Adaptive Architecture Research Articles

MMIR: an open-source software for the registration of multimodal histological images

BackgroundMultimodal histology image registration is a process that transforms into a common coordinate system two or more images obtained from different microscopy modalities. The combination of information from various modalities can contribute to a comprehensive understanding of tissue specimens, aiding in more accurate diagnoses, and improved research insights. Multimodal image registration in histology samples presents a significant challenge due to the inherent differences in characteristics and the need for tailored optimization algorithms for each modality.ResultsWe developed MMIR a cloud-based system for multimodal histological image registration, which consists of three main modules: a project manager, an algorithm manager, and an image visualization system.ConclusionOur software solution aims to simplify image registration tasks with a user-friendly approach. It facilitates effective algorithm management, responsive web interfaces, supports multi-resolution images, and facilitates batch image registration. Moreover, its adaptable architecture allows for the integration of custom algorithms, ensuring that it aligns with the specific requirements of each modality combination. Beyond image registration, our software enables the conversion of segmented annotations from one modality to another.

BERMAD: batch effect removal for single-cell RNA-seq data using a multi-layer adaptation autoencoder with dual-channel framework.

Removal of batch effect between multiple datasets from different experimental platforms has become an urgent problem, since single-cell RNA sequencing (scRNA-seq) techniques developed rapidly. Although there have been some methods for this problem, most of them still face the challenge of under-correction or over-correction. Specifically, handling batch effect in highly nonlinear scRNA-seq data requires a more powerful model to address under-correction. In the meantime, some previous methods focus too much on removing difference between batches, which may disturb the biological signal heterogeneity of datasets generated from different experiments, thereby leading to over-correction. In this article, we propose a novel multi-layer adaptation autoencoder with dual-channel framework to address the under-correction and over-correction problems in batch effect removal, which is called BERMAD and can achieve better results of scRNA-seq data integration and joint analysis. First, we design a multi-layer adaptation architecture to model distribution difference between batches from different feature granularities. The distribution matching on various layers of autoencoder with different feature dimensions can result in more accurate batch correction outcome. Second, we propose a dual-channel framework, where the deep autoencoder processing each single dataset is independently trained. Hence, the heterogeneous information that is not shared between different batches can be retained more completely, which can alleviate over-correction. Comprehensive experiments on multiple scRNA-seq datasets demonstrate the effectiveness and superiority of our method over the state-of-the-art methods. The code implemented in Python and the data used for experiments have been released on GitHub (https://github.com/zhanglabNKU/BERMAD) and Zenodo (https://zenodo.org/records/10695073) with detailed instructions.

End-to-end deep reinforcement learning for control of an autonomous underwater robot with an undulating propulsor

This paper focuses on the development and implementation of control algorithms for positioning an Autonomous Underwater Vehicle (AUV) with an undulating propulsor, using reinforcement learning methods. It provides an analysis and overview of works incorporating reinforcement learning methods such as Actor-only, Critic-only, and Actor-Critic. The paper primarily focuses on the Deep Deterministic Policy Gradient method and its implementation using deep neural networks to train the Actor-Critic agent. In the agent's architecture, a replay buffer and target neural networks were utilized to address the data correlation issue that induces training instability. An adaptive ar-chitecture was proposed for training the agent to force the robot to move from the initial point to any target point. Additionally, a random target point generator was incorporated at the training stage so as not to retrain the agent when the target points change. The training objective is to optimize the actor's policy by optimizing the critic and maximizing the reward function. Reward function is determined as the distance from the robot's center of mass to the target points. Consequently, the reward received by the agent increases when the robot gets closer to the target point and becomes maximal when the target point is reached with an acceptable error.

Exploring the connection between adaptive architecture and artificial intelligence

Interactive adaptive architecture refers to the design of systems that can adapt to changes in their operating environment through interaction with users or other agents in the system. Artificial intelligence, on the other hand, refers to the ability of computer systems to perform tasks that would otherwise require human intelligence. Interactive adaptive architecture and artificial intelligence are two interconnected concepts that have evolved with the development of cybernetics. The development of interactive architecture based on the use of artificial intelligence will not mean the disappearance of architecture, as we know it, but rather a change in the way it is perceived. In the future, architecture will be participatory, with an active role of inhabitants in shaping and transforming architectural habitats, but there will always be a need for a moderator, the architect, to design and set the rules of the game for humans.

