Segmentation Of Structures Research Articles

Artificial intelligence methods in dental radiography

Introduction: The application of artificial intelligence in dentistry can significantly improve the outcome of patient treatment and reduce human error. Computer systems and certain methods that use complex algorithms are able to encode digital data about the radiographic image, obtained by X-ray radiation, and transfer them to a computer language for processing a large set of data. Applied as an auxiliary tool, it can give information about the patient and diagnostics of teeth and surrounding dental structures on radiographic images (two-dimensional and three-dimensional). The aim of the paper: The aim of the paper is to present the development and application of artificial intelligence in dentistry and at the same time emphasizing the importance of segmentation methods and methods of improving the quality of dental radiographic images. Discussion: The common goal of the development and application of artificial intelligence methods in dental radiography is directed on obtaining radiograms from which diagnostically valuable information will be collected faster and easier for the purpose of successful treatment of the patients. Methods of segmentation of a dental structures have been developed to automatically number and localize the tooth or independently mark and display the desired structures on the image. Methods of image quality enhancement tend to work on, for example, improving image resolution or reducing metal artifacts. Research has shown that artificial intelligence can be applied in everyday clinical practice, but it has certain limitations. Conclusion: Artificial intelligence methods in dental radiography are an useful auxiliary tool in clinical dental practice because they can speed up the process of retrieving patient data. Automatic localization of dental structure reduces the dentist's time for manual image analysis, but cannot completely replace human knowledge. For the full implementation of the mentioned methods, more development standards and resources are needed to overcome the ethical and legal problems of replacing humans with computer systems.

ScribSD+: Scribble-supervised medical image segmentation based on simultaneous multi-scale knowledge distillation and class-wise contrastive regularization

Despite that deep learning has achieved state-of-the-art performance for automatic medical image segmentation, it often requires a large amount of pixel-level manual annotations for training. Obtaining these high-quality annotations is time-consuming and requires specialized knowledge, which hinders the widespread application that relies on such annotations to train a model with good segmentation performance. Using scribble annotations can substantially reduce the annotation cost, but often leads to poor segmentation performance due to insufficient supervision. In this work, we propose a novel framework named as ScribSD+ that is based on multi-scale knowledge distillation and class-wise contrastive regularization for learning from scribble annotations. For a student network supervised by scribbles and the teacher based on Exponential Moving Average (EMA), we first introduce multi-scale prediction-level Knowledge Distillation (KD) that leverages soft predictions of the teacher network to supervise the student at multiple scales, and then propose class-wise contrastive regularization which encourages feature similarity within the same class and dissimilarity across different classes, thereby effectively improving the segmentation performance of the student network. Experimental results on the ACDC dataset for heart structure segmentation and a fetal MRI dataset for placenta and fetal brain segmentation demonstrate that our method significantly improves the student’s performance and outperforms five state-of-the-art scribble-supervised learning methods. Consequently, the method has a potential for reducing the annotation cost in developing deep learning models for clinical diagnosis.

3D measurement of precise part in complex circumstance using line structured light and improved U-Net

Abstract Line structured light scanning is extensively utilized for the 3D measurement of precise metal parts, but the curvature surfaces and specific materials of these parts generate specular reflection, making it challenging to accurately extract the center of the laser stripe in complex circumstances. Therefore, the primary challenges are the noise separation in the laser stripe image and the extraction of the laser stripe center under complex circumstances. To solve the above problems, an improved U-Net semantic segmentation algorithm is proposed by adding an attention mechanism and modifying skip connections to the classical U-Net network structure for accurate laser stripe segmentation. Secondly, the dual smoothing method of mean smoothing and Savitzky–Golay smoothing is combined with the Hessian matrix algorithm to complete the subpixel extraction of the center point of the laser stripe. Finally, taking the blade and shaft part as the measurement objects, the experimental results demonstrate that the method can obtain more complete, smoother, and denser results than the traditional method under highly reflective surfaces, vital interference spots, and strong ambient light. The proposed method is used for the 3D measurement of the shaft parts, and the diameter measurement maximum error is 0.029 mm, verifying the feasibility of the proposed method.

