Pre-defined Model Research Articles

Situated AR Simulations of a Lantern Festival Using a Smartphone and LiDAR-Based 3D Models

A lantern festival was 3D-scanned to elucidate its unique complexity and cultural identity in terms of Intangible Cultural Heritage (ICH). Three augmented reality (AR) instancing scenarios were applied to the converted scanned data from an interaction to the entire site; a forward additive instancing and interactions with a pre-defined model layout. The novelty and contributions of this study are three-fold: documentation, development of an AR app for situated tasks, and AR verification. We presented ready-made and customized smartphone apps for AR verification to extend the model’s elaboration of different site contexts. Both were applied to assess their feasibility in the restructuring and management of the scene. The apps were implemented under a homogeneous and heterogeneous combination of contexts, originating from an as-built event description to a remote site as a sustainable cultural effort. A second reconstruction of screenshots in an AR loop process of interaction, reconstruction, and confirmation verification was also made to study the manipulated result in 3D prints.

Prototyping operational autonomy for Space Traffic Management

Current state of the art in Space Traffic Management (STM) relies on a handful of providers for surveillance and collision prediction, and manual coordination between operators. Neither is scalable to support the expected 10x increase in active spacecraft population in less than 10 years, nor does it support automated maneuver planning. We present a software prototype of an STM architecture based on open Application Programming Interfaces (APIs), and drawing on insights from NASA's architecture for low-altitude Unmanned Aerial System Traffic Management. The STM architecture is designed to provide structure to the interactions between spacecraft operators, various regulatory bodies, and STM service suppliers, while maintaining flexibility of these interactions and the ability for new market participants to enter easily. Autonomy will be an indispensable part of the proposed architecture in enabling efficient data sharing, coordination between STM participants and safe flight operations (e.g. select spacecraft maneuvers to prevent impending conjunctions between multiple spacecraft).The STM prototype is based on modern micro-service architecture adhering to OpenAPI standards and deployed in industry-standard virtualized containers, facilitating easy communication between different participants or services. The system architecture is designed to facilitate adding and replacing services with minimal disruption. We have implemented some example participant services (e.g. a space situational awareness/SSA provider, a conjunction assessment supplier/CAS, an automated maneuver advisor/AMA) within the prototype. Different services, with creative algorithms folded into them, can fulfill similar functional roles within the STM architecture by flexibly connecting to it using pre-defined APIs and data models, thereby lowering the barrier to entry of new players in the STM marketplace.We demonstrate the STM prototype on a multiple conjunction scenario with multiple maneuverable spacecraft, where an example CAS and AMA can recommend optimal maneuvers to the spacecraft operators, based on a pre-defined reward function. Such tools can intelligently search the space of potential collision avoidance maneuvers with varying parameters like lead time and propellant usage, to optimize a customized reward function, and be implemented as a scheduling service within the STM architecture. The case study shows an example of autonomous maneuver planning using the API-based framework. As satellite populations and predicted conjunctions increase, an STM architecture can facilitate seamless information exchange related to collision prediction and mitigation among various service applications on different platforms and servers. The availability of such an STM network also opens up new research topics on satellite maneuver planning, scheduling and negotiation across disjoint entities.

Obelisc: an identical-by-descent mapping tool based on SNP streak.