CAT: Continual Adapter Tuning for aspect sentiment classification

Humans can continually acquire, improve, and transfer knowledge throughout their lifespan so that they can accurately identify sentiment polarities of the data attributed to different domains. However, the continual learning of incrementally available aspect sentiment classification (ASC) tasks from different domains with non-stationary data distributions remains a long-standing challenge for learning-based models due to the catastrophic forgetting problem, which is the tendency to completely and abruptly forget previously learned knowledge upon learning new knowledge. In this work, we present Continual Adapter Tuning (CAT), a parameter-efficient framework that not only avoids catastrophic forgetting but also enables knowledge transfer from learned ASC tasks to new ASC tasks. To avoid catastrophic forgetting, we only learn and store a task-specific adapter for each ASC task while freezing the backbone pre-trained model. To promote new task learning, we propose a continual adapter initialization technique to transfer knowledge from preceding tasks. Besides, we also develop a novel label-aware contrastive learning to simultaneously learn the features of input samples and the parameters of classifiers in the same space so that we can efficiently classify a sample with the help of label semantics. To eliminate the need for task IDs in testing, we propose a simple yet efficient majority sentiment polarity voting strategy to obtain final sentiment polarities according to the polarities predicted by all reasoning paths in the adapter architecture. Experimental results show the high effectiveness of our CAT by achieving new state-of-the-art performance.

Conditional generative adversarial networks for groundwater contamination characterization and source identification

Accurate characterization of sources and spatial distribution of contamination remains crucial in groundwater survey and remediation efforts. Nevertheless, conventional geo-statistical methods, like kriging, usually produce overly smooth concentration fields that violate the transport physics. Furthermore, the identification of contaminant sources based on kriging methods may be misleading. As a deep learning approach, Generative adversarial networks (GANs) is being increasingly employed to extract patterns and insights from intricate geospatial data. The conditional variant of GANs (CGANs), can reconstruct complete spatial data from sparse point observations. In this study, we propose a deep learning framework with CGANs for the purpose of groundwater contamination characterization and source identification. In ensure compliance with the principles of transport physics, we use a numerical transport model as the underlying prior model for training the CGANs. Through the adaptation of network architecture, the generator transforms low-dimensional random variables and observation data into concentration images, while the discriminator is responsible for extracting source information from these images. The developed framework is tested through two numerical cases with channelized random hydraulic conductivity fields, which pose a challenge due to the complexity of the concentration fields. The results demonstrate that the proposed method can give accurate results for both contaminant characterization and source identification in groundwater. CGANs excel in reconstructing concentration fields compared to traditional interpolation methods such as Kriging and inverse distance weighting (IDW). Additionally, CGANs offer the capability to estimate uncertainty and exhibit a certain level of error robustness.

Antiviral fibrils of self-assembled peptides with tunable compositions

The lasting threat of viral pandemics necessitates the development of tailorable first-response antivirals with specific but adaptive architectures for treatment of novel viral infections. Here, such an antiviral platform has been developed based on a mixture of hetero-peptides self-assembled into functionalized β-sheets capable of specific multivalent binding to viral protein complexes. One domain of each hetero-peptide is designed to specifically bind to certain viral proteins, while another domain self-assembles into fibrils with epitope binding characteristics determined by the types of peptides and their molar fractions. The self-assembled fibrils maintain enhanced binding to viral protein complexes and retain high resilience to viral mutations. This method is experimentally and computationally tested using short peptides that specifically bind to Spike proteins of SARS-CoV-2. This platform is efficacious, inexpensive, and stable with excellent tolerability.

СТАЛИЙ РОЗВИТОК ТА АРХІТЕКТУРА: ЕКОНОМІЧНА ДОЦІЛЬНІСТЬ ВИКОРИСТАННЯ ДУАЛІСТИЧНОЇ КОНСТРУКЦІЇ НА ПРАКТИЦІ

The characteristics, principles, and directions of sustainable development in architecture have been identified. The economic feasibility of implementing sustainable development in architecture has been substantiated. The Fourth Industrial Revolution has been considered in the context of sustainable development. The authors suggest utilising the latest technologies and innovations to enhance production efficiency, reduce emissions through the implementation of clean technologies, and utilise resources rationally. By combining the opportunities presented by the Fourth Industrial Revolution with sustainable development approaches, it is possible to achieve a balance between economic, social, and environmental aspects of development. Sustainable architecture aims to minimise negative environmental impacts and improve building performance by consuming fewer resources, reducing waste, and creating a functional and productive environment. The text discusses the importance of sustainable principles in architecture, including adaptability of architecture, recycling of resources, use of sustainable and renewable materials, and energy efficiency. The economic feasibility of applying sustainable principles in design has also been considered and justified. Additionally, the text provides examples of innovative architectural designs. This is a crucial step towards implementing sustainable building practices. The adaptability of architecture enhances space efficiency by enabling areas to serve various functions based on user requirements. Resource recycling minimises waste and the expense of extracting new materials. The use of sustainable and renewable materials aids in reducing the ecological impact of construction. Energy efficiency helps to decrease energy consumption and GHG emissions. Further research and innovative architectural designs can accelerate the development of sustainable construction and promote its wider adoption in the future. The authors suggest that taking into account the costs of building, maintaining, and operating buildings, as well as a balanced cost-benefit ratio, can significantly influence decision-making in the design and construction process. The authors argue that implementing environmental solutions in architecture may raise the initial project cost. However, in the long run, it can lead to significant economic benefits by reducing energy and maintenance costs, as well as increasing the overall value of the building.