Mechanism of Abnormal Activation of MEK1 Induced by Dehydroalanine Modification.

Mitogen-activated protein kinase kinase 1 (MAPK kinase 1, MEK1) is a key kinase in the mitogen-activated protein kinase (MAPK) signaling pathway. MEK1 mutations have been reported to lead to abnormal activation that is closely related to the malignant growth and spread of various tumors, making it an important target for cancer treatment. Targeting MEK1, four small-molecular drugs have been approved by the FDA, including Trametinib, Cobimetinib, Binimetinib, and Selumetinib. Recently, a study showed that modification with dehydroalanine (Dha) can also lead to abnormal activation of MEK1, which has the potential to promote tumor development. In this study, we used molecular dynamics simulations and metadynamics to explore the mechanism of abnormal activation of MEK1 caused by the Dha modification and predicted the inhibitory effects of four FDA-approved MEK1 inhibitors on the Dha-modified MEK1. The results showed that the mechanism of abnormal activation of MEK1 caused by the Dha modification is due to the movement of the active segment, which opens the active pocket and exposes the catalytic site, leading to sustained abnormal activation of MEK1. Among four FDA-approved inhibitors, only Selumetinib clearly blocks the active site by changing the secondary structure of the active segment from α-helix to disordered loop. Our study will help to explain the mechanism of abnormal activation of MEK1 caused by the Dha modification and provide clues for the development of corresponding inhibitors.

Deterministic preparation of a dual-species two-ion crystal

The demand for efficient preparation methods for dual-species ion crystals is rapidly expanding across quantum technology and fundamental physics applications with trapped ions. We present a deterministic and efficient technique to produce such crystals, utilizing the segmented structure of a linear Paul trap. By precisely tailoring the trapping potentials, we can split, move, and discard parts of an ion chain. This process is automated in a sequence that converts a larger ion sample into the desired configuration. A critical component of our approach is the accurate identification of crystal constituents. This is achieved by matching the measured positions of fluorescing ions against theoretical expectations for larger crystals, thus facilitating the detection of nonfluorescing ions and enabling accurate ion counting. We demonstrate that our method reliably produces two-ion crystals within minutes. These results represent a significant advance in the production of two-species ion crystals with applications ranging from quantum logic spectroscopy and optical clocks to quantum computing and simulations with trapped ions. Published by the American Physical Society 2024

Respiratory virus infections and adenovirus characteristics during acute exacerbation of chronic obstructive pulmonary disease.

Chronic obstructive pulmonary disease (COPD) is a common respiratory disease globally, characterized by obstructive ventilatory disorder under pulmonary function tests. Recent years have witnessed a yearly increase in the prevalence of COPD. To investigate the impact of respiratory virus infections on patients with acute exacerbation of chronic obstructive pulmonary disease (AECOPD), and to perform sequencing typing and mutation analysis of viruses with high detection rate. A total of 1523 inpatients with AECOPD admitted to our hospital from April 1,2020 to March 30,2022 were collected and divided into two groups: the infected group (n= 532) and the non-infected group (n= 991). The related indexes between the two groups were collected and compared (including clinical characteristics and laboratory tests that blood cell count, PCT, CRP, adenovirus, respiratory syncytial virus, rhinovirus, influenza A virus, influenza B virus, etc.). In the infected group, the proportion of patients with palpitations (49.44% VS 8.07%, P< 0.001), lipid metabolism abnormalities (18.42% VS 39.96%, P< 0.001), heart failure (39.85% VS 29.87%, P< 0.001), disease duration (17.48 ± 7.47 VS 12.45 ± 11.43 d, P< 0.001), and poor prognosis (69.55% VS 17.15%, P< 0.001) were higher than those in the non-infected group; Adenovirus (ADV) accounted for 75.94% (404/532) of all infected viruses. 31 virus strains could be categorized into 16 ADV-C1, one ADV-C5, two ADV-B3, three ADV-B7, two ADV-D17, two ADV-D19, and five ADV-D27, which were similar to the serotypes reported in severe pneumonia. Furthermore, three strains of C1 adenovirus were found to be highly homologous to the original strain AF534906 by sequencing, and the phylogenetic trees of the three main structural genes were all on the same branch as the original strain. Base mutations and amino acid variants were found in each structural gene segment. In clinical data, it's found that patients with mutations are worse than those without mutations. Respiratory viruses are common in patients with poor prognosis of AECOPD, especially adenovirus, respiratory syncytial virus. Respiratory virus infections will lead to the deterioration of patients with AECOPD, accompanied by longer treatment cycles and poor prognosis.

