Information Input Research Articles

Deep learning-based three-dimensional crack damage detection method using point clouds without color information

Automated high-precision crack detection on building structures under poor lighting conditions poses a significant challenge for traditional image-based methods. Overcoming this challenge is crucial to enhance the practical applicability of structural health monitoring and rapid damage assessment, especially in post-disaster scenarios like earthquakes. To address this challenge, this paper presents a deep learning-based three-dimensional crack detection method that utilizes light detection and ranging (LiDAR) point cloud data. Our method is specifically designed to address crack detection without relying on color information input, resulting in high-precision and robust apparent damage detection. The key contribution of this paper is the NL-3DCrack model, which enables automated three-dimensional crack semantic segmentation. This model comprises a feature embedding module, an incomplete neighbor feature extraction module, a decoder, and morphological filtering. Notably, we introduce an innovative incomplete neighbor mechanism to effectively mitigate the impact of outliers. To validate the effectiveness of our proposed method, we establish two three-dimensional crack detection datasets, namely the Luding dataset and the terrestrial laser scanner dataset, which are based on earthquake disasters. Experimental results demonstrate that our method achieves remarkable performance, with an intersection-over-union of 39.62% and 51.33% on the respective test sets, surpassing existing point cloud-based semantic segmentation models. Ablation experiments further confirm the effectiveness of our approach. In summary, our method showcases exceptional crack detection performance on LiDAR data using only XYZI channels. With its high precision and reliable results, it offers significant utility in real-world applications, contributing to improved structural health monitoring and rapid damage assessment after disasters, particularly in post-earthquake scenarios.

Dynamic High‐Capacity Structural‐Color Encryption Via Inkjet Printing and Image Recognition

AbstractStimuli‐responsive structural‐color materials have received widespread attention in information encryption due to the significant color changes under different stimuli. However, the trade‐off between the capacity of information input, security level, cost, and large‐area manufacturing greatly limits the application of structural colors in encryption. Herein, dynamic high‐capacity and high‐resolution encryption are achieved by implementing printed total internal reflection (TIR) structural color and computer‐aided image recognition. The printed TIR microstructures are prepared with relative humidity (RH) responsive polymer, which form a heterogeneous wettability system, and can exhibit vibrant color variation with humidity. As the implemented RH is changed, the printed microstructures will expand or shrink precisely, enabling a full‐color modulation across the visible light range. With the color change, each structural‐color pixel can be specifically encoded, allowing for this to encrypt dynamic information within the same pattern at different RHs. Furthermore, This study can precisely integrates tremendous different pixels and easily prepare various encrypted patterns, which guarantee the high‐capacity information input in a low‐cost way. Moreover, through computer programming and algorithm reading, the structural‐color patterns can be decoded and decrypted in real‐time, thus offering great potential for further encryption, anti‐counterfeiting, multiplexing encoding, and data storage.

A Decentralized Multi-Agent Path Planning Approach Based on Imitation Learning and Selective Communication

Abstract Multi-agent path planning (MAPP) is crucial for large-scale mobile robot systems to work safely and properly in complex environments. Existing learning-based decentralized MAPP approaches allow each agent to gather information from nearby agents, leading to more efficient coordination among agents. However, these approaches often struggle with reasonably handling local information inputs for each agent, and their communication mechanisms between agents need to be further refined to treat those congested traffic scenarios effectively. To address these issues, we propose a decentralized MAPP approach based on imitation learning and selective communication. Our approach adopts an imitation learning architecture that enables agents to rapidly learn complex behaviors from expert planning experience. The information extraction layer is integrated with convolutional neural network (CNN) and gated recurrent unit (GRU) for capturing features from local field-of-view observations. A two-stage selective communication process based on graph attention neural network (GAT) is developed to reduce the required neighbor agents in inter-agent communication. In addition, an adaptive strategy switching mechanism utilizing local expert-planned paths is designed to support robots to escape from local traps. The effectiveness of our proposed approach is evaluated in simulated grid environments with varying map sizes, obstacle densities, and numbers of agents. Experimental results show that our approach outperforms other decentralized path planning methods in success rate while maintaining the lowest flowtime variation and communication frequency. Furthermore, our approach is computationally efficient and scalable, making it suitable for real-world applications.

