3D Virtual Reconstruction Research Articles

Histological assessment of the shared vascular wall in anomalous aortic origin of coronary arteries with an interarterial course

Abstract Background In anomalous aortic origin of a coronary artery (AAOCA) an intramural segment is present if the proximal AAOCA courses through the aortic wall. This is defined as a shared tunica media of the aorta and AAOCA in the ‘intramural septum’. This definition is based on autopsy studies of selected AAOCA patients who suffered sudden cardiac death and is difficult to apprehend on imaging modalities. Therefore, the aim was to identify the histological features and variants of the interarterial vascular wall in living adult patients, and correlate this to CT-angiography (CTA). Methods This prospective multicenter study included consecutive adult AAOCA patients who underwent surgical unroofing between 2021 to 2024. The excised interatrial vascular wall tissue was embedded, sectioned, and immunohistochemically stained for physiological components, e.g. smooth muscle cells and connective tissue, and pathological components such as fibrosis. Microscopic examination and quantification was performed by two independent observers and findings were correlated with CTA. Results Fifteen patients (mean age 42.9±14.0 years, 60% female) and 1 postmortem specimen were included. Fourteen (93%) had a right-AAOCA and 1 (7%) a left-AAOCA, with a mean intramural length of 8.4±4.1mm. Histopathologically, disorganized elastic lamellae with fragmentation, presence of fibrosis and loss of nuclei were observed. No vasa vasorum or nervi vasorum were identified. Post-sectioning virtual 3D reconstructions enhanced the spatial insight in correlation of histological sections with CTA. Conclusion This first study of the intramural septum in adults with AAOCA reveals the histological structure of the shared vascular wall between the aorta and AAOCA. Marked vascular wall abnormalities such as lack of vasa and nervi vasorum, fragmentation of elastic lamellae and fibrotic changes were observed. These findings support a propensity to local vascular wall pathology in patients with AAOCA. 3D reconstruction supported the translation of histological data to clinical imaging techniques.Central Figure

Poliface: A Multi-pose Synchronous Imaging System

To enhance the accuracy of face recognition technology in real-world scenarios, it is necessary to train deep learning models on datasets that contain a large number of labeled human face images under multiple poses, lighting, and accessory variations. In this paper, we introduce a novel acquisition system named the Poliface. This system can capture multiple high-resolution images simultaneously around the human head. We designed this system with a well-built aluminum structure, control electronic circuits, and high-performing in-house software. The results demonstrate the precise operation and exceptional stability of this system. Using this Poliface system, we have collected over 6 million photos, which can be used to train and evaluate facial recognition models, and exploited for three-dimensional (3D) virtual face reconstruction.

The virtual reconstruction of Cervara di Roma fortress: methods and tools for the dissemination of the past

Three-dimensional (3D) technologies play a crucial role in the philological reconstruction of archaeological contexts. The integration of technology and archaeological evidence has made it possible to develop 3D virtual reconstructions, providing valuable insights into the temporal evolution of archaeological sites. This innovation serves as a significant research and analytical resource, enabling the visualisation of possible scenarios related to archaeological contexts that have significantly changed over time. Moreover, understanding these changes, the past and its relevance to the current context could pose a challenge to the general public. This work aims to illustrate a scalable rendering system using open-source Web3D apps and platforms, allowing access to information, 3D models, and descriptions to enhance the experience of artworks.

Evaluation of anode support microstructure in solid oxide fuel cells using virtual 3D reconstruction: A simulation study

In this study, the impact of microstructural properties of the anode support layer (ASL) composed of catalyst, electrolyte and pore phases on the performance of solid oxide fuel cell (SOFC) is investigated. Synthetic SOFC anode microstructures, comprising two layers including the anode functional layer (AFL), where electrochemical reactions take place, are generated using an open-source software. Different configurations of the anode support layer with various particle sizes and volume fractions of the catalyst and electrolyte phases for different porosities are combined with a consistent AFL to isolate the effect of ASL microstructural parameters. The triple phase boundary (TPB) density, chosen as a useful metric monitoring the anode performance, is then quantitively determined for each anode layer and entire anode in all reconstructed microstructures. The results demonstrate that the microstructure of ASL significantly influences the anode and thus cell performance by impacting reactant gas supply and current collection, emphasizing the necessity for meticulous design. Specifically, ASL microstructures with low volume fractions of Ni and the pore phase, and/or substantial differences between the volume fractions of the phases, are observed to result in discontinuous phase networks, which deactivate TPBs in AFL.