Structure focused neurodegeneration convolutional neural network for modelling and classification of Alzheimer’s disease

Alzheimer’s disease (AD), the predominant form of dementia, is a growing global challenge, emphasizing the urgent need for accurate and early diagnosis. Current clinical diagnoses rely on radiologist expert interpretation, which is prone to human error. Deep learning has thus far shown promise for early AD diagnosis. However, existing methods often overlook focal structural atrophy critical for enhanced understanding of the cerebral cortex neurodegeneration. This paper proposes a deep learning framework that includes a novel structure-focused neurodegeneration CNN architecture named SNeurodCNN and an image brightness enhancement preprocessor using gamma correction. The SNeurodCNN architecture takes as input the focal structural atrophy features resulting from segmentation of brain structures captured through magnetic resonance imaging (MRI). As a result, the architecture considers only necessary CNN components, which comprises of two downsampling convolutional blocks and two fully connected layers, for achieving the desired classification task, and utilises regularisation techniques to regularise learnable parameters. Leveraging mid-sagittal and para-sagittal brain image viewpoints from the Alzheimer’s disease neuroimaging initiative (ADNI) dataset, our framework demonstrated exceptional performance. The para-sagittal viewpoint achieved 97.8% accuracy, 97.0% specificity, and 98.5% sensitivity, while the mid-sagittal viewpoint offered deeper insights with 98.1% accuracy, 97.2% specificity, and 99.0% sensitivity. Model analysis revealed the ability of SNeurodCNN to capture the structural dynamics of mild cognitive impairment (MCI) and AD in the frontal lobe, occipital lobe, cerebellum, temporal, and parietal lobe, suggesting its potential as a brain structural change digi-biomarker for early AD diagnosis. This work can be reproduced using code we made available on GitHub.

Making Arrangements: A Sketch of a ‘Big Package’

ABSTRACT This article examines the making of arrangements for future joint activities as a “big package” of talk in interaction. Arrangement-making is here characterized as having a segmented structure, a series of sequences in which the details of time, place, and manner are addressed, often discretely, in sequences that are boundaried off from one another and variable in their ordering. The analysis also addresses the launching of the package and the varying ways it can emerge out of a prior remote proposal sequence or an already presumed activity, also, some of the contingencies impacting the body of the arrangement, and the sequential mechanisms through which segments and the overall package are closed. It is suggested that the segmented “sequence of sequences” structure may apply to the analysis of other big packages. Data are in English.

Novel Reactive Polyhedral Oligomeric Silsesquioxane-Reinforced and Toughened Epoxy Resins for Advanced Composites.