A Parameter Assessment of Teaching Quality Indicators Based on Data Class Mining Fuzzy K-Mean Type Clustering

This paper proposes a data-based mining and hesitant fuzzy C-canopy-K mean clustering degree algorithm and uses it in the parameter assessment model of teaching quality indicators. Simulation and training are carried out through data class mining, and information input, followed by combining the hesitant fuzzy K-mean classification assessment method, which involves a hesitant fuzzy type evaluation system, a neural network identification and prediction system, and an application system for module identity verification. The simulation results show that the results of the six simulation conditions are consistent with the actual results, with only slight differences in some amplitudes, and a high degree of consistency in the overall trend, the change rule, and the average peak value. Through the prediction model processing in this paper, the teaching quality index parameter assessment has high accuracy and can reach more than 95.0%, in addition, the development of the law also fits very well. a, b, c, d four kinds of teaching quality parameter assessment of the average calculation of the assessment speed increased by 52.5%. In addition, the assessment test after the integrated design of the module shows that the system can effectively identify the four clustering identification processes that can be seen as excellent, good, medium, and poor; at the same time, the test data show that the system class effectively for teaching quality indicator parameter assessment.

Urban Infrastructure Innovation: The Contribution Ratio-Based Distance-to-Target Prioritization Method for Advanced Systems

As urban infrastructure increasingly benefits from adopting advanced systems, several approaches for prioritizing their utilization in urban infrastructure planning have been proposed. However, more research focusing on prioritizing advanced systems in urban infrastructure is required, and many existing prioritization methods rely on subjective inputs to determine the weights (i.e., relative importance) of evaluation criteria. This paper introduces a contribution ratio-based distance-to-target (CR-DTT) method for prioritizing advanced systems to assist urban infrastructure innovation. The novel prioritization method combines DTT normalization and ratio-based weights, and consists of three modules: (a) hierarchy table preparation; (b) input of advanced systems information; and (c) advanced systems prioritization. Each module involves multiple steps that generate outputs, which serve as inputs for the subsequent modules. We applied the prioritization method to a case study involving urban roads to demonstrate its effectiveness in real-world settings. The newly presented CR-DTT prioritization method contributes to the current knowledge in the area of developing prioritization methods by developmentally combining the DTT normalization and the ratio-based weights approaches into a practical data-driven prioritization method, which addresses the temporal aspect of the input parameter and provides objective insights into how best to prioritize advanced systems to assist transparent and credible decision-making in relation to investment in urban infrastructure innovation.

Recognition of Amino Acid Salts by Temperature-Dependent Allosteric Binding with Stereodynamic Urea Receptors.

Positive homotropic artificial allosteric systems are important for the regulation of cooperativity, selectivity and nonlinear amplification. Stereodynamic homotropic allosteric receptors can transmit and amplify induced chirality by the first ligand binding to axial chirality between two chromophores. We herein report stereodynamic allosteric urea receptors consisting of a rotational shaft as the axial chirality unit, terphenyl units as structural transmission sites and four urea units as binding sites. NMR titration experiments revealed that the receptor can bind two carboxylate guests in a positive homotropic allosteric manner attributed to the inactivation by intramolecular hydrogen-bonding between urea units within the receptor. In addition, the VT-CD spectra observed upon binding of the urea receptor with l- or D-amino acid salts in MeCN showed interesting temperature-dependent Cotton effects, based on the differences of the receptor shaft unit and the guest structure. The successful discrimination of hydrocarbon-based side chains of amino acid salts indicated that the input of chiral and steric information for the guest was amplified as outputs of the Cotton effect and the temperature-dependence of VT-CD spectra through cooperativity of positive allosteric binding.

Incorporating Entity Type-Aware and Word–Word Relation-Aware Attention in Generative Named Entity Recognition