MotivationGenetic linkage analysis has made a huge contribution to the genetic mapping of Mendelian diseases. However, most previously available linkage analysis methods have limited applicability. Since parametric linkage analysis requires predefined model of inheritance with a fixed set of parameters, it is inapplicable without fully structured pedigree information. Furthermore, the analytical results are dependent on the specification of model parameters. While non-parametric linkage analysis can avoid these problems, the runs of homozygosity (ROH) mapping, a widely used non-parametric linkage analysis method, can only deal with recessive inheritance. The implementation of non-parametric linkage analyses capable of dealing with both dominant and recessive inheritance has been required.ResultsWe have developed the Obelisc (Observational linkage scan), a flexibly applicable user-friendly non-parametric linkage analysis tool, which also provides an intuitive visualization of the analytical results. Obelisc is based on the SNP streak approach, which does not require any predefined inheritance model with parameters. In contrast to the ROH mapping, the SNP streak approach is applicable to both dominant and recessive traits. To illustrate the performance of Obelisc, we generated a pseudo-pedigree from the publicly available BioBank Japan Project genome-wide genotype dataset (n > 180 000). By applying Obelisc to this pseudo-pedigree, we successfully identified the regions with inherited identical-by-descent haplotypes shared among the members of the pseudo-pedigree, which was validated by the population-based haplotype phasing approach.Availability and implementationObelisc is feely available at https://github.com/qsonehara/Obelisc as a python package with example datasets.Supplementary information Supplementary data are available at Bioinformatics online.

Local Human Impacts Disrupt Relationships Between Benthic Reef Assemblages and Environmental Predictors

Human activities are changing ecosystems at an unprecedented rate, yet large-scale studies into how local human impacts alter natural systems and interact with other aspects of global change are still lacking. Here we provide empirical evidence that local human impacts fundamentally alter relationships between ecological communities and environmental drivers. Using tropical coral reefs as a study system, we investigated the influence of contrasting levels of local human impact using a spatially extensive dataset spanning 62 outer reefs around inhabited Pacific islands. We tested how local human impacts (low versus high determined using a threshold of 25 people km−2 reef) affected benthic community (i) structure, and (ii) relationships with environmental predictors using pre-defined models and model selection tools. Data on reef depth, benthic assemblages, and herbivorous fish communities were collected from field surveys. Additional data on thermal stress, storm exposure, and market gravity (a function of human population size and reef accessibility) were extracted from public repositories. Findings revealed that reefs subject to high local human impact were characterised by relatively more turf algae (>10% higher mean absolute coverage) and lower live coral cover (9% less mean absolute coverage) than reefs subject to low local human impact, but had similar macroalgal cover and coral morphological composition. Models based on spatio-physical predictors were significantly more accurate in explaining the variation of benthic assemblages at sites with low (mean adjusted-R2 = 0.35) rather than high local human impact, where relationships became much weaker (mean adjusted-R2 = 0.10). Model selection procedures also identified a distinct shift in the relative importance of different herbivorous fish functional groups in explaining benthic communities depending on the local human impact level. These results demonstrate that local human impacts alter natural systems and indicate that projecting climate change impacts may be particularly challenging at reefs close to higher human populations, where dependency and pressure on ecosystem services are highest.

Low-frequency dominant electrical conductivity imaging of in vivo human brain using high-frequency conductivity at Larmor-frequency and spherical mean diffusivity without external injection current

Diffusion weighted imaging based on random Brownian motion of water molecules within a voxel provides information on the micro-structure of biological tissues through water molecule diffusivity. As the electrical conductivity is primarily determined by the concentration and mobility of ionic charge carriers, the macroscopic electrical conductivity of biological tissues is also related to the diffusion of electrical ions. This paper aims to investigate the low-frequency electrical conductivity by relying on a pre-defined biological model that separates the brain into the intracellular (restricted) and extracellular (hindered) compartments. The proposed method uses B1 mapping technique, which provides a high-frequency conductivity distribution at Larmor frequency, and the spherical mean technique, which directly estimates the microscopic tissue structure based on the water molecule diffusivity and neurite orientation distribution. The total extracellular ion concentration, which is separated from the high-frequency conductivity, is recovered using the estimated diffusivity parameters and volume fraction in each compartment. We propose a method to reconstruct the low-frequency dominant conductivity tensor by taking into consideration the extracted extracellular diffusion tensor map and the reconstructed electrical parameters. To demonstrate the reliability of the proposed method, we conducted two phantom experiments. The first one used a cylindrical acrylic cage filled with an agar in the background region and four anomalies for the effect of ion concentration on the electrical conductivity. The other experiment, in which the effect of ion mobility on the conductivity was verified, used cell-like materials with thin insulating membranes suspended in an electrolyte. Animal and human brain experiments were conducted to visualize the low-frequency dominant conductivity tensor images. The proposed method using a conventional MRI scanner can predict the internal current density map in the brain without directly injected external currents.