Most toughening methods for epoxy resins are usually used at the expense of other properties. Some polyhedral oligomeric silsesquioxanes (POSSs) with both a rigid Si-O-Si structure and flexible organic chain segments could be expected to be effective toughening agents. In this study, three reactive polyhedral oligomeric silsesquioxanes with a thiol group (OMPPS), a carboxyl group (OCOPS), and an epoxy group (OGCPS) were synthesized and characterized. They were utilized as modifiers to toughen 3-(oxiran-2-ylmethoxy)-N,N-bis(oxiran-2-ylmethyl)aniline (AFG-90MH)/4,4'-methylenebis(2-ethylaniline) (MOEA) (epoxy resin) with different molar ratios to obtain hybrid resins named OMPPS-EP-i, OCOPS-EP-j, and OGCPS-EP-k. The effects of the amount of modifier added and the length of the organic chain on the cage structure on various properties of the hybrid resins were investigated. The results show that all three modifiers show good compatibility with the epoxy resin. The hybrid resins have a low viscosity at 45~85 °C and can be cured at a low temperature (110 °C). The cured hybrid resins display improved toughness. Typically, the critical stress intensity factor (KIC) and impact strength of OGCPS-EP-0.6-C are 2.54 MPa∙m-1/2 and 19.33 kJ∙m-2, respectively, which increased by 58.75% and 22.48% compared with the pristine epoxy resin, respectively. In addition, the glass transition temperature and flexural strength of the hybrid resins are basically unchanged.

Effect of transbronchial or intravenous administration of indocyanine green on resection margins during near-infrared-guided segmentectomy: a review.

For early-stage non-small cell lung cancer, surgical resection remains the best treatment option. Currently, sublobar resection, including segmentectomy, is recommended in these cases, as it provides a better quality of life with the same oncological outcomes; however, is requires adequate resection margins. Accurate preoperative planning and proper identification of the intersegmental planes during thoracic surgery are crucial for ensuring precise surgical management and adequate resection margins. Three dimensional computed tomography reconstruction and near-infrared-guided intersegmental plane identification can greatly facilitate the surgical procedures. Three-dimensional computed tomography reconstruction can simulate both the resection and resection margins. Indocyanine green is one of the most frequently used and affordable fluorophores. There are two ways to identify the intersegmental planes using indocyanine green: intravenous and transbronchial administration. Intravenous application is simple; however, its effectiveness may be affected by underlying lung disease, and it requires the isolation of segmental structures before administration. Transbronchial use requires appropriate bronchoscopic skills and preoperative planning; however, it also allows for delineation deep in the parenchyma and can be used for complex segmentectomies. Both methods can be used to ensure adequate resection margins and, therefore, achieve the correct oncological radicality of the surgical procedure. Here, we summarise these applications and provide an overview of their different possibilities.

Impact of Block Ratio, Polymer Architecture, and Soft Segment Structure on Modified Asphalt Rheological Performance

Impact of Block Ratio, Polymer Architecture, and Soft Segment Structure on Modified Asphalt Rheological Performance

Beckett’s Not I as a Dramatic Rendition of Kristeva’s Notion of the Semiotic Chora

Samuel Beckett as a dramatist is well-known with his contributions to the landscape of British theatre. His plays reverberate with voices of traumatized, marginalized and war-stricken characters, which function as an implicit criticism of the turbulent atmosphere of the 1900s. When Beckett addresses the problems of his time in a covert way, the playwright deconstructs conventional elements of drama by specifying no plot, no setting or no theme. This Beckettian style also becomes manifest in his use of language which is characterized by the presence of segmented structures, pauses, ellipses and even silence. Considering these tenets of Beckettian drama, the thematic concerns of his theatrical productions are assumed to primarily revolve around the issues of nihilism, language, and ontology. Therefore, Beckett’s plays are thought to be read through the lenses of distinguished theorists such as Sigmund Freud, Carl Jung, Jacques Lacan and Julia Kristeva. From among these theorists, the concepts of the semiotic, the symbolic and the semiotic chora of Julia Kristeva as a poststructuralist thinker provide a fertile ground for a theoretical reading of Beckett’s play Not I (1972). Relevantly, this paper principally examines the applicability of Kristeva’s “semiotic chora” into Beckett’s Not I in all aspects.