Named entity recognition (NER) is a critical subtask in natural language processing. It is particularly valuable to gain a deeper understanding of entity boundaries and entity types when addressing the NER problem. Most previous sequential labeling models are task-specific, while recent years have witnessed the rise of generative models due to the advantage of tackling NER tasks in the encoder–decoder framework. Despite achieving promising performance, our pilot studies demonstrate that existing generative models are ineffective at detecting entity boundaries and estimating entity types. In this paper, a multiple attention framework is proposed which introduces the attention of entity-type embedding and word–word relation into the named entity recognition task. To improve the accuracy of entity-type mapping, we adopt an external knowledge base to calculate the prior entity-type distributions and then incorporate the information input to the model via the encoder’s self-attention. To enhance the contextual information, we take the entity types as part of the input. Our method obtains the other attention from the hidden states of entity types and utilizes it in self- and cross-attention mechanisms in the decoder. We transform the entity boundary information in the sequence into word–word relations and extract the corresponding embedding into the cross-attention mechanism. Through word–word relation information, the method can learn and understand more entity boundary information, thereby improving its entity recognition accuracy. We performed experiments on extensive NER benchmarks, including four flat and two long entity benchmarks. Our approach significantly improves or performs similarly to the best generative NER models. The experimental results demonstrate that our method can substantially enhance the capabilities of generative NER models.

Research on the Design of Online Course Teaching Interfaces and Knowledge Presentation Methods under Balanced Cognitive Load

This paper aims to explore the impact of high or low cognitive load on learners' learning effectiveness during the online learning process, as well as to achieve the goal of learners efficiently completing online learning tasks. Based on the theory of balanced cognitive load, the paper delves into the factors that may cause changes in learners' cognitive load within online courses. Through methods such as in-depth user interviews, negative experience points in the online learning process were collected and a detailed analysis of the corresponding types of cognitive load was conducted. The PAAS scale was utilized for the measurement and assessment of cognitive load, determining the extent of cognitive load at each negative experience point. In response to the issues identified, the paper proposes strategies to improve or eliminate learners' negative touchpoints from the following aspects: first, optimizing the hierarchical design of the instructional interface; second, implementing multi-channel information input; third, reinforcing the guidance of key information; and fourth, split-attention effect. These measures are intended to optimize the presentation of information on the instructional interface, reduce the cognitive burden on learners, and thereby enhance their cognitive validity, ensuring the efficiency of online learning.

데이터로 본 판소리의 性 인지 체계: 신재효本을 중심으로

This study analyzes the sexual cognitive system in pansori works using Shin Jae-hyo’s pansori texts as the research subject. To that end, I first extracted excerpts of sexual expressions from Shin Jae-hyo’s pansori texts such as Chunhyang-ga(NamChang), Simcheong-ga, Bak-Taryeong, Tobyol-ga, Jeokbyeok-ga, and Byeongangsoe-ga. Then, I input them on a sentence-by-sentence basis with metadata including the speaker (narrator), target of description, related characters, works, and excerpts. As a preliminary step for the quantitative analysis, I performed morphological analysis and preprocessing on the data entered in the text column. This made the language of the pansori texts suitable for digital analysis tools with the input of new information or stop words into the user dictionary.
 Three analyses were attempted using the data extracted from Shin Jae-hyo’s pansori texts. First, basic statistical analysis was conducted to visualize which work had the most occurrences of sexual expressions and to examine the frequency of speaker-description target pairs when sexual expressions were uttered and their targets. Second, Latent Dirichlet Allocation (LDA) topic modeling was performed to extract topics from a large collection of documents. Third, I used co-occurrence word analysis to reveal the semantic network of words constituting sexual expressions.
 The findings revealed that the sexual cognitive system in pansori works is based on “deviation and return.” That is, while enjoying pansori, one can realize that forbidden desires are, indeed, common desires and can satisfy sexual fantasies. However, such desires should not persist beyond exceptional circumstances and should ultimately be concealed and transformed to abide byto societal norms. Listening to and reading pansori involves repeatedly deviating from and returning to sexuality as a theme that persists throughout the whole story.

Chinese college students’ post-traumatic growth during the COVID-19: a grounded theory study

During the Corona Virus Disease 2019 (COVID-19) pandemic, both high levels of post-traumatic stress and positive personal changes, a phenomenon known as post-traumatic growth (PTG), were experienced by some people. What remains unknown are the constructs and facilitators of PTG during the COVID-19 pandemic. Therefore, this study uses a grounded theory methodology to elucidate the behavioral and cognitional dimensions of PTG in Chinese college students. Twenty-four individual semi-structured interviews were conducted with Chinese college students who had positive experiences during the epidemic. Transcripts were analyzed using initial and focused coding to identify themes that explained the PTG model. The coding results revealed that the Chinese college students’ PTG consisted of four core themes: reflections on the country and society, self-awareness changes, increased social behavior, and lifestyle modifications. Additionally, participants described external factors such as information input and patriotism, which were identified as influencing their PTG during the pandemic. Findings from this study extend prior theory and research by conceptualizing a novel theoretical model for PTG among college students during the pandemic and provide important implications for research and practice.