Measurement, analysis and prediction of amoxicillin oral dose stability from integrated molecular description approach and accelerated predictive stability (APS)

AbstractBackgroundStability of low amoxicillin oral dosage form (5 mg) used in reintroduction drug test was not fully documented. Furthermore, the impact of (1) salt moiety of amoxicillin and (2) amoxicillin – excipient interactions upon the antibiotic formulation stability during the storage was not characterized so that the estimation of the pharmaceutical expiration date from shelf-life was uncertain. Thus, the main goal of this study was to estimate the shelf-life of two formulations of amoxicillin, using a semi-predictive methodology.MethodsAmoxicillin sodium (AS) and amoxicillin trihydrate (ATH), corresponding to 5-mg amoxicillin, were compounded with microcrystalline cellulose (MCC) in oral hard capsules which were, then, submitted to four environmental conditions (25 °C / 60% or 80% relative humidity (RH); 40 °C / 75% RH; 60 °C / 5% RH) in climatic chambers for 45 and 84 days. Therefore, the characterization of amoxicillin-MCC mixture was assessed by attenuated total reflectance Fourier-transform infrared spectroscopy (ATR-FTIR) The profiles of amoxicillin content (determined by stability indicating chromatographic method) as a function of storage time, temperature and RH were fitted to pre-defined kinetic models performed by accelerated predictive stability (APS).ResultsATR-FTIR analysis of AS, ATH, MCC and bulk specimens stored in heated and humid atmosphere confirmed water sorption to cellulose described by a broad and unresolved 3600 to 3000 cm−1 band associated with (1) general intramolecular and intermolecular hydrogen bonding between water and hydroxyl groups of the cellulose, and with (2) free hydroxyl in cellulose. Moreover, a dramatic decrease of absorption at 1776 and 1687 cm−1 respectively characteristic of the β-lactam ring (νC=O) and amide group (νC=O), was revealed as a consequence of AS and ATH degradation caused by moisturization of bulk. Amoxicillin degradation was established by chromatographic analysis showing faster AS degradation than ATH throughout time exposure. The combined effects of temperature – RH were successfully modeled by APS, where AS and ATH showed accelerated (auto-catalysis degradation mechanism) and linear degradation, respectively. The faster AS degradation was assumed to be linked to lower hydrogen donor to hydrogen acceptor count ratio and polar surface than ATH, increasing the probability of AS hydrolysis by water adsorption to AS-MCC solid dispersion (e.g., by reduction of protective intramolecular hydrogen bonds between AS molecules). Furthermore, the compounding which involved a drastic homogenization of solids may have affected the crystalline degree of MCC with an increase of amorphous phase more sensitive to water adsorption.ConclusionsThe improvement of amoxicillin compounding for oral dose forms might be rationalized by taking into account the molecular descriptors of salt moiety and excipients, improved by the choice of an appropriate process of production, characterized from infrared vibrational spectroscopy and chromatographic analysis and finally predicted from accelerated stability assays.

Genetic programming in civil engineering: advent, applications and future trends

Over the past two decades, machine learning has been gaining significant attention for solving complex engineering problems. Genetic programing (GP) is an advanced framework that can be used for a variety of machine learning tasks. GP searches a program space instead of a data space without a need to pre-defined models. This method generates transparent solutions that can be easily deployed for practical civil engineering applications. GP is establishing itself as a robust intelligent technique to solve complicated civil engineering problems. This paper provides a review of the GP technique and its applications in the civil engineering arena over the last decade. We discuss the features of GP and its variants followed by their potential for solving various civil engineering problems. We finally envision the potential research avenues and emerging trends for the application of GP in civil engineering.