The role of 3D printing in chest wall reconstruction.

Technology is advancing fast, and chest wall surgery finds particular benefit in the broader availability of three-dimensional (3D) reconstruction and printing. An increasing number of reports are being published on the use of these resources in virtual 3D reconstructions of chest walls in computed tomography (CT) scans, virtual surgeries, 3D printing of real-size models for surgical planning, practice, and education, and of note, the manufacture of customized 3D printed implants, changing the fundamental conception from a surgery that fits all, to a surgery for each patient. In this review, we explore the evidence published on simple chest wall reconstruction, including the use of 3D technology to assist in the improvement of the repair of the most frequent chest wall deformities: pectus excavatum and carinatum. Current studies are oriented to the automatization and customization of transthoracic implants, as well as education on real-size models. Next, we investigate the implementation of 3D printing in the repair of complex chest wall reconstruction, comprised of infrequent chest wall deformities such as pectus arcuatum and Poland syndrome. These malformations are very heterogeneous resulting in a high degree of improvisation during the surgical repair. In this setting, 3D technology plays a role in the standardization of a process that contemplates customization, concepts that may seem contradictory. Finally, 3D printing with biocompatible materials is rapidly becoming the first choice for the reconstruction of wide chest wall oncological resections. In this work, we review the first and most important current publications on the subject.

Estimating body volumes and surface areas of animals from cross-sections.

Body mass and surface area are among the most important biological properties, but such information is lacking for some extant organisms and most extinct species. Numerous methods have been developed for body size estimation of animals for this reason. There are two main categories of mass-estimating approaches: extant-scaling approaches and volumetric-density approaches. Extant-scaling approaches determine the relationships between linear skeletal measurements and body mass using regression equations. Volumetric-density approaches, on the other hand, are all based on models. The models are of various types, including physical models, 2D images, and 3D virtual reconstructions. Once the models are constructed, their volumes are acquired using Archimedes' Principle, math formulae, or 3D software. Then densities are assigned to convert volumes to masses. The acquisition of surface area is similar to volume estimation by changing math formulae or software commands. This article presents a new 2D volumetric-density approach called the cross-sectional method (CSM). The CSM integrates biological cross-sections to estimate volume and surface area accurately. It requires a side view or dorsal/ventral view image, a series of cross-sectional silhouettes and some measurements to perform the calculation. To evaluate the performance of the CSM, two other 2D volumetric-density approaches (Graphic Double Integration (GDI) and Paleomass) are compared with it. The CSM produces very accurate results, with average error rates around 0.20% in volume and 1.21% in area respectively. It has higher accuracy than GDI or Paleomass in estimating the volumes and areas of irregular-shaped biological structures. Most previous 2D volumetric-density approaches assume an elliptical or superelliptical approximation of animal cross-sections. Such an approximation does not always have good performance. The CSM processes the true profiles directly rather than approximating and can deal with any shape. It can process objects that have gradually changing cross-sections. This study also suggests that more attention should be paid to the careful acquisition of cross-sections of animals in 2D volumetric-density approaches, otherwise serious errors may be introduced during the estimations. Combined with 2D modeling techniques, the CSM can be considered as an alternative to 3D modeling under certain conditions. It can reduce the complexity of making reconstructions while ensuring the reliability of the results.

Mandibular Reconsctruction using Double Barrel Costal Non Vascularized Bone Graft with Stereolitographic Three Dimensional Printed Models

Surgeons face problems in reconstructing the facial profile after mandibular resection aesthetically and practically. The stereolithographic method was used in the 3D virtual reconstruction program for preoperative planning. This study aims to document the use of 3D printed models in the reconstruction of mandibular ameloblastoma patients under general anesthesia after segmental resection with non-vascularized autogenous costal bone grafts arranged in double-barrel sequence at Airlangga University Hospital, Surabaya, Indonesia. A single ameloblastoma from an anterior mandibular patient undergoing segmental mandibular resection is reported here. Autodesk Fusion 360 was used preoperatively to simulate anatomical markers, surgical margins, and plate morphology to create 3D models used for plate pre-bending. The graft used for reconstruction is a non-vascularized autogenous costal bone graft which is then stacked with an anteroposterior double-barrel. The patient did not show any severe complications associated with the surgery. Stereolithographic helps determine the extent of the lesion in the mandible. 3D printing improves graft viability by enabling optimal planning of the template, reduces operating room time by pre-bent screw plates, and increases procedural precision. The photo was taken a long time to process using virtual reconstruction software, but it put the surgeon at ease.