Cultural Influence on Architectural Evolution in Nigeria: A Case Study of Igala Indigenous Homestead

Architecture and culture are two concepts that represent the identification of any society. Architecture is the manifestation and expression of culture. So, Architecture and Culture are interdependent and inseparable. Research has identified culture as the main factor in the evolution of the architecture of a people and it plays an important role in the environmental and social development and sustainability of a place. The culture of any society is based on their beliefs, history, civilization, climate and customs, whilst architecture symbolizes a society’s culture that has evolved. This research explores the intricate relationship between culture and architecture in the evolution of traditional architecture in Nigeria, with a specific focus on the evolution of the indigenous homestead of the Igala people of North Central Nigeria. Employing a narrative approach, the study utilized historical research, field surveys, and interviews to explore how cultural beliefs, norms, practices, and socio-cultural factors have profoundly influenced the design and planning of Igala traditional dwellings. By investigating the interplay between indigenous knowledge systems, community practices, and material availability, the research uncovers the unique architectural features that have emerged from the rich tapestry of Igala culture to include the long straight entrance path (ojikpologu), the circular curvilinear homestead and buildings, courtyard (anuku/okolo), reception house (atakpa), and food storage barn (aka). Others include shrine (achękwu), grinding hut (ǫdǫ-okuta), animals’ pen (unyi-amęñwu-ǫrę), and security fence (ϙgba), all deeply rooted in Igala culture. This research provides valuable insights into the adaptability of traditional architecture to culture. It further offers essential knowledge for architects, preservationists, and policymakers, guiding their efforts in heritage preservation, sustainable architectural design, and urban planning within multicultural contexts.

Genetic architecture of ecological divergence between Oryza rufipogon and Oryza nivara.

Ecological divergence due to habitat difference plays a prominent role in the formation of new species, but the genetic architecture during ecological speciation and the mechanism underlying phenotypic divergence remain less understood. Two wild ancestors of rice (Oryza rufipogon and Oryza nivara) are a progenitor-derivative species pair with ecological divergence and provide a unique system for studying ecological adaptation/speciation. Here, we constructed a high-resolution linkage map and conducted a quantitative trait locus (QTL) analysis of 19 phenotypic traits using an F2 population generated from a cross between the two Oryza species. We identified 113 QTLs associated with interspecific divergence of 16 quantitative traits, with effect sizes ranging from 1.61% to 34.1% in terms of the percentage of variation explained (PVE). The distribution of effect sizes of QTLs followed a negative exponential, suggesting that a few genes of large effect and many genes of small effect were responsible for the phenotypic divergence. We observed 18 clusters of QTLs (QTL hotspots) on 11 chromosomes, significantly more than that expected by chance, demonstrating the importance of coinheritance of loci/genes in ecological adaptation/speciation. Analysis of effect direction and v-test statistics revealed that interspecific differentiation of most traits was driven by divergent natural selection, supporting the argument that ecological adaptation/speciation would proceed rapidly under coordinated selection on multiple traits. Our findings provide new insights into the understanding of genetic architecture of ecological adaptation and speciation in plants and help effective manipulation of specific genes or gene cluster in rice breeding.

Genomewide architecture of adaptation in experimentally evolved Drosophila characterized by widespread pleiotropy

Genomewide architecture of adaptation in experimentally evolved Drosophila characterized by widespread pleiotropy

The contribution of the Accessibility Program: include for the insertion of people with disabilities in higher education

This article aims to highlight the contribution of the Higher Education Accessibility Program – Include Program, which is conceived as an expression of accessibility policies for people with disabilities in the academic environment. The methodology was carried out through documentary research – analysis of notices and bibliography – on the portal of the Coordination for the Improvement of Higher Education Personnel (CAPES), covering the period from 2001 to 2022. It was understood from the documentary analysis that two actions stand out and strengthen the insertion of these people: architectural adaptation and implementation and/or strengthening of accessibility centers. From the bibliographical research, it was learned that inclusion actions are strengthened in the academic environment with the significant performance of Accessibility Centers, which mediate the relationship between academics with disabilities and the scientific community. This mediation takes place through guidance regarding the need to change traditional methodology, issues of pedagogical accessibility, different methodologies, among others. Therefore, it is urgent that the entire academic community becomes aware of its inclusive role in Society.

