Continuous Segments Research Articles

BIOM-53. LOW-PASS WHOLE GENOME SEQUENCING OF CEREBROSPINAL FLUID IDENTIFIES TUMOR AGNOSTIC TP53 ABERRATIONS IN LEPTOMENINGEAL METASTATIC DISEASE

Abstract Leptomeningeal metastatic disease (LMD) is the spread of cancer cells into the cerebrospinal fluid (CSF)-filled spaces surrounding the central nervous system (CNS). Once LMD occurs, patients endure devastating neurologic symptoms and survive for only a few weeks to months. The clinical diagnosis of LMD has risen as patient survival from non-CNS metastatic disease improves. Unfortunately, detection of LMD even in symptomatic patients is <50% using the current gold standard – CSF cytology. Therefore, a strong clinical need exists for the sensitive and early detection of LMD. Here, we sought to develop a minimally invasive liquid biopsy approach to diagnose LMD using next-generation sequencing. Because copy number variations (CNVs) are common in cancer and increase after chemoradiation, we hypothesized that CNV detection in CSF may offer a means to detect LMD since patients have already undergone treatment for the underlying disease. Specifically, we sought to develop low-pass whole genome sequencing (lpWGS) of cell-free DNA in CSF to detect CNVs associated with LMD. A novel lpWGS algorithm was developed using FASTQ files trimmed to 30 million total paired reads. The genome was partitioned into one million continuous base pair segments excluding problematic areas yielding 2,445 regions across 22 autosomes. Control samples were used to model the read depth for each region to establish the euploid state. Read depth across the genome for all samples was 0.6X. CSF samples from lung cancer, breast cancer, and melanoma patients with LMD demonstrated prominent evidence of monosomies and trisomies throughout the genome. Lung cancer exhibited frequent amplifications (>4-fold). Notably, we observed 17p monosomy in CSF regardless of cancer type which indicates TP53 haploinsufficiency may support LMD progression, a conjecture supported by a codeletion of TP53 in one patient. lpWGS detects LMD in solid tumor cancers and may also identify novel mechanisms for LMD progression.

Exploiting Temporal Features in Calculating Automated Morphological Properties of Spiky Nanoparticles Using Deep Learning

Object segmentation in images is typically spatial and focuses on the spatial coherence of pixels. Nanoparticles in electron microscopy images are also segmented frame by frame, with subsequent morphological analysis. However, morphological analysis is inherently sequential, and a temporal regularity is evident in the process. In this study, we extend the spatially focused morphological analysis by incorporating a fusion of hard and soft inductive bias from sequential machine learning techniques to account for temporal relationships. Previously, spiky Au nanoparticles (Au-SNPs) in electron microscopy images were analyzed, and their morphological properties were automatically generated using a hourglass convolutional neural network architecture. In this study, recurrent layers are integrated to capture the natural, sequential growth of the particles. The network is trained with a spike-focused loss function. Continuous segmentation of the images explores the regressive relationships among natural growth features, generating morphological statistics of the nanoparticles. This study comprehensively evaluates the proposed approach by comparing the results of segmentation and morphological properties analysis, demonstrating its superiority over earlier methods.

Analysis of ST segment alarms in children admitted to the paediatric and cardiac intensive care units and cardiac progressive care unit: a single-centre retrospective study.

ST segment monitoring in the adult population allows for the early detection of myocardial ischaemia. In children admitted to the paediatric intensive care unit (PICU), cardiac intensive care unit (CICU), and cardiac progressive care unit (CPCU), it is unclear if continuous ST segment alarm monitoring is necessary in all patients. All patients admitted to the PICU, CICU, and CPCU during the study period were included. Children with any ST segment alarms were compared with those without an alarm during their stay. The electrocardiogram confirmed true ST segment alarms were compared with all other ST segment alarms. Demographic and clinical data were extracted from the medical record. Medical interventions and procedures occurring around ST segment alarms were recorded for multivariable analysis assessing for the association of true ST segment. Logistic regression was used to evaluate the associations with ST segment alarms during hospital stays. ST segment alarms occurred in 36% of hospital stays, and only 3.4% were considered true. True alarms were significantly more common among patients with a cardiac-related diagnosis, located in both cardiac units, and having received an intervention with any vasoactive medication. In the multivariable logistic regression, patients 11 years or older, hypotension, supraventricular tachycardia, and initiation/escalation of any vasoactive were independently associated with a true ST segment alarm. True ST segment alarms were infrequent, occurring in 1.2% of stays during the study period. Alarm monitoring may be beneficial in those with an underlying cardiac diagnosis.