Domain composition and attention network trained with synthesized unlabeled images for generalizable medical image segmentation

Despite that deep learning models have achieved remarkable performance in medical image segmentation, their performance is often limited on testing images from new centers with a domain shift. To achieve Domain Generalization (DG) for medical image segmentation, we propose a Domain Composition and Attention-based Network (DCA-Net) combined with structure- and style-based data augmentation that generates unlabeled synthetic images for training. First, DCA-Net represents features in one certain domain by a linear combination of a set of basis representations that are learned by parallel domain preceptors with a divergence constraint. The linear combination is used to calibrate the feature maps of an input image, which enables the model to generalize to unseen domains. Second, considering the number of domains and images for training is limited, we employ generative models to synthesize images with a higher structure diversity, and to leverage the unlabeled synthetic images, we introduce a consistency constraint for their predictions under style augmentation based on frequency amplitude mixture. Additionally, a Test-Time Frequency Augmentation (TTFA) is proposed to neutralize the domain shift from the target to source domains. Experimental results on two multi-domain datasets for fundus structure and nasopharyngeal carcinoma segmentation showed that: (1) our method significantly outperformed several existing DG methods, and (2) the model’s generalizability was largely improved by domain composition and attention modules; (3) by leveraging the unlabeled synthetic images and the TTFA, the model could better deal with images from unseen domains.

Synthesis, electro‐optic properties, and performance of novel fluorinated polyurethaneimide

AbstractUrethane imide anhydride oligomers with terminal groups as anhydride groups were designed and synthesized using second‐order nonlinear optical chromophore molecule, bisphenol AF‐type diethylidene dianhydride, and toluene diisocyanate as monomers. Polyurethaneimide (PUI) second‐order nonlinear optical polymers with the structure of imide and amino ester chain segments were then synthesized by the polycondensation reaction between urethane imide anhydride oligomers and aromatic diamine monomers. The glass transition temperature (Tg) of PUI reached 186°C, and the 5% heat‐loss decomposition temperature was 320°C. The corona‐polarized PUI had an electro‐optic (EO) coefficient (γ33) of 82 pm/V tested by the reflection method. The loss of light transmission of the polymer waveguide prepared with PUI as electro‐optical material was 1.13 dB/cm at 1550 nm. Mach–Zehnder type polymer waveguide electro‐optical modulators have been prepared based on the properties of the electro‐optic material PUI. The fabricated polymer EO modulators had favorable electro‐optic modulation response performance for low‐frequency electrical signals. The good modulation characteristics of the manufactured electro‐optical modulators proved that the synthesized PUI was a waveguide polymer material with good comprehensive performance.Highlights Electro‐optic fluorinated polyurethaneimide was synthesized. Polyurethaneimide as core material, polymer waveguide EO modulators were fabricated. The prepared modulators had good electro‐optic modulation performance.

Influence of genetic mutations to atria vulnerability to atrial fibrillation: An in-silico 3D human atria study

