by increased integration. Further subgroup analysis of patients in different stages revealed this randomization pattern was also observed in patients with recurrent MDD, while the first-episode drug naïve patients exhibited decreased segregation. Altered nodal properties in several brain regions which have a key role in both emotion regulation and executive control were also found in MDD patients compared with HCs. The abnormalities in inferior temporal gyrus were not influenced by any specific site. Moreover, antidepressants increased nodal efficiency in the anterior ventromedial prefrontal cortex. The MDD patients at different stages exhibit distinct patterns of randomization in their brain networks, with increased integration during illness progression. These findings provide valuable insights into the disruption in structural brain networks that occurs in patients with MDD and might be useful to guide future therapeutic interventions.

Community capacity for social enterprise development: Empirical evidence from community forest enterprises (CFEs) in Cameroon

With increasing forest devolution globally, community forest enterprises (CFEs) are emerging as potential options for local development based on forest resources. CFEs trade to meet their economic, social, and environmental goals; however, empirical studies have highlighted capacity deficiencies as key drawbacks to their development. Knowledge of what these capacity gaps are is low. This paper uses a systematic framework to explore capacity and deficiencies in capacity and the relation between the two within CFEs in Cameroon. Using the contextualised organisational capacity theory in combination with asset-based mobilisation theory, data was gathered from 31 CFEs based on focus group discussions. Principal component analysis and descriptive statistics were used to evaluate community capacity at individual, organisational (CFE), and network levels. Pearson correlation tests were used to explore the relationships between different domains of community capacity. The results indicate that community members and development practitioners agree that communities are weak in creating partnerships, networking, and resource mobilisation. The participatory community evaluation highlights major capacity gaps in infrastructure, members' skills and knowledge, and sense of community. This confirms that capacity gaps were larger at the individual and social network level, while organisational capacity scored higher except for a sense of community and resource mobilization. Although the community capacities at the individual level are generally low, these capacities correlate strongly with organisational capacity. The potential of the CFEs to meet objectives depends a lot on the capacities of individual members and networking capacity, indicating that these should be prioritised in development efforts, which should involve multiple stakeholders, with policy support and participation by the entire community.

Associations between postmigration living situation and symptoms of common mental disorders in adult refugees in Europe: updating systematic review from 2015 onwards

BackgroundRefugees and asylum seekers have a high prevalence of psychiatric disorders such as post-traumatic stress disorder (PTSD), anxiety, and depression. The postmigration context inheres different risk and protective factors for mental health of refugees and asylum seekers in host countries. We conducted a systematic review to update knowledge on the association between characteristics of the postmigration living situation (PMLS) and mental health outcomes in Europe since 2015.MethodsWe searched in five databases according to the PRISMA statement. From a total of 5,579 relevant studies published in 2015–22, 3,839 were included for title and abstract screening, and 70 full texts screened for eligibility. Out of these, 19 studies on refugees and asylum seekers conducted in European countries after 2014 were included in this systematic review. The quality of studies was assessed by using the Mixed Methods Appraisal Tool (MMAT) – version 2018. We performed a narrative synthesis using the four layers of the social determinants of health framework.ResultsA wide range of risk and protective factors for mental health in the PMLS were identified as exposure measures, which included individual factors (e.g., language skills), social and community networks (e.g., family concerns, loneliness and social support, discrimination), living and working conditions (e.g., legal status, duration of residence, unemployment and financial hardship, housing) as well as general socio-economic, cultural and environmental factors (e.g., social status, acculturation). We found postmigration stressors are positively associated with symptoms of depression, anxiety, and PTSD, albeit not consistently so. Especially, the general socio-economic, cultural and environmental factors showed weak associations with mental health.ConclusionsHeterogenous study characteristics likely explain the inconsistent associations between characteristics of the PMLS and mental health outcomes. However, broken down in its component layers, most risk and protective factors of the PMLS were significantly associated with symptoms of mental disorders showing the same direction of association across the included studies, while the association between some stressors or resources of the PMLS and mental health turns out to be less homogeneous than expected. Characteristics of the PMLS contribute to the high prevalence of mental diseases of refugees and asylum seekers. Disadvantages in general socio-economic conditions, living and working conditions, in access to social and community networks need to be redressed, in addition to better access to health care.