WiLife: Long-term Daily Status Monitoring and Habit Mining of the Elderly Leveraging Ubiquitous Wi-Fi Signals

The global aging demographic underscores the imperative for continuous in-home monitoring of the empty-nest elderly, ensuring their safety and well-being. The widespread deployment of Wi-Fi infrastructure has paved the way to monitor the elderly in a non-intrusive and privacy-preserving manner. Numerous studies have explored the potential of utilizing Wi-Fi signals to address urgent life safety concerns such as fall detection and vital sign monitoring. However, apart from these acute safety issues, the early detection of potential disease symptoms and managing the progression of chronic diseases are also crucial for elderly care, which calls for long-term and continuous monitoring of the elderly’s daily routines. Unfortunately, challenges like continuous activity segmentation and location/orientation dependencies have hindered the implementation of a long-term, around-the-clock activity monitoring system for the elderly. This work introduces ”WiLife”, a cutting-edge Wi-Fi based framework for continuous monitoring of the elderly’s spatio-temporal daily status information. Specifically, WiLife adopts a strategy of partitioning living spaces into functional areas and categorizing daily activities into atomic states . By encapsulating daily life status into a unique series of triple unit format: \(\left\langle{{Time, Area, State}}\right\rangle\) , WiLife is able to offer valuable insights into when, where, and how activities occur. Field implementations spanning 1,080 hours ( \(45days\times 24hours\) ) in real-world home environments highlight WiLife’s exceptional capability in understanding individual living habits and timely detection of irregularities.

Using contrasts in horizontal P-wave reflectivity to map the base of the continental lithosphere: Results for the central and eastern U.S.

Vertical-incidence seismic reflection profiles generated from global phases for 16 earthquakes recorded by stations of the Transportable Array (TA) show distinctive patterns of P-wave reflectivity in the uppermost mantle beneath the central and eastern United States. The overall distribution of reflections identified objectively using the sign test statistic applied to bootstrapped stacks is consistent with a westward increase in depth of the lithosphere-asthenosphere boundary (LAB) from roughly 110 to 250 km that is marked within the lower lithosphere by piecewise continuous segments of elevated horizontal P-wave reflectivity. For some profiles, the onsets of zones of increased reflectivity closely match depths corresponding to the maximum negative S-wave velocity gradients found by surface-wave tomography, suggesting that P-wave reflectivity can be used to help characterize properties of the lower lithosphere. We suggest that the vertical change in horizontal reflectivity straddles the lithosphere-asthenosphere transition, encompassing a broad zone of layering caused by increased strain in the lower lithosphere as well as drag-induced flow in the asthenosphere. Some of the lines also show waveforms that fall within the depth range of arrivals identified as midlithospheric discontinuities (MLDs) in overlapping Sp receiver-function profiles. The reflection waveforms observed in the TA lines are mostly multicyclic with a mix of polarities indicating a layered transition, consistent with previous observations and model studies that show the breakup of single Sp waveforms into a series of less prominent, shorter-period P-wave reflections as the dominant frequency of incident energy is increased.

Dual-stage semantic segmentation of endoscopic surgical instruments.

Endoscopic instrument segmentation is essential for ensuring the safety of robotic-assisted spinal endoscopic surgeries. However, due to the narrow operative region, intricate surrounding tissues, and limited visibility, achieving instrument segmentation within the endoscopic view remainschallenging. This work aims to devise a method to segment surgical instruments in endoscopic video. By designing an endoscopic image classification model, features of frames before and after the video are extracted to achieve continuous and precise segmentation of instruments in endoscopicvideos. Deep learning techniques serve as the algorithmic core for constructing the convolutional neural network proposed in this study. The method comprises dual stages: image classification and instrument segmentation. MobileViT is employed for image classification, enabling the extraction of key features of different instruments and generating classification results. DeepLabv3+ is utilized for instrument segmentation. By training on distinct instruments separately, corresponding model parameters are obtained. Lastly, a flag caching mechanism along with a blur detection module is designed to effectively utilize the image features in consecutive frames. By incorporating specific parameters into the segmentation model, better segmentation of surgical instruments can be achieved in endoscopicvideos. The classification and segmentation models are evaluated on an endoscopic image dataset. In the dataset used for instrument segmentation, the training set consists of 7456 images, the validation set consists of 829 images, and the test set consists of 921 images. In the dataset used for image classification, the training set consists of 2400 images and the validation set consists of 600 images. The image classification model achieves an accuracy of 70% on the validation set. For the segmentation model, experiments are conducted on two common surgical instruments, and the mean Intersection over Union (mIoU) exceeds 98%. Furthermore, the proposed video segmentation method is tested using videos collected during surgeries, validating the effectiveness of the flag caching mechanism and blur detection module. Experimental results on the dataset demonstrate that the dual-stage video processing method excels in performing instrument segmentation tasks under endoscopic conditions. This advancement is significant for enhancing the intelligence level of robotic-assisted spinal endoscopic surgeries.