Background and ObjectivePersonalized 3D computer models of atria have been extensively implemented in the last yearsas a tool to facilitate the understanding of the mechanisms underlying different forms of arrhythmia, such as atrial fibrillation (AF). Meanwhile, genetic mutations acting on potassium channel dynamics were demonstrated to induce fibrillatory episodes in asymptomatic patients. This research study aims at assessing the effects and the atrial susceptibility to AF of three gain-of-function mutations – namely, KCNH2 T895M, KCNH2 T436M, and KCNE3-V17M – associated with AF outbreaks, using highly detailed 3D atrial models with realistic wall thickness and heterogenous histological properties. MethodsThe 3D atrial model was generated by reconstructing segmented anatomical structures from CT scans of an AF patient. Modified versions of the Courtemanche human atrial myocyte model were used to reproduce the electrophysiological activity of the WT and of the three mutant cells. Ectopic foci (EF) were simulated in sixteen locations across the atrial mesh using an S1-S2 protocol with two S2 basic cycle lengths (BCL) and eleven coupling intervals in order to induce arrhythmias. ResultsThe three genetic mutations at 3D level reduced the APD90. The KCNE3-V17M mutation provoked the highest shortening (55 % in RA and LA with respect to WT), followed by KCNH2 T895M (14 % in RA and 18 % LA with respect to WT)and KCNH2 T436M (7 % in RA and 9 % LA with respect to WT). The KCNE3-V17M mutation led to arrhythmia in 67 % of the cases simulated and in 94 % of ectopic foci considered, at S2 BCL equal to 100 ms. The KCNH2 T436M and KCNH2 T895M mutations increased the vulnerability to AF in a similar way, leading to arrhythmic episodes in 7 % of the simulated conditions, at S2 BCL set to 160 ms. Overall, 60 % of the arrhythmic events generated arise in the left atrium. Spiral waves, multiple rotors and disordered electrical pattern were elicited in the presence of the KCNE3-V17M mutation, exhibiting an instantaneous mean frequency of 7.6 Hz with a mean standard deviation of 1.12 Hz. The scroll waves induced in the presence of the KCNH2 T436M and KCNH2 T895M mutations showed steadiness and regularity with an instantaneous mean frequencies in the range of 4.9 – 5.1 Hz and a mean standard deviation within 0.19 – 0.53 Hz. ConclusionsThe pro-arrhythmogenicity of the KCNE3-V17M, KCNH2 T895M and KCNH2 T436M mutations was studied and proved on personalized 3D cardiac models. The three genetic mutations were demonstrated to increase the predisposition of atrial tissue to the formation of AF-susceptible substrate in different ways based on their effects on electrophysiological properties of the atria.

Integrating self-organizing feature map with graph convolutional network for enhanced superpixel segmentation and feature extraction in non-Euclidean data structure

Integrating self-organizing feature map with graph convolutional network for enhanced superpixel segmentation and feature extraction in non-Euclidean data structure

A Label-Efficient Framework for Automated Sinonasal CT Segmentation in Image-Guided Surgery.

Segmentation, the partitioning of patient imaging into multiple, labeled segments, has several potential clinical benefits but when performed manually is tedious and resource intensive. Automated deep learning (DL)-based segmentation methods can streamline the process. The objective of this study was to evaluate a label-efficient DL pipeline that requires only a small number of annotated scans for semantic segmentation of sinonasal structures in CT scans. Retrospective cohort study. Academic institution. Forty CT scans were used in this study including 16 scans in which the nasal septum (NS), inferior turbinate (IT), maxillary sinus (MS), and optic nerve (ON) were manually annotated using an open-source software. A label-efficient DL framework was used to train jointly on a few manually labeled scans and the remaining unlabeled scans. Quantitative analysis was then performed to obtain the number of annotated scans needed to achieve submillimeter average surface distances (ASDs). Our findings reveal that merely four labeled scans are necessary to achieve median submillimeter ASDs for large sinonasal structures-NS (0.96 mm), IT (0.74 mm), and MS (0.43 mm), whereas eight scans are required for smaller structures-ON (0.80 mm). We have evaluated a label-efficient pipeline for segmentation of sinonasal structures. Empirical results demonstrate that automated DL methods can achieve submillimeter accuracy using a small number of labeled CT scans. Our pipeline has the potential to improve pre-operative planning workflows, robotic- and image-guidance navigation systems, computer-assisted diagnosis, and the construction of statistical shape models to quantify population variations. N/A.