A systematic review of virtual 3D reconstructions of Cultural Heritage in immersive Virtual Reality

AbstractImmersive Virtual Reality (iVR) devices are increasingly affordable and accessible to consumers. The widespread adoption of this technology for professional training is now finding its way into various other fields. One field that is gaining significant popularity is Cultural Heritage (CH), where iVR enables the reconstruction and exploration of lost heritage. However, an up-to-date systematic review of iVR within this field will be of great benefit. Hence, the present review of 94 papers published between 2013 and 2022 that follows PRISMA methodology on virtual reconstruction of CH for iVR. The aim is to identify the key factors behind the development of these applications and their standards. To do so, a statistical analysis on the following topics was performed: (1) nationality, publication date, and article type; (2) heritage type and its current state of preservation; (3) the area of final application and the features of the reconstructions; (4) the characteristics of the iVR experience; and (5) the assessment of the iVR applications. Finally, a roadmap of best practices is outlined for the virtual reconstruction of CH using iVR and some of the most promising future research lines are outlined.

Comprehensive design of the 3D printing process for architectural models. A case study: the medieval walled enclosure of Priego De Cordoba (E1:200)

AbstractNowadays, an object can be viewed both physically and virtually. In both cases, the visualization process consists in a continuous variation of the observer’s perspective. When looking at objects, visual interactivity is achieved by the tandem formed by human vision and the position of the observer. 3D technology allows the viewer to delve into this aspect: starting from a virtual reconstruction. It is possible to go from the virtual 3D reconstruction of a model shown on a display to a real element materialized by means of a scale model or prototype. The innovation of this research lies in the development of an efficient method for printing 3D models, with a specific emphasis on material and time savings. This study aims to obtain the three-dimensional physical model of the walled enclosure of the city of Priego de Córdoba in the Middle Ages, at its time of greatest historical development, i.e., in the late 15th century. To this end, 3D printing technology was applied, which is an innovative method for the realization of architectural scale models. The performance of 3D printers, with different printing formats, was evaluated in terms of efficiency and quality of the architectural scale models. Our theoretical assumption showed that material saving is determined by the orography slopes of the model. Obtaining slopes with values in the range of 1–5%, a large format printer is the most suitable option; in the range of 5–15%, the medium format printer is recommended; and, for a slope greater than 15%, the small format printer should be used.

Based on 3D Virtual Reconstruction of Modern City Landscape Sculpture Planning Design

INTRODUCTION: With the continuous advancement of urbanization, urban landscape sculpture plays an increasingly important role in modern urban planning. Traditional planning and design methods make it challenging to demonstrate the three-dimensional sense and artistry of sculpture fully; therefore, this study explores a new method of planning and designing modern urban landscape sculpture based on three-dimensional virtual reconstruction.OBJECTIVES: This study aims to enhance the three-dimensional sense and artistry of urban landscape sculpture planning and design through three-dimensional virtual reconstruction technology to meet the needs of modern urban development better. By using advanced technical means, the planning and design can be made more intuitive and specific and provide urban residents with a more artistic public space.METHODS: The study adopts advanced three-dimensional virtual reconstruction technology, combined with urban planning and design theory, to plan and design modern urban landscape sculpture. Firstly, relevant literature on urban planning and sculpture design is collected to understand the existing design concepts and technical means. Secondly, a detailed virtual reconstruction of the sculpture is carried out by using three-dimensional modeling software to show the three-dimensional effect of the sculpture. Finally, the design scheme is optimized and improved through fieldwork and expert review.RESULTS: Through three-dimensional virtual reconstruction technology, this study successfully shows the whole picture of modern urban landscape sculpture. The design scheme not only has a three-dimensional sense but it has also been improved in artistry. The results of fieldwork and expert evaluation show that the new design scheme is more in line with the needs of urban development and adds a unique artistic atmosphere to the urban space.CONCLUSION: This study has achieved positive results in the field of modern urban landscape sculpture planning and design through 3D virtual reconstruction technology. The new design method not only provides a more specific tool for urban planners but also creates a more creative and artistic public space for urban residents. In the future, the application of this method in different urban contexts can be further explored and expanded to inject more innovation and vitality into urban planning and sculpture design.