Preliminary design of a viaduct on new Highspeed line RS2 VRT Jižní Morava

AbstractThe newly proposed high‐speed rail in Czech Republic creates a set of new challenges for engineers in design overall. In this case, the most significant challenge was to design a viaduct over several obstacles due to the crossing of Special Area of Conservation (SAC), part of Natura 2000. This aims to limit impacts in short‐term, during construction and more importantly in the long term. Construction of the viaduct will mitigate an impact to the natural surrounding area, whilst allowing immigration of its natural habitats. Nevertheless, the viaduct design and operational speeds will reach 320 km/h with possible raising to 350 km/h limit, with minimal impact within the area.Initial length of the viaduct was over 1 300 m and the main requirement given by an infrastructure manager was to design a viaduct without railway expansion joints. Due to such a long distance, the viaduct had to be separated into several dilatation segments in order to avoid the use of rail expansion joint. The most susceptible parts of SAC were in proximity of the river Šatava and surrounding wetlands. This area created an obstacle, which had to be crossed by a long span avoiding the implementation of bridge piers. Poor geological conditions close to the watercourse had to be taken into consideration for design of pier foundation, therefore the span had to be shortened to a compromised length.The final design of the viaduct consists of 16 segments, of which 14 are continuous segments and 4 are single spans. Overall, there are 29 spans over the length of the viaduct. Construction method of incremental launching had been chosen to reduce the short‐term impact in the SAC.

Continuous hand gesture segmentation and acknowledgement of hand gesture path for innovative effort interfaces

Human-computer interaction (HCI) has revolutionized the way we interact with computers, making it more intuitive and user-friendly. It is a dynamic field that has found it is applications in various industries, including multimedia and gaming, where hand gestures are at the forefront. The advent of ubiquitous computing has further heightened the interest in using hand gestures as input. However, recognizing continuous hand gestures presents a set of challenges, primarily stemming from the variable duration of gestures and the lack of clear starting and ending points. Our main objective is to propose a solution: the framework for “continuous palm motion analysis and retrieval” based on “Spatial-temporal and path knowledge”. Framework harnesses the power of cognitive deep learning networks (DLN), offering a significant advancement in the continuous hand gesture recognition domain. we conducted rigorous experiments using a diverse video dataset capturing hand gestures for boasting an impressive F-score of up to 0.99. The potential of our framework to significantly enhance the accuracy and reliability of hand gesture recognition in real-world applications.

Floor plan reconstruction from indoor 3D point clouds using iterative RANSAC line segmentation

Indoor Floor Plans (IFPs) are crucial in many fields, such as architectural design, BIM generation, indoor navigation, as well as reliability analysis and safety assessment in civil engineering. The traditional manual creation of IFPs is labor-intensive and time-consuming, hence the growing interest in their automatic generation. To automate IFP creation and ensure their consistency with original indoor layouts, a method for generating IFPs from indoor 3D point clouds is proposed. This method uses a complete point cloud from an indoor scene as the sole input and produces a corresponding vectorized IFP. It first identifies wall points from the 3D point cloud, and projects them onto a 2D plane. An iterative RANSAC-based clustering algorithm is then used to detect and segment line segments from the 2D point cloud. Finally, we restore the topological structure of the discrete line segment set and outputs a vectorized IFP composed of continuous line segments. When tested on two different scale indoor point cloud datasets, this new method achieved over 90 % of accuracy in line segment reconstruction and 97 % of IoU in IFPs reconstruction, proving its effectiveness in non-Manhattan indoor scenes. Compared with other representative methods, Floor-sp and ASIP, this new method offers more detail and higher accuracy.