Adaptive Control for Virtual Synchronous Generator Parameters Based on Soft Actor Critic.

This paper introduces a model-free optimization method based on reinforcement learning (RL) aimed at resolving the issues of active power and frequency oscillations present in a traditional virtual synchronous generator (VSG). The RL agent utilizes the active power and frequency response of the VSG as state information inputs and generates actions to adjust the virtual inertia and damping coefficients for an optimal response. Distinctively, this study incorporates a setting-time term into the reward function design, alongside power and frequency deviations, to avoid prolonged system transients due to over-optimization. The soft actor critic (SAC) algorithm is utilized to determine the optimal strategy. SAC, being model-free with fast convergence, avoids policy overestimation bias, thus achieving superior convergence results. Finally, the proposed method is validated through MATLAB/Simulink simulation. Compared to other approaches, this method more effectively suppresses oscillations in active power and frequency and significantly reduces the setting time.

Field Independence and Dependence and High School English Reading Teaching

The English Curriculum Standards for General High Schools (2017 Edition) (hereinafter referred to as the "Standards") clearly stipulate that reading is one of the basic skills mastered by high school students, and it accounts for a relatively large proportion in the college entrance examination. At the high school level, students should develop the five basic skills of listening, speaking, reading, writing and translation, focusing on the cultivation of reading ability. Students should be able to read materials about biographies, fables, etc. With the help of dictionaries, students should be able to grasp the main facts and central ideas, and grasp the main logical clues of the material they read. Reading is a complex cognitive process. Reading is an important means of information input and a cognitive activity. Cognitive style is the preferred information processing method in individual cognitive activities, and field independence and field dependence cognitive style are the core of cognitive style. Through the discussion of field dependence and field independence cognitive style, this article analyses the differences between field independence and field dependence students’ learning behavior to explore the influence of students' cognitive style on teachers' reading teaching strategies and reading material choices, so as to improve students' reading ability.

Soybean polypeptide content detection based on fusion of spectral and image information

Soybean polypeptide is a protein hydrolysate. It has multiple benefits, such as anticancer, antioxidant, etc. Addressing the limitations of traditional methods, such as time-consuming procedures and high costs, this study proposed a novel approach for the rapid, nondestructive detection of soybean polypeptide content. This approach utilizes the fusion of spectral and image information. First, seven preprocessing methods were applied to the spectral data. The results indicated that the model accuracy was highest after applying the standard normal variate preprocessing method. Second, a combined method, consisting of synergy interval partial least squares and iteratively retains informative variables, was proposed to extract characteristic variables. A two-layer Stacking framework ensemble learning model was constructed to compare the detection effects between single spectral input and fusion of spectral and image information input. The results showed that the fusion input model achieved the best performance. The coefficient of determination (R 2) of prediction of the fusion input model reached 0.952, an improvement of 0.061 compared to single spectral input. And root mean square error of prediction (RMSEP) of the fusion input model reached 1.803 × 10−4, a reduction of 7.43 × 10−5 compared to single spectral input. These results indicate that the fusion of spectral and image information can effectively improve the model’s accuracy, offering valuable technical support for soybean breeding and quality detection.

BiVi-GAN: Bivariate Vibration GAN.

In the domain of prognosis and health management (PHM) for rotating machinery, the criticality of ensuring equipment reliability cannot be overstated. With developments in artificial intelligence (AI) and deep learning, there have been numerous attempts to use those methodologies in PHM. However, there are challenges to applying them in practice because they require huge amounts of data. This study explores a novel approach to augment vibration data-a primary component in traditional PHM methodologies-using a specialized generative model. Recognizing the limitations of deep learning models, which often fail to capture the intrinsic physical characteristics vital for vibration analysis, we introduce the bivariate vibration generative adversarial networks (BiVi-GAN) model. BiVi-GAN incorporates elements of a physics-informed neural network (PINN), emphasizing the specific vibration characteristics of rotating machinery. We integrate two types of physical information into our model: order analysis and cross-wavelet transform, which are crucial for dissecting the vibration characteristics of such machinery. Experimental findings show the effectiveness of our proposed model. With the incorporation of physics information (PI) input and PI loss, the BiVi-GAN showed a 70% performance improvement in terms of JS divergence compared with the baseline biwavelet-GAN model. This study maintains the potential and efficacy of complementary domain-specific insights with data-driven AI models for more robust and accurate outcomes in PHM.