Moderation, Networks, and Anti-Social Behavior Online

Major open platforms, such as Facebook, Twitter, Instagram, and Tik Tok, are bombarded with postings that violate platform community standards, offend societal norms, and cause harm to individuals and groups. To manage such sites requires identification of content and behavior that is anti-social and action to remove content and sanction posters. This process is not as straightforward as it seems: what is offensive and to whom varies by individual, group, and community; what action to take depends on stated standards, community expectations, and the extent of the offense; conversations can create and sustain anti-social behavior (ASB); networks of individuals can launch coordinated attacks; and fake accounts can side-step sanctioning behavior. In meeting the challenges of moderating extreme content, two guiding questions stand out: how do we define and identify ASB online? And, given the quantity and nuances of offensive content: how do we make the best use of automation and humans in the management of offending content and ASB? To address these questions, existing studies on ASB online were reviewed, and a detailed examination was made of social media moderation practices on major media. Pros and cons of automated and human review are discussed in a framework of three layers: environment, community, and crowd. Throughout, the article adds attention to the network impact of ASB, emphasizing the way ASB builds a relation between perpetrator(s) and victim(s), and can make ASB more or less offensive.

Individual modelling of haematotoxicity with NARX neural networks: A knowledge transfer approach

Cytotoxic cancer therapy often results in dose-limiting haematotoxic side effects. Predicting an individual's risk is a major objective in precision medicine of cancer treatment. In this regard, patient heterogeneity presents a significant challenge. In this paper, we explore the use of hypothesis-free machine learning models based on recurrent nonlinear auto-regressive networks with exogenous inputs (NARX) as an approach to achieve this goal. Also, we propose a knowledge transfer approach to ameliorate the issue of sparse individual data, which typically hampers learning of individual networks. We demonstrate the feasibility of our approach based on a virtual patient population generated using a semi-mechanistic model of haematopoiesis and imposing different cytotoxic therapy scenarios on it. Employing different techniques of model optimisation, we derive robust and parsimonious individual networks with good generalisation performances. Moreover, we analyse in detail possible factors influencing the generalisation performance. Results suggest that our transfer learning approach using NARX networks can provide robust predictions of individual patient's response to treatment. As a practical perspective, we apply our approach to individual time series data of two patients with non-Hodgkin's lymphoma.

Temporal Data Pooling With Meta-Initialization for Individual Short-Term Load Forecasting

The growth of advanced metering infrastructure (AMI) deployment has enabled intelligent power control and management at the customer level; Highly accurate individual short-term load forecasting is crucial for this precise control per customer. Recently, deep learning has been widely adopted to improve forecasting accuracy. However, training an individual deep network (local training) has overfitting issues due to data paucity per customer. Thus, a pooling scheme has been introduced to augment a training dataset by batching several customers’ data. Nevertheless, there is room for existing pooling approaches to further improve accuracy by considering distribution heterogeneity within a customer dataset. In addition, their static pool assignment only with a customer’s training dataset may cause accuracy degradation under concept drift in serving time. To overcome these, we propose a Temporal Data Pooling (TDP) that constructs data pools at the data sample level with a novel distribution inference method and theoretical analysis. It allows the most probable forecasting model to serve predictions while resolving data shortage issues in local training. The TDP outperforms the other six competing methods for point and probabilistic forecasting; it shows robust accuracy under concept drift. Moreover, it demonstrates superior accuracy for unseen customers without additional training, proving its scalability.

COAC: Cross-Layer Optimization of Accelerator Configurability for Efficient CNN Processing

To achieve high accuracy, convolutional neural networks (CNNs) are increasingly growing in complexity and diversity in layer types and topologies. This makes it very challenging to efficiently deploy such networks on custom processor architectures for resource-scarce edge devices. Existing mapping exploration frameworks enable searching for the optimal execution schedules or hardware mappings of individual network layers, by optimizing each layer’s spatial (dataflow parallelization) and temporal unrolling (TU, execution order). However, these tools fail to take into account the overhead of supporting different unrolling schemes within a common hardware architecture. Using a fixed unrolling scheme across all layers is also not ideal, as this misses significant opportunities for energy and latency savings from optimizing the mapping of diverse layer types. A balanced approach assesses the right amount of mapping flexibility needed across target neural networks, while taking into account the overhead to support multiple unrollings. This article, therefore, presents cross-layer optimization of accelerator configurability (COAC), a cross-layer design space exploration and mapping framework to optimize the flexibility of neural processing architectures by balancing configurability overhead against resulting energy and latency savings for end-to-end inference. COAC does not only provide a systematical analysis of the architectural overhead in function of the supported spatial unrollings (SUs), but also builds an automated flow to find the best unrolling combination(s) for efficient end-to-end inference with limited hardware overhead. Results demonstrate that architectures with carefully optimized flexibility can achieve up to 38% energy-delay-product (EDP) savings for a set of six neural networks at the expense of a relative area increase of 9.5%.