Whole Spine Segmentation Using Object Detection and Semantic Segmentation.

Virtual and augmented reality have enjoyed increased attention in spine surgery. Preoperative planning, pedicle screw placement, and surgical training are among the most studied use cases. Identifying osseous structures is a key aspect of navigating a 3-dimensional virtual reconstruction. To automate the otherwise time-consuming process of labeling vertebrae on each slice individually, we propose a fully automated pipeline that automates segmentation on computed tomography (CT) and which can form the basis for further virtual or augmented reality application and radiomic analysis. Based on a large public dataset of annotated vertebral CT scans, we first trained a YOLOv8m (You-Only-Look-Once algorithm, Version 8 and size medium) to detect each vertebra individually. On the then cropped images, a 2D-U-Net was developed and externally validated on 2 different public datasets. Two hundred fourteen CT scans (cervical, thoracic, or lumbar spine) were used for model training, and 40 scans were used for external validation. Vertebra recognition achieved a mAP50 (mean average precision with Jaccard threshold of 0.5) of over 0.84, and the segmentation algorithm attained a mean Dice score of 0.75 ± 0.14 at internal, 0.77 ± 0.12 and 0.82 ± 0.14 at external validation, respectively. We propose a 2-stage approach consisting of single vertebra labeling by an object detection algorithm followed by semantic segmentation. In our externally validated pilot study, we demonstrate robust performance for our object detection network in identifying individual vertebrae, as well as for our segmentation model in precisely delineating the bony structures.

The Application of Image Graphic Visual Communication Techniques in Artistic Creation from the Perspective of Digital Media

Abstract This paper first studies the connection between computer image graphics and visual communication and analyses the use of image graphics and visual communication technology in artistic creation. It denoises the image and proposes a method for enhancing blurred light images based on visual communication technology. It also outlines the process of applying computer graphic image processing technology to the field of visual communication design. Then VR technology is combined with visual communication and art design, a 3D image virtual reconstruction system based on visual communication effects is designed, and a 3D image graphical visual art creation method is constructed. Finally, we analyzed the impact of virtual reality art creation in terms of aesthetic evaluation, experience impact, and satisfaction. The results showed that the simple main effect of expression was significant when intensity was perceived for three different types of emotions, as shown by the simple effect analysis of expression on the type of emotion, which were FA(2, 160)=43.05, p<0.001, FB(2, 160)=10.45, p<0.001, and FA*B(2, 160)=12.36, p <0.001. This study complements the research on virtual aesthetics and, at the same time, provides a new research direction for the diversified and integrated development of art.

Virtual reconstruction of stone tool refittings by using 3D modelling and the Blender Engine: The application of the "ReViBE" protocol to the archaeological record.

Visual representation of material culture plays a crucial role in prehistoric archaeology, from academic research to public outreach and communication. Scientific illustration is a valuable tool for visualising lithic artefacts and refittings, where technical attributes must be drawn to enhance our understanding of their significance. However, the representation of lithic refittings, which involve dynamic and sequential transformations of a volume, requires an alternative approach to traditional two-dimensional models such as photography or illustration. Advances in imaging technologies have improved our ability to capture and communicate the multifaceted nature of archaeological artefacts. In this context, we present the ReViBE protocol (Refitting Visualisation using Blender Engine), which integrates photogrammetry, 3D modelling and the animation software Blender© for the virtual representation of lithic refittings. This protocol allows the sequential study of core reduction phases and their associated flakes, as well as other aspects related to knapping decision making (core rotations, surface modifications, and direction and position of impact points). Thus, this method allows the visualisation of techno-cognitive aspects involved in core reduction through a step-by-step animation process. In addition, the 3D models and virtual reconstructions generated by ReViBE can be accessed through open repositories, in line with the principles of open science and FAIR (Findable, Accessible, Interoperable, and Reusable) data. This accessibility ensures that data on lithic technology and human behaviour are widely available, promoting transparency and knowledge sharing, and enabling remote lithic analysis. This in turn breaks down geographical barriers and encourages scientific collaboration.

Multisensor and Multiscale Data Integration Method of TLS and GPR for Three-Dimensional Detailed Virtual Reconstruction.