Corticostriatal causality analysis in children and adolescents with attention-deficit/hyperactivity disorder.

The effective connectivity between the striatum and cerebral cortex has not been fully investigated in attention-deficit/hyperactivity disorder (ADHD). Our objective was to explore the interaction effects between diagnosis and age on disrupted corticostriatal effective connectivity and to represent the modulation function of altered connectivity pathways in children and adolescents with ADHD. We performed Granger causality analysis on 300 participants from a publicly available Attention-Deficit/Hyperactivity Disorder-200 dataset. By computing the correlation coefficients between causal connections between striatal subregions and other cortical regions, we estimated the striatal inflow and outflow connection to represent intermodulation mechanisms in corticostriatal pathways. Interactions between diagnosis and age were detected in the superior occipital gyrus within the visual network, medial prefrontal cortex, posterior cingulate gyrus, and inferior parietal lobule within the default mode network, which is positively correlated with hyperactivity/impulsivity severity in ADHD. Main effect of diagnosis exhibited a general higher cortico-striatal causal connectivity involving default mode network, frontoparietal network and somatomotor network in ADHD compared with comparisons. Results from high-order effective connectivity exhibited a disrupted information pathway involving the default mode-striatum-somatomotor-striatum-frontoparietal networks in ADHD. The interactions detected in the visual-striatum-default mode networks pathway appears to be related to the potential distraction caused by long-term abnormal information input from the retina in ADHD. Higher causal connectivity and weakened intermodulation may indicate the pathophysiological process that distractions lead to the impairment of motion planning function and the inhibition/control of this unplanned motion signals in ADHD.

Implementation and evaluation of a decentralized medical data exchange system based on the IOTA Tangle and approximate algorithms for optimal data placement


 
 
 
 
 
 
 
 © Voronin V., Rybalchenko A., Shmatko O., Kolomiitsev O., Tretiak V., Kliuchka Ya., 2023
 
 
 
 
 
 
 
 
 Facilitating the transfer of patient medical information across healthcare providers is crucial for ensuring high-quality care. However, this process encounters obstacles pertaining to privacy, security, and centralised control. This article outlines the creation of a decentralised system for exchanging medical data. The system utilises Distributed Ledger Technology, specifically the IOTA Tangle (open protocol (and network) for data and value transfer). This document outlines the structure and fundamental elements of a system designed for the secure and unalterable storage and transfer of medical records. The system employs masking and encryption methodologies to protect patient confidentiality while enabling healthcare practitioners to access complete information with patient authorization. The IOTA Tangle facilitates transactions without any fees and ensures the verification of data integrity through its Directed Acyclic Graph (DAG) topology. Simulation trials validate the system’s capability to securely communicate medical data on a large scale, while incurring lower resource costs than typical blockchain systems. The system demonstrates the feasibility of a decentralised and self-governing method for efficient and confidential sharing of medical data, utilising Distributed Ledger Technology (DLT). The secure data sharing platform facilitates the implementation of novel care and research frameworks, all the while upholding patient confidentiality and adhering to healthcare ethical standards. Also, in the development of billing Online Transaction Processing (OLTP) systems, which are designed for input, structured storage and processing of information in real time, the use of cloud technology is proposed. The problems that arise and their relevance to the solution of an integer linear programming problem with Boolean variables are shown. Approximate algorithms for optimal data placement and mathematical models for optimizing the structure of a distributed database of a cloud system are proposed, taking into account the limitations on the amount of node memory, the available costs of renting cloud resources, and the number of replicas of fragments of the distributed database.