Enhancing well-being in modern education: A comprehensive eHealth proposal for managing stress and anxiety based on machine learning

In the evolving education landscape, institutions are transitioning to a hybrid model encompassing physical and virtual classes. In this scenario, as working hours and exhausting routines increase, individuals accumulate psychological and physical challenges over time. This change requires students and educators to adjust to changes in teaching and learning routines, leading to moments of stress and anxiety. Cultivating self-awareness about unhealthy habits emerges as a pragmatic strategy to address this issue. In this work, a comprehensive eHealth proposal is developed to provide individuals with pertinent health information, assisting educational agents (students and teachers) in identifying and managing these stressful instances during their daily activities. This proposal acquires, processes, and disseminates vital sign data using IoT devices, supported by a Fog Computing architecture for scalability and adaptability. IoT plays a pivotal role in this eHealth proposal, facilitating the seamless collection and transmission of real-time data from various devices. Connected wearables and sensors enable the continuous monitoring of vital signs, enhancing the accuracy and responsiveness of the system. Heart rate data from educational agents were collected for implementation, allowing the evaluation of system performance. Machine learning models were leveraged to discern behavioral profiles and predict possible irregularities in vital signs. The results confirm the system’s ability to perform its intended functions, giving users quick and accurate insights into their evolving behavioral patterns. Integrating these research perspectives underscores the importance of adaptive systems in navigating the challenges of modern education environments. The incorporation of IoT technology not only enhances data collection but also opens avenues for real-time interventions and personalized feedback, ultimately contributing to a more proactive approach in addressing health-related concerns within the educational context.

Fuzzy Logic Systems for Healthcare Applications

Artificial Intelligence Technologies (AITs) have found application in several domains, including the area of medicine. Within this context, AITs have been leveraged for purposes such as illness diagnosis and treatment, patient monitoring, and risk evaluation. By using Artificial Intelligence Technologies (AITs), it becomes feasible to create systems that facilitate the development of intelligent models for predicting not only patients' response to therapy but also the risk of illness. Due to the intricate and uncertain nature of these domains, a multitude of scholars have developed AITs, including genetic algorithms, artificial immune systems, Artificial Neural Networks (ANN), and fuzzy logic. The integration of Fuzzy Logic Systems and ANN allows the construction of intelligent and flexible systems. ANN gain novel information by changing the connections among its distinct layers. Fuzzy logic inference frameworks provide a computational model that is grounded on fuzzy set rules, theory, and fuzzy reasoning. The amalgamation of many adaptive architectures gives rise to a "Neuro-Fuzzy" system. This research paper examines fuzzy network topologies, exploring their possible applications in the medical field. Researchers have recognized that this convergence has promise for the discovery of medical patterns.

Identification of difficult laryngoscopy using an optimized hybrid architecture

BackgroundIdentification of difficult laryngoscopy is a frequent demand in cervical spondylosis clinical surgery. This work aims to develop a hybrid architecture for identifying difficult laryngoscopy based on new indexes.MethodsInitially, two new indexes for identifying difficult laryngoscopy are proposed, and their efficacy for predicting difficult laryngoscopy is compared to that of two conventional indexes. Second, a hybrid adaptive architecture with convolutional layers, spatial extraction, and a vision transformer is proposed for predicting difficult laryngoscopy. The proposed adaptive hybrid architecture is then optimized by determining the optimal location for extracting spatial information.ResultsThe test accuracy of four indexes using simple model is 0.8320. The test accuracy of optimized hybrid architecture using four indexes is 0.8482.ConclusionThe newly proposed two indexes, the angle between the lower margins of the second and sixth cervical spines and the vertical direction, are validated to be effective for recognizing difficult laryngoscopy. In addition, the optimized hybrid architecture employing four indexes demonstrates improved efficacy in detecting difficult laryngoscopy.Trial registrationEthics permission for this research was obtained from the Medical Scientific Research Ethics Committee of Peking University Third Hospital (IRB00006761-2015021) on 30 March 2015. A well-informed agreement has been received from all participants. Patients were enrolled in this research at the Chinese Clinical Trial Registry (http://www.chictr.org.cn, identifier: ChiCTR-ROC-16008598) on 6 June 2016.