Blockchain Business Model and Security

Abstract: Blockchain has become the most mentioned or a trending topic on the internet. Today, it is used for many purposes such as smart contracts, keeping records of financial data for banks, data management for government agencies or servicesbased companies, etc. Blockchains are phlebotomizing money. Now a day there is a plethora of different networks or tokens available. Each and every token and network has its own pros and cons. You can use this cryptocurrency to purchase goods or services on the platform or transfer it to various exchanges. But the point is how this token or cryptocurrency generates revenue from providing different services. How are individual networks able to be upheld value in the blockchain market? This research aims to know what are the different ways through which a blockchain company makes money and what things are important to run a profitable blockchain company and also researches what kind of vulnerability and problems occur in the blockchain. Blockchain networks are not immune to cyberattacks and fraud. Through this, we hope to increase the transparency of the Blockchain Business model to enhance the public's trust in blockchain networks and help to make a decision on which blockchain networks are profitable and good for future investments.

Network Analysis Using the Established Database (K-herb Network) on Herbal Medicines Used in Clinical Research on Heart Failure

Objectives: Heart failure is a chronic disease with increasing prevalence rates despite advancements in medical technology. Korean medicine utilizes herbal prescriptions to treat heart failure, but little is known about the specific herbal medicines comprising the network of herbal prescriptions for heart failure. This study proposes a novel methodology that can efficiently develop prescriptions and facilitate experimental research on heart failure by utilizing existing databases.Methods: Herbal medicine prescriptions for heart failure were identified through a PubMed search and compiled into a Google Sheet database. NetMiner 4 was used for network analysis, and the individual networks were classified according to the herbal medicine classification system to identify trends. K-HERB NETWORK was utilized to derive related prescriptions.Results: Network analysis of heart failure prescriptions and herbal medicines using NetMiner 4 produced 16 individual networks. &lt;i&gt;Uhwangcheongsim-won&lt;/i&gt; (牛黃淸心元), &lt;i&gt;Gamiondam-tang&lt;/i&gt; (加味溫膽湯), &lt;i&gt;Bangpungtongseong-san&lt;/i&gt; (防風通聖散), and &lt;i&gt;Bunsimgi-eum&lt;/i&gt; (分心氣飮) were identified as prescriptions with high similarity in the entire network. A total of 16 individual networks utilized K-HERB NETWORK to present prescriptions that were most similar to existing prescriptions. The results provide 1) an indication of existing prescriptions with potential for use to treat heart failure and 2) a basis for developing new prescriptions for heart failure treatment.Conclusion: The proposed methodology presents an efficient approach to developing new heart failure prescriptions and facilitating experimental research. This study highlights the potential of network pharmacology methodology and its possible applications in other diseases. Further studies on network pharmacology methodology are recommended.

Physical activity is associated with behavioral and neural changes across the lifespan

Physical activity is known to positively impact brain structure and function, but its effects on resting-state functional connectivity (rsFC) and its relationship with complex tasks as a function of age remain unclear. Here, we address these issues in a large population-based sample (N = 540) from the Cambridge Centre for Ageing and Neuroscience (Cam-CAN) repository. We relate levels of physical activity to rsFC patterns in magnetoencephalographic (MEG) and functional magnetic resonance imaging (fMRI) data, and to measures of executive function and visuomotor adaptation, across the lifespan. We show that higher self-reported daily physical activity is associated with lower alpha-band (8–12 Hz) global coherence, indicating weaker synchrony of neural oscillations in this band. Physical activity affected between-network connectivity of resting-state functional networks, although its effects on individual networks did not survive correction for multiple comparisons. Furthermore, our results indicate that greater engagement in day-to-day physical activity is associated with better visuomotor adaptation, across the lifespan. Overall, our findings indicate that rsFC metrics indexed by MEG and fMRI are sensitive indicators of the brain’s response to physical activity, and that a physically active lifestyle affects multiple aspects of neural function across the lifespan.

Platform Rhetoric and Fan Labour as the Building Blocks of &lt;em&gt;LEGO Ideas&lt;/em&gt;

Platform Rhetoric and Fan Labour as the Building Blocks of &lt;em&gt;LEGO Ideas&lt;/em&gt;