Autonomous Driving System Driven by Artificial Intelligence Perception Fusion

Perception, as the information input module of the automatic driving system, determines the lower limit of the entire automatic driving system. Both autonomous driving perception and robot perception are constantly approaching the real physical world through digital methods, and this real physical world is only limited to the scope of human perception, such as lane lines, traffic lights, driving obstacles, and so on. The main premise of this process is that humans already know the categories or properties of the physical world, and only allow machines and systems to replicate human responses. Whether it is a pure visual route or a multi-source fusion route, the essence is the difference between the perceptual system schemes, one focusing on the vertical and the other on the horizontal. The pure vision solution represented by Tesla or the multi-source sensor fusion file represented by Waymo. In fact, the perception module of the automatic driving system usually has multiple sensors to achieve information redundancy and information complementarity through multiple dimensions, but there is the possibility of information conflict between different sensors. This paper aims at the advantages of perception-driven artificial intelligence to achieve breakthroughs in autonomous driving innovation, and analyzes how perception fusion drive is applied to the practical application of autonomous driving, so as to analyze the future development prospects of artificial intelligence.

Will Machines Innovate for and with Us - What Kind of Strategic Themes Could Belong to Innovation Automation?

The Internet economy and computer-aided innovation enable improvements in the quality and quantity of outcomes of innovation processes. Traditional “research pipes” are often too slow to fit with contemporary business logic. In this paper, we focus on the intersection of innovation and automation and the potential they create together. Innovation automation represents a next generation of automation that has structural implications. Automation in the innovation context is about maintaining the richness of creative innovation processes while also absorbing a greater amount of data, information, and knowledge inputs and producing more holistic outputs that meet customer needs better and are faster on the market. The paper builds a novel academic “playground” for the research on innovation automation as the efficient and effective use of co-creative intelligence—the fusion and mixture of artificial intelligence, human intelligence, and the intelligence of crowds. Covering the wide field of innovation automation requires various future research programs. The main focus areas in this paper are related to understanding innovation automation, enabling the way to new management of innovation and ecosystem development. We also propose relevant research themes for the future.

Estimating Winter Arctic Sea Ice Motion Based on Random Forest Models

Sea ice motion (SIM) plays a crucial role in setting the distribution of the ice cover in the Arctic. Limited by images’ spatial resolution and tracking algorithms, challenges exist in obtaining coastal sea ice motion (SIM) based on passive microwave satellite sensors. In this study, we developed a method based on random forest (RF) models to obtain Arctic SIM in winter by incorporating wind field and coastal geographic location information. These random forest models were trained using Synthetic Aperture Radar (SAR) SIM data. Our results show good consistency with SIM data retrieved from satellite imagery and buoy observations. With respect to the SAR data, compared with SIM estimated with RF model training using reanalysis surface wind, the results by additional coastal information input had a lower root mean square error (RMSE) and a higher correlation coefficient by 31% and 14% relative improvement, respectively. The latter SIM result also showed a better performance for magnitude, especially within 100 km of the coastline in the north of the Canadian Arctic Archipelago. In addition, the influence of coastline on SIM is quantified through variable importance calculation, at 22% and 28% importance of all RF variables for east and north SIM components, respectively. These results indicate the great potential of RF models for estimating SIM over the whole Arctic Ocean in winter.

A Lightweight Pre-Crash Occupant Injury Prediction Model Distills Knowledge From Its Post-Crash Counterpart.

Accurate occupant injury prediction in near-collision scenarios is vital in guiding intelligent vehicles to find the optimal collision condition with minimal injury risks. Existing studies focused on boosting prediction performance by introducing deep-learning models but encountered computational burdens due to the inherent high model complexity. To better balance these two traditionally contradictory factors, this study proposed a training method for pre-crash injury prediction models, namely, knowledge distillation (KD)-based training. This method was inspired by the idea of knowledge distillation, an emerging model compression method. Technically, we first trained a high-accuracy injury prediction model using informative post-crash sequence inputs (i.e., vehicle crash pulses) and a relatively complex network architecture as an experienced "teacher". Following this, a lightweight pre-crash injury prediction model ("student") learned both from the ground truth in output layers (i.e., conventional prediction loss) and its teacher in intermediate layers (i.e., distillation loss). In such a step-by-step teaching framework, the pre-crash model significantly improved the prediction accuracy of occupant's head abbreviated injury scale (AIS) (i.e., from 77.2% to 83.2%) without sacrificing computational efficiency. Multiple validation experiments proved the effectiveness of the proposed KD-based training framework. This study is expected to provide reference to balancing prediction accuracy and computational efficiency of pre-crash injury prediction models, promoting the further safety improvement of next-generation intelligent vehicles.