Integrated TLS and GPR data can provide multisensor and multiscale spatial data for the comprehensive identification and analysis of surficial and subsurface information, but a reliable systematic methodology associated with data integration of TLS and GPR is still scarce. The aim of this research is to develop a methodology for the data integration of TLS and GPR for detailed, three-dimensional (3D) virtual reconstruction. GPR data and high-precision geographical coordinates at the centimeter level were simultaneously gathered using the GPR system and the Global Navigation Satellite System (GNSS) signal receiver. A time synchronization algorithm was proposed to combine each trace of the GPR data with its position information. In view of the improved propagation model of electromagnetic waves, the GPR data were transformed into dense point clouds in the geodetic coordinate system. Finally, the TLS-based and GPR-derived point clouds were merged into a single point cloud dataset using coordinate transformation. In addition, TLS and GPR (250 MHz and 500 MHz antenna) surveys were conducted in the Litang fault to assess the feasibility and overall accuracy of the proposed methodology. The 3D realistic surface and subsurface geometry of the fault scarp were displayed using the integration data of TLS and GPR. A total of 40 common points between the TLS-based and GPR-derived point clouds were implemented to assess the data fusion accuracy. The difference values in the x and y directions were relatively stable within 2 cm, while the difference values in the z direction had an abrupt fluctuation and the maximum values could be up to 5 cm. The standard deviations (STD) of the common points between the TLS-based and GPR-derived point clouds were 0.9 cm, 0.8 cm, and 2.9 cm. Based on the difference values and the STD in the x, y, and z directions, the field experimental results demonstrate that the GPR-derived point clouds exhibit good consistency with the TLS-based point clouds. Furthermore, this study offers a good future prospect for the integration method of TLS and GPR for comprehensive interpretation and analysis of the surficial and subsurface information in many fields, such as archaeology, urban infrastructure detection, geological investigation, and other fields.

LOW-COST WORKFLOW FOR 3D URBAN FOREST VIRTUAL RECONSTRUCTION

Abstract. This contribution addresses, on one hand, the increasing need and demand of 3D data (e.g., point clouds) in urban forestry – for monitoring, analysis, and visualisation purposes –, on the other, the lack of standardised procedures to manage them, also towards engaging and empowering the broader audience. Practitioners in landscape architecture – as a potential bridging field between forestry, geomatics and computer science – could ‘smartly’ manage dynamic ecosystems as forests, parks, and gardens in urban contexts if efficiently provided with digital tools, technologies and techniques for design-oriented information. To this end, a low-cost workflow for 3D virtual reconstruction of under-storey urban vegetation is here proposed to allow exploiting the aesthetic, analytical, and generative potential of 3D point clouds in landscape and urban design. As a proof-of-concept, a combination of a wide range of methods (e.g., spherical photogrammetry, point cloud modelling, virtual reality), tools (e.g., 360° camera, tablet with lidar sensor), and software (e.g., Agisoft Metashape, 3DF Zephyr, CloudCompare, Unity) is tested in the development of a virtual plot of the Mesiano University park in Trento (Italy). Covering all the different steps of the information management activities (namely acquisition, processing, and visualisation), the presented workflow enables a GIS-based, texturized, real-world 3D point cloud rendering of urban forests to be intuitively and interactively experienced in Virtual Reality.

Digital Approach in Conservation of Heritage: 3D Virtual Reconstruction Applications in Ancient Cities

Digital Approach in Conservation of Heritage: 3D Virtual Reconstruction Applications in Ancient Cities

3D multiplexed tissue imaging reconstruction and optimized region of interest (ROI) selection through deep learning model of channels embedding.

Introduction: Tissue-based sampling and diagnosis are defined as the extraction of information from certain limited spaces and its diagnostic significance of a certain object. Pathologists deal with issues related to tumor heterogeneity since analyzing a single sample does not necessarily capture a representative depiction of cancer, and a tissue biopsy usually only presents a small fraction of the tumor. Many multiplex tissue imaging platforms (MTIs) make the assumption that tissue microarrays (TMAs) containing small core samples of 2-dimensional (2D) tissue sections are a good approximation of bulk tumors although tumors are not 2D. However, emerging whole slide imaging (WSI) or 3D tumor atlases that use MTIs like cyclic immunofluorescence (CyCIF) strongly challenge this assumption. In spite of the additional insight gathered by measuring the tumor microenvironment in WSI or 3D, it can be prohibitively expensive and time-consuming to process tens or hundreds of tissue sections with CyCIF. Even when resources are not limited, the criteria for region of interest (ROI) selection in tissues for downstream analysis remain largely qualitative and subjective as stratified sampling requires the knowledge of objects and evaluates their features. Despite the fact TMAs fail to adequately approximate whole tissue features, a theoretical subsampling of tissue exists that can best represent the tumor in the whole slide image. Methods: To address these challenges, we propose deep learning approaches to learn multi-modal image translation tasks from two aspects: 1) generative modeling approach to reconstruct 3D CyCIF representation and 2) co-embedding CyCIF image and Hematoxylin and Eosin (H&E) section to learn multi-modal mappings by a cross-domain translation for minimum representative ROI selection. Results and discussion: We demonstrate that generative modeling enables a 3D virtual CyCIF reconstruction of a colorectal cancer specimen given a small subset of the imaging data at training time. By co-embedding histology and MTI features, we propose a simple convex optimization for objective ROI selection. We demonstrate the potential application of ROI selection and the efficiency of its performance with respect to cellular heterogeneity.