An input-controlled model system for identification of MHC bound peptides enabling laboratory comparisons of immunopeptidome experiments

Characterization of MHC-bound peptides by mass spectrometry (MS) is an essential technique for immunologic studies. Many efforts have been made to quantify the number of MHC-presented ligands by MS and to define the limits of detection of a specific MHC ligand. However, these experiments are often complex and comparisons across different laboratories are challenging. Therefore, we compared and orthogonally validated quantitation of peptide:MHC complexes by radioimmunoassay and flow cytometry using TCR mimic antibodies in three model systems to establish a method to control the experimental input of peptide MHC:complexes for MS analysis. Following isolation of MHC-bound peptides we identified and quantified an MHC ligand of interest with high correlation to the initial input. We found that the diversity of the presented ligandome, as well as the peptide sequence itself affected the detection of the target peptide. Furthermore, results were applicable from these model systems to unmodified cell lines with a tight correlation between HLA-A*02 complex input and the number of identified HLA-A*02 ligands. Overall, this framework provides an easily accessible experimental setup that offers the opportunity to control the peptide:MHC input and in this way compare immunopeptidome experiments not only within but also between laboratories, independent of their experimental approach. SignificanceAlthough immunopeptidomics is an essential tool for the characterization of MHCbound peptides on the cell surface, there are no easily applicable established protocols available that allow comparison of immunopeptidome experiments across laboratories. Here, we demonstrate that controlling the peptide:MHC input for immunopurification and LC-MS/MS experiments by flow cytometry in pre-defined model systems allows the generation of qualitative and quantitative data that can easily be compared between investigators, independently of their methods for MHC ligand isolation for MS.

Novel differential protection using model recognition and unsymmetrical vector reconstruction for the transmission line with wind farms connection

The large-scale integration of different kinds of wind farms (WFs) in a transmission grid can bring special adaptability issues to the traditional phasor-based line differential protection in both the transient and steady states. In this paper, a novel composite fault model based differential protection (CFMDP) is proposed. The CFMDP detects the fault status by evaluating the consistency between the sampled fault data and the pre-defined differential model. Since the full-frequency fault information in time-domain is directly used in the consistency calculation, the CFMDP method performs better transient behaviours than the conventional scheme. Besides, the compensated currents are used to build the differential model, which overcomes the 'over angle difference' caused steady-state adaptability issue in the phasor based differential relay. This steady-state protective issue mainly happens with the interphase faults in large-scale Type-IV wind turbine generators (WTGs) integrated power grid. The electro-magnetic transient (EMT) simulation results demonstrate the effectiveness and merits of the proposed CFMDP schemes.

HPAC model studies of selected Jack Rabbit II (JRII) releases and comparisons to test data

Chlorine transport and dispersion (T&D) modeling using the Hazard Prediction and Assessment Capability (HPAC) software suite was performed by the Defense Threat Reduction Agency (DTRA) Technical Reachback Division to generate concentration plumes as a function of time using the Jack Rabbit II (JRII) test series releases for Trials 1, 6, and 7. The results of the concentration time-histories and spatial location with respect to the JRII sensor arcs and pre-defined common modeling parameters are presented here in order to facilitate a model inter-comparison with other T&D modeling groups. Also, results of sensitivity studies and quantification of performance using different weather sets are presented here with the goal of identifying significant stressors in the context of an operationally relevant (real-world) framework.