Study on dynamic and failure characteristics of sand and anti-floating safety of tunnel structures under long-term vibration loads of the Beijing subway

Most research into sand liquefaction has predominantly focused on the effects of earthquake ground motion, with a limited exploration into the possibility of sand liquefaction induced under long-term train vibration loads. Therefore, in order to explore the possibility of saturated sand liquefaction under long-term vibration loads of the Beijing subway, this article selected Tongzhou sand to conduct dynamic triaxial tests and obtained the dynamic and failure characteristics of saturated sand under different confining pressures and dynamic stresses. The test results showed that the confining pressure has a greater impact on sand liquefaction. When the confining pressure exceeds 100 kPa, the possibility of liquefaction of Tongzhou sand is fairly limited. Furthermore, a numerical model of an operating tunnel under the influence of long-term train vibration loads is established, and the dynamic response characteristics of the saturated sand and tunnel structure are studied. Numerical simulation results show that the local liquefaction of sand caused by train vibration will not cause the tunnel structure to float. Under local liquefaction conditions at the extremely shallow tunnel base, the soil did not completely lose its bearing capacity, and the displacement of the soil and structure was still dominated by settlement. Changes in tunnel burial depth have a greater impact on segment structure than train vibration loads. Tunnels with a more than 10 m burial depth will not liquefy in saturated sand within one year, and the segment structure is safe. The research results can provide a reference for the anti-floating safety of tunnel structures in similar saturated sand strata.

S5Utis: Structured State-Space Sequence SegNeXt UNet-like Tongue Image Segmentation in Traditional Chinese Medicine.

Intelligent Traditional Chinese Medicine can provide people with a convenient way to participate in daily health care. The ease of acceptance of Traditional Chinese Medicine is also a major advantage in promoting health management. In Traditional Chinese Medicine, tongue imaging is an important step in the examination process. The segmentation and processing of the tongue image directly affects the results of intelligent Traditional Chinese Medicine diagnosis. As intelligent Traditional Chinese Medicine continues to develop, remote diagnosis and patient participation will play important roles. Smartphone sensor cameras can provide irreplaceable data collection capabilities in enhancing interaction in smart Traditional Chinese Medicine. However, these factors lead to differences in the size and quality of the captured images due to factors such as differences in shooting equipment, professionalism of the photographer, and the subject's cooperation. Most current tongue image segmentation algorithms are based on data collected by professional tongue diagnosis instruments in standard environments, and are not able to demonstrate the tongue image segmentation effect in complex environments. Therefore, we propose a segmentation algorithm for tongue images collected in complex multi-device and multi-user environments. We use convolutional attention and extend state space models to the 2D environment in the encoder. Then, cross-layer connection fusion is used in the decoder part to fuse shallow texture and deep semantic features. Through segmentation experiments on tongue image datasets collected by patients and doctors in real-world settings, our algorithm significantly improves segmentation performance and accuracy.

Design and movement mechanism analysis of a multiple degree of freedom bionic crocodile robot based on the characteristic of “death roll”

AbstractThis paper introduces a multi‐degree of freedom bionic crocodile robot designed to tackle the challenge of cleaning pollutants and debris from the surfaces of narrow, shallow rivers. The robot mimics the “death roll” motion of crocodiles which is a technique used for object disintegration. First, the design incorporated a swinging tail mechanism using a multi‐section oscillating guide‐bar mechanism. By analyzing three‐, four‐, and five‐section tail structures, the four‐section tail was identified as the most effective structure, offering optimal strength and swing amplitude. Each section of the tail can reach maximum swing angles of 8.05°, 20.95°, 35.09°, and 43.84°, respectively, under a single motor's drive. Next, the robotic legs were designed with a double parallelogram mechanism, facilitating both crawling and retracting movements. In addition, the mouth employed a double‐rocker mechanism for efficient closure and locking, achieving an average torque of 5.69 N m with a motor torque of 3.92 N m. Moreover, the robotic body was designed with upper and lower segment structures and waterproofing function was also considered. Besides, the kinematic mechanism and mechanical properties of the bionic crocodile structure were analyzed from the perspectives of modeling and field tests. The results demonstrated an exceptional kinematic performance of the bionic crocodile robot, effectively replicating the authentic movement characteristics of a crocodile.