Partitioning surface wave propagation on reconfigurable porous plane

This paper introduces a novel reconfigurable technique for partitioning the propagation of surface waves by utilizing a T-shaped structure and pathways established through the introduction of fluid metal or metal pins into evenly spaced cylindrical cavities within a porous surface wave platform. Notably, the co-printing of metal and dielectric materials via 3D printing is employed, resulting in an expedited fabrication process. Extensive 3D electromagnetic simulations and experimental investigations validate the proposed approach’s efficacy in achieving surface wave division while minimizing interference. The study encompasses an exploration of diverse power distribution ratios achievable within the distributed surface waves. Critical physical parameters of the T-junction are comprehensively examined, including partition depth, junction geometry, output port symmetry, and asymmetry. Additionally, the research delves into the frequency-dependent behaviours of asymmetric T-junctions and pathways. These findings establish the groundwork for adaptable architectures, facilitating concurrent communication among multiple devices within a unified surface wave communication network. This innovation holds potential to enhance various applications through improved communication capabilities.

The Intelligent Built-Environment as Cyber-Physical System

This thesis presents an alternative approach to intelligence in the built-environment, departing from the two established yet divergent branches in the discourse: the Technical, centered around Information and Communication Technologies (ICTs), and represented by Ambient Intelligence (AmI) and Ambient Assisted Living (AAL); and the Architectural, centered around architectural / spatial experiences and considerations, and represented by Interactive Architecture and Adaptive Architecture. The promise of both AmI/AAL and IA/AA is constrained by rigid and increasingly outdated assumptions in their approaches—i.e., AmI’s / AAL’s approach to the built-environment, and IA’s / AA’s approach to ICTs. Moreover, it is impossible to combine them to yield a cohesive system due to disparity in their typical Technology Readiness Levels. That is, the sophistication of a system depends on that of its mutually complementing subsystems; and two or more subsystems may not mutually complement, sustain, and/or support one another if their levels of development do not correspond. Consequently, the presented alternative conceives the intelligent built-environment as a Cyber- Physical System. Under this approach, ICTs and Architectural considerations in conjunction instantiate intelligence fundamentally. The presented approach’s promise is illustrated via its application to a constrained use-case focused on the elderly. Twelve proof-of-concept demonstrators are developed based on key parameters pertaining to Indoor Environmental Quality and Quality of Life. While each demonstrator is presented as a discrete proof-of-concept, all build on the same core System Architecture and technological ecosystem, and are intended to be viewed as a collection of systems and services expressed within a same hypothetical environment.

Can the Copenhagen Adduction Exercise Prevent Groin Injuries in Soccer Players? A Critically Appraised Topic.

Injuries that affect the groin region are among the most common in football players. To prevent this condition, studies have focused on strengthening the adductors, hip flexors, or abdominal muscles. Recent investigations have used an eccentric-biased exercise (Copenhagen Adduction Exercise [CAE]) that promotes functional and architectural adaptations in the muscle tissue, though its effect on injury risk reduction is unknown. Can the Copenhagen Adduction Exercise prevent groin injuries in soccer players? The literature was searched for studies investigating the potential groin injury risk reduction effect of the CAE. (1)Three studies met the inclusion criteria and were used for this appraisal; (2)one study observed a significantly lower injury rate ratio favoring the group that used the CAE program; and (3) 2 studies found similar or higher injury rates in the intervention groups, not supporting the inclusion of the CAE as a preventative tool. There is conflicting evidence that usage of the CAE is superior to not performing adductor strengthening exercises in mitigating the risk of sustaining groin injuries. Given the evidence supporting these findings, it is advisable to exercise caution when contemplating the incorporation of the CAE into training regimens aimed at preventing groin injuries. There is Grade B evidence to suggest that inclusion of the CAE may not be associated with reduced injury rates.

Streaming Readout and Data-Stream Processing With ERSAP

With the exponential growth in the volume and complexity of data generated at high-energy physics and nuclear physics research facilities, there is an imperative demand for innovative strategies to process this data in real or near-real-time. Given the surge in the requirement for high-performance computing, it becomes pivotal to reassess the adaptability of current data processing architectures in integrating new technologies and managing streaming data. This paper introduces the ERSAP framework, a modern solution that synergizes flow-based programming with the reactive actor model, paving the way for distributed, reactive, and high performance in data stream processing applications. Additionally, we unveil a novel algorithm focused on time-based clustering and event identification in data streams. The efficacy of this approach is further exemplified through the data-stream processing outcomes obtained from the recent beam tests of the EIC prototype calorimeter at DESY.

A Heterogeneous and Adaptive Architecture for Decision-tree-based ACL Engine on FPGA

A Heterogeneous and Adaptive Architecture for Decision-tree-based ACL Engine on FPGA