Haery: A Hadoop Based Query System on Accumulative and High-Dimensional Data Model for Big Data

Column-oriented stores, known for their scalability and flexibility, are a common NoSQL database implementation and are increasingly used in big data management. In column-oriented stores, a “full-scan” query strategy is inefficient and the search space can be reduced if data is well partitioned or indexed; however, there is no pre-defined schema for building and maintaining partitions and indexes at lower cost. We leverage an accumulative and high-dimensional data model, a sophisticated linearization algorithm, and an efficient query algorithm, to solve the challenge of how a pre-defined and well-partitioned data model can be applied to flexible and time-varied key-value data. We adapt a high-dimensional array as the data model to partition the key-value data without additional storage and massive calculation; improve the Z-order linearization algorithm, which map multidimensional data to one dimension while preserving locality of the data points, for flexibility; efficiently build an expansion mechanism for the data model to support time-varied data. The result is Haery, a column-oriented store, based on a distributed file system and computing framework. In experiments, Haery is compared with Hive, HBase, Cassandra, MongoDB, PostgresXL, and HyperDex in terms of query performance. With results indicating Haery on average performs 4.57x, 4.23x, 3.55x, 1.79x, 1.82x, and 120.6x faster, respectively.

A new behavior-assisted semantic recognition method for smart home

In the recent Smart Home (SH) research work, intelligent service recommendation technique based on behavior recognition, it has been extensively preferred by researchers. However, most current research uses the Semantic recognition to construct the user's basic behavior model. This method is usually restricted by environmental factors, the way these models are built makes it impossible for them to dynamically match the services that might be provided in the user environment. To solve this problem, this paper proposes a Semantic behavior assistance (Semantic behavior assistance, SBA). By joining the semantic model on the intelligent gateway, building an SA model, in this way, a logical Internet networks for smart home is established. At the same time, a behavior assistant method based on SBA model is proposed, among them, the user environment-related entities, sensors, devices, and user-related knowledge models exist in the logical interconnection network of the SH system through the semantic model. In this paper, the data simulation experiment is carried out on the method. The experimental results show that the SBA model is better than the knowledge-based pre-defined model.

IoT meets BPM: a bidirectional communication architecture for IoT-aware process execution

Business processes are frequently executed within application systems that involve humans, computer systems as well as objects of the Internet of Things (IoT). Nevertheless, the usage of IoT technology for system supported process execution is still constrained by the absence of a common system architecture that manages the communication between both worlds. In this paper, we introduce an integrated approach for IoT-aware business process execution that exploits IoT for BPM by providing IoT data in a process-compatible way, providing an IoT data provenance framework, considering IoT data for interaction in a pre-defined process model, and providing wearable user interfaces with context-specific IoT data provision. The approach has been implemented on top of contemporary BPM modeling concepts and system technology. The introduced technique has evaluated extensively in different use cases in industry.

A consistent variable screening procedure with family-wise error control

ABSTRACTVariable selection in ultra-high-dimensional data analysis is of critical importance in diverse scientific fields. There are many methods proposed for this problem in the past few decades. However, no attention has been paid on the model size produced by screening procedures and the difficulty of determining the screening threshold for practical use, which is key to real applications. Specifically, the pre-defined model size in a screening procedure lacks theoretical support, and users have no idea about the true positives and false positives in the final results. To solve this problem, we developed a new consistent independence screening procedure called HHG-CIS. The screening threshold of HHG-CIS can be set to explicitly control the family-wise error rate, thus it is easy to use in practice. Furthermore, HHG-CIS is model-free, robust to outliers, and outperforms other methods in models with interactions. We showed these advantages of HHG-CIS through extensive simulations and a real data application.

Anthropometric clothing measurements from 3D body scans

We propose a full processing pipeline to acquire anthropometric measurements from 3D measurements. The first stage of our pipeline is a commercial point cloud scanner. In the second stage, a pre-defined body model is fitted to the captured point cloud. We have generated one male and one female model from the SMPL library. The fitting process is based on non-rigid iterative closest point algorithm that minimizes overall energy of point distance and local stiffness energy terms. In the third stage, we measure multiple circumference paths on the fitted model surface and use a nonlinear regressor to provide the final estimates of anthropometric measurements. We scanned 194 male and 181 female subjects, and the proposed pipeline provides mean absolute errors from 2.5 to 16.0 mm depending on the anthropometric measurement.