An Adapted NURBS Interpolator with a Switched Optimized Method of Feed-Rate Scheduling

With the increasing demand for processing precision in the manufacturing industry, feed-rate scheduling is a crucial component in achieving the processing quality of complex surfaces. A smooth feed-rate profile not only guarantees machining quality but also improves machining efficiency. Although the typical offline feed-rate scheduling method possesses good processing efficiency, it may not provide an optimal solution due to the NP-hard problem caused by the feed-rate scheduling of continuous curve segments, which easily results in excess kinetic limitations and feed-rate fluctuations in a real-time interpolation. Instead, the FIR (Finite Impulse Response) method is widely used to realize interpolation in real-time processing. However, the FIR method will filter out a large number of high-frequency signals, leading to a low-processing efficiency. Further, greater acceleration or deceleration is required to ensure the interpolation passes through the segment end at a predefined feed rate and the deceleration in the feed rate profile appears earlier, which allows the interpolation to easily exceed the kinetic limitation. At present, a simple offline or online method cannot realize the global optimization of the feed-rate profile and guarantee the machining efficiency. Moreover, the current feed-rate scheduling that considers both offline and online methods does not consider the situation that the call of offline data and online prediction data will lead to a decrease in the real-time performance of the CNC system. Further, real-time feed-rate scheduling data tend to dominate the whole interpolation process, thus reducing the effect of the offline feed-rate scheduling data. Hence, based on the tool path with C3 continuity (Cubic Continuously Differentiable), this paper first presents a basic interpolation unit relevant to the S-type interpolation feed-rate profile. Then, an offline local smooth strategy is proposed to smooth the feed-rate profile and reduce the exceeding of kinetic limitations and feed-rate fluctuations caused by frequent acceleration and deceleration. Further, a global online smoothing strategy based on the data generated by offline pre-interpolation is presented. What is more, FIR login and logout conditions are proposed to further smooth the feed-rate profile and improve the real-time performance and machining efficiency. The case study validates that the proposed method performs better in kinetic results compared with the typical offline and FIR methods in both the simulation experiment and actual machining experiments. Especially, in actual processing experiments, the proposed method obtains a 28% reduction in contour errors. Further, the proposed method compared with the FIR method obtains a 15% increase in machining efficiency but only a 4% decrease compared with the typical offline method.

The Crack Diffusion Model: An Innovative Diffusion-Based Method for Pavement Crack Detection

Pavement crack detection is of significant importance in ensuring road safety and smooth traffic flow. However, pavement cracks come in various shapes and forms which exhibit spatial continuity, and algorithms need to adapt to different types of cracks while preserving their continuity. To address these challenges, an innovative crack detection framework, CrackDiff, based on the generative diffusion model, is proposed. It leverages the learning capabilities of the generative diffusion model for the data distribution and latent spatial relationships of cracks across different sample timesteps and generates more accurate and continuous crack segmentation results. CrackDiff uses crack images as guidance for the diffusion model and employs a multi-task UNet architecture to predict mask and noise simultaneously at each sampling step, enhancing the robustness of generations. Compared to other models, CrackDiff generates more accurate and stable results. Through experiments on the Crack500 and DeepCrack pavement datasets, CrackDiff achieves the best performance (F1 = 0.818 and mIoU = 0.841 on Crack500, and F1 = 0.841 and mIoU = 0.862 on DeepCrack).

Improved BIO-Based Chinese Automatic Abstract-Generation Model

With its unique information-filtering function, text summarization technology has become a significant aspect of search engines and question-and-answer systems. However, existing models that include the copy mechanism often lack the ability to extract important fragments, resulting in generated content that suffers from thematic deviation and insufficient generalization. Specifically, Chinese automatic summarization using traditional generation methods often loses semantics because of its reliance on word lists. To address these issues, we proposed the novel BioCopy mechanism for the summarization task. By training the tags of predictive words and reducing the probability distribution range on the glossary, we enhanced the ability to generate continuous segments, which effectively solves the above problems. Additionally, we applied reinforced canonicality to the inputs to obtain better model results, making the model share the sub-network weight parameters and sparsing the model output to reduce the search space for model prediction. To further improve the model’s performance, we calculated the bilingual evaluation understudy (BLEU) score on the English dataset CNN/DailyMail to filter the thresholds and reduce the difficulty of word separation and the dependence of the output on the word list. We fully fine-tuned the model using the LCSTS dataset for the Chinese summarization task and conducted small-sample experiments using the CSL dataset. We also conducted ablation experiments on the Chinese dataset. The experimental results demonstrate that the optimized model can learn the semantic representation of the original text better than other models and performs well with small sample sizes.