On Roth’s “human fossil” from Baradero, Buenos Aires Province, Argentina: morphological and genetic analysis

The “human fossil” from Baradero, Buenos Aires Province, Argentina, is a collection of skeleton parts first recovered by the paleontologist Santiago Roth and further studied by the anthropologist Rudolf Martin. By the end of the nineteenth century and beginning of the twentieth century it was considered one of the oldest human skeletons from South America's southern cone. Here, we present the results of an interdisciplinary approach to study and contextualize the ancient individual remains. We discuss the context of the finding by first compiling the available evidence associated with the historical information and any previous scientific publications on this individual. Then, we conducted an osteobiographical assessment, by which we evaluated the sex, age, and overall preservation of the skeleton based on morphological features. To obtain a 3D virtual reconstruction of the skull, we performed high resolution CT-scans on selected skull fragments and the mandible. This was followed by the extraction of bone tissue and tooth samples for radiocarbon and genetic analyses, which brought only limited results due to poor preservation and possible contamination. We estimate that the individual from Baradero is a middle-aged adult male. We conclude that the revision of foundational collections with current methodological tools brings new insights and clarifies long held assumptions on the significance of samples that were recovered when archaeology was not yet professionalized.

Low-Cost 3D Virtual and Dynamic Reconstruction Approach for Urban Forests: The Mesiano University Park

Urban forests, parks, and gardens are fundamental components of urban sustainability, resilience, and regenerative dynamics. Designers, architects, and landscape architects could smartly manage these dynamic ecosystems if efficiently provided with design-oriented digital tools, technologies, and techniques. However, practitioners lack knowledge and standardized procedures for their uses. The rise of low-cost sensors to generate 3D data (e.g., point clouds) in forestry can also effectively support monitoring, analysis, and visualization purposes for greenery in urban contexts. Adopting an interdisciplinary approach—involving the fields of forestry, geomatics, and computer science—this contribution addresses these issues and proposes a low-cost workflow for 3D virtual reconstructions of urban forests to support information management activities and thus landscape architecture applications. By connecting a wide range of methods (i.e., spherical photogrammetry, point cloud modeling), tools (i.e., 360° camera, tablet with lidar sensor), and software (i.e., Agisoft Metashape, CloudCompare, Autodesk AutoCAD), the proposed workflow is defined and tested in the development of dynamic virtual representations for a plot of the Mesiano University park in Trento (Italy). Finally, comparing acquisition, processing, and elaboration methodologies and their results, the possibility of developing digital twins of urban forests is envisioned.

Attention Mechanism Used in Monocular Depth Estimation: An Overview

Monocular depth estimation (MDE), as one of the fundamental tasks of computer vision, plays important roles in downstream applications such as virtual reality, 3D reconstruction, and robotic navigation. Convolutional neural networks (CNN)-based methods gained remarkable progress compared with traditional methods using visual cues. However, recent researches reveal that the performance of MDE using CNN could be degraded due to the local receptive field of CNN. To bridge the gap, various attention mechanisms were proposed to model the long-range dependency. Although reviews of MDE algorithms based on CNN were reported, a comprehensive outline of how attention boosts MDE performance is not explored yet. In this paper, we firstly categorize recent attention-related works into CNN-based, Transformer-based, and hybrid (CNN–Transformer-based) approaches in the light of how the attention mechanism impacts the extraction of global features. Secondly, we discuss the details and contributions of attention-based MDE methods published from 2020 to 2022. Then, we compare the performance of the typical attention-based methods. Finally, the challenges and trends of the attention mechanism used in MDE are discussed.