Assistive Technology for Patient with Congenital Foot Deformity

An assistive technology (AT) is any item, equipment or product used to increase, maintain or improve the functional capabilities of people with disabilities. The aim of this study was to design and develop two ATs in order to assist the needs of a male student from Universidad de Monterrey experiencing gait dysfunction and pain due to a congenital foot deformity preventing him from normal performance. These ATs included personalized orthopedic insoles to improve the participant’s posture and stability as well as two ankle-foot orthosis (AFO) to reduce the pain he presented. In order to design the orthopedic insoles, it was necessary to scan the participant’s feet; this was achieved using the photogrammetry technique. For the design of the AFOs, anthropometric measurements of the lower limbs were taken in order to modify a predefined 3D human model and obtain a digital model of the lower limbs. Both devices were manufactured using 3D printing technology. In order to analyze the participant’s progress and validate the effectiveness of the ATs, we developed a methodology for movement analysis based on the marker-less motion capture system Kinect 2. Data obtained was imported into Matlab in order to calculate lower limb joint angles and compare gait before and after using the ATs. Significant improvement was seen in the participant’s gait after two weeks of using the ATs. Moreover, we were able to demonstrate that the use of orthopedic insoles improved participant’s posture based on the correct alignment (180°) of the heel with the ankle. We believe these posture improvements could further impact on participant’s gait performance. Therefore, we expect a significant improvement on participant’s gait after constant use of both ATs developed.

A review on epileptic foci localization using resting-state functional magnetic resonance imaging.

Epilepsy is a brain syndrome caused by synchronous abnormal discharge of brain neurons. As an effective treatment for epilepsy, successful surgical resection requires accurate localization of epileptic foci to avoid iatrogenic disability. Previous studies have demonstrated the potential of restingstate functional magnetic resonance imaging (rs-fMRI) technique to localize epileptic foci though clinical applications of rs-fMRI are still at an early stage of development. fMRI data analysis approaches seek pre-defined regressors modeling contributions to the voxel time series, including the BOLD response following neuronal activation. In present study, localization strategies of epileptic foci in rs-fMRI technology were classified and summarized. To begin with, data-driven approaches attempting to determine the intrinsic structure of the data were discussed in detail. Then, as novel fMRI data analysis methods, deconvolution algorithms such as total activation (TA) and blind deconvolution were discussed, which were applied to explore the underlying activity-inducing signal of the BOLD signal. Lastly, effective connectivity approaches such as autocorrelation function method and Pearson correlation coefficient have also been proposed to identify the brain regions driving the generation of seizures within the epileptic network. In the future, fMRI technology can be used as a supplement of intraoperative subdural electrode method or combined with traditional epileptic focus localization technologies, which is one of the most attractive aspect in clinic. It may also play an important role in providing diagnostic information for epilepsy patients.

Novel data-driven, equation-free method captures spatio-temporal patterns of neurodegeneration in Parkinson's disease: Application of dynamic mode decomposition to PET.

Most neurodegenerative disorders are characterized by progressive loss of neurons throughout the course of disease in the form of specific spatio-temporal patterns. To capture and quantify these coherent patterns across both space and time, traditionally one would either fit a pre-defined model with spatial and temporal parameters or apply analysis in the spatial and temporal domains separately. In this work, we introduce and validate the use of dynamic mode decomposition (DMD), a data-driven multivariate approach, to extract coupled spatio-temporal patterns simultaneously. We apply the method to examine progressive dopaminergic degeneration in 41 patients with Parkinson's disease (PD) using [11C](±)dihydrotetrabenazine (DTBZ) Positron Emission Tomography (PET). DMD decomposed the progressive dopaminergic changes in the putamen into two orthogonal temporal progression curves associated with distinct spatial patterns: 1) an anterior-posterior gradient, the expression of which decreased gradually with disease progression with a higher initial expression in the less affected side; 2) a dorsal-ventral gradient in the less affected side, which was present in early disease stage only. In the caudate, we found a head-tail gradient analogous to the anterior-posterior gradient seen in the putamen; as in the putamen, the expression of this gradient decreased gradually with disease progression with higher expression in the less affected side. Our results with DTBZ PET data show the applicability and relevance of the proposed method for extracting spatio-temporal patterns of neurotransmitter changes due to neurodegeneration. The method is able to decompose known PD-induced dopaminergic denervation into orthogonal (and thus loosely independent) temporal curves, which may be able to reflect and separate either different mechanisms underlying disease progression and disease initiation, or differential involvement of striatal sub-regions at different disease stages, in a completely data driven way. It is expected that this method can be easily extended to other PET tracers and neurodegenerative disorders and may help to elucidate disease mechanisms in more details compared to traditional approaches.

Big Data and Official Statistics

Big data is a component of the Fourth Industrial Revolution. The deep penetration of digital technology has turned data into an essential component of the production process. Data are automatically generated by machines during the course of operation and during interactions with humans. This paper describes the concept and composition of big data. Most of the big data are unstructured and include text, audio-video files, images, emails, log files, etc. Statisticians are more interested in structured data presented in a pre-defined database model. Big data offer new sources and opportunities that cannot be discounted. However, the use of big data requires proper assessment in terms of quality dimensions such as accuracy, comparability and methodological soundness. Against the backdrop of arguments regarding big data, some users view big data as a replacement of official statistics. Such a conclusion is premature for at least two reasons: first, only a small part of big data can be used for decision-making. Second, theory and practice prove that a small sample based on scientific methods can yield much more reliable and accurate estimates than the results obtained from the processing of large amounts of unstructured data. The paper assesses the possibility of using big data for Sustainable Development Goals (SDG) monitoring, which is a nationally owned process, and NSOs are accountable for the SDG data they report. If the data are derived from a big data source, irrespective of the level of technical sophistication used in data transformation, the reliability of such data might be questioned by the national institutions. The paper concludes that the reliability of data obtained from big data sources hinges on the quality of tools and methods applied to data transformation. Statisticians can play an important role in alerting society, decision-making bodies of the government and businesses about the reliability of information derived from the different sources.

Deep learning-based smart task assistance in wearable augmented reality

• Combining deep learning with wearable AR for innovative and user-centered task assistance. • Reconstructing 2.5D and 3D replicas without pre-defined models for effective 3D annotation and sharing. • Applying instance segmentation and 3D spatial mapping in wearable AR for more intuitive task guidance. • Conducting experimental analyses through realistic manufacturing tasks for proving usability and usefulness. Wearable augmented reality (AR) smart glasses have been utilized in various applications such as training, maintenance, and collaboration. However, most previous research on wearable AR technology did not effectively supported situation-aware task assistance because of AR marker-based static visualization and registration. In this study, a smart and user-centric task assistance method is proposed, which combines deep learning-based object detection and instance segmentation with wearable AR technology to provide more effective visual guidance with less cognitive load. In particular, instance segmentation using the Mask R-CNN and markerless AR are combined to overlay the 3D spatial mapping of an actual object onto its surrounding real environment. In addition, 3D spatial information with instance segmentation is used to provide 3D task guidance and navigation, which helps the user to more easily identify and understand physical objects while moving around in the physical environment. Furthermore, 2.5D or 3D replicas support the 3D annotation and collaboration between different workers without predefined 3D models. Therefore, the user can perform more realistic manufacturing tasks in dynamic environments. To verify the usability and usefulness of the proposed method, we performed quantitative and qualitative analyses by conducting two user studies: 1) matching a virtual object to a real object in a real environment, and 2) performing a realistic task, that is, the maintenance and inspection of a 3D printer. We also implemented several viable applications supporting task assistance using the proposed deep learning-based task assistance in wearable AR.