Lumbar spine segmentation in MR images: a dataset and a public benchmark

This paper presents a large publicly available multi-center lumbar spine magnetic resonance imaging (MRI) dataset with reference segmentations of vertebrae, intervertebral discs (IVDs), and spinal canal. The dataset includes 447 sagittal T1 and T2 MRI series from 218 patients with a history of low back pain and was collected from four different hospitals. An iterative data annotation approach was used by training a segmentation algorithm on a small part of the dataset, enabling semi-automatic segmentation of the remaining images. The algorithm provided an initial segmentation, which was subsequently reviewed, manually corrected, and added to the training data. We provide reference performance values for this baseline algorithm and nnU-Net, which performed comparably. Performance values were computed on a sequestered set of 39 studies with 97 series, which were additionally used to set up a continuous segmentation challenge that allows for a fair comparison of different segmentation algorithms. This study may encourage wider collaboration in the field of spine segmentation and improve the diagnostic value of lumbar spine MRI.

Predicting adequate segmental lordosis correction in lumbar spinal stenosis patients undergoing oblique lumbar interbody fusion: a focus on the discontinuous segment

PurposeTo identify the factors associated with a correction of the segmental angle (SA) with a total change greater than 10° in each level following minimally invasive oblique lumbar interbody fusion (MIS-OLIF).MethodsPatients with lumbar spinal stenosis who underwent single- or two-level MIS-OLIF were reviewed. Segments with adequate correction of the SA >10° after MIS-OLIF in immediate postoperative radiograph were categorized as discontinuous segments (D segments), whereas those without such improvement were assigned as continuous segments (C segments). Clinical and radiological parameters were compared, and multivariate logistic regression analysis was performed to identify factors associated with SA correction >10° after MIS-OLIF.ResultsOf 211 segments included, 38 segments (18.0%) were classified as D segments. Compared with C segments, D segments demonstrated a significantly smaller preoperative SA (mean ± standard deviation [SD], − 1.1° ± 6.7° vs. 6.6° ± 6.3°, p < 0.001), larger change of SA (mean ± SD, 13.5° ± 3.4° vs. 3.1° ± 3.9°, p < 0.001), and a higher rate of presence of facet effusion (76.3% vs. 48.6%, p = 0.002). Logistic regression revealed preoperative SA (odds ratio (OR) [95% confidence interval (CI)]:0.733 [0.639–0.840], p < 0.001) and facet effusion (OR [95% CI]:14.054 [1.758–112.377], p = 0.027) as significant predictors for >10° SA correction after MIS-OLIF.ConclusionPreoperative kyphotic SA and facet effusion can predict SA correction >10° following MIS-OLIF. For patients with lordotic SA and no preoperative facet effusion, supplemental procedures, such as anterior column release or posterior osteotomy, should be prepared for additional lumbar lordosis correction required for remnant global sagittal imbalance after MIS-OLIF.

Runs of Homozygosity Detection and Selection Signature Analysis for Local Goat Breeds in Yunnan, China.

Runs of Homozygosity (ROH) are continuous homozygous DNA segments in diploid genomes, which have been used to estimate the genetic diversity, inbreeding levels, and genes associated with specific traits in livestock. In this study, we analyzed the resequencing data from 10 local goat breeds in Yunnan province of China and five additional goat populations obtained from a public database. The ROH analysis revealed 21,029 ROH segments across the 15 populations, with an average length of 1.27 Mb, a pattern of ROH, and the assessment of the inbreeding coefficient indicating genetic diversity and varying levels of inbreeding. iHS (integrated haplotype score) was used to analyze high-frequency Single-Nucleotide Polymorphisms (SNPs) in ROH regions, specific genes related to economic traits such as coat color and weight variation. These candidate genes include OCA2 (OCA2 melanosomal transmembrane protein) and MLPH (melanophilin) associated with coat color, EPHA6 (EPH receptor A6) involved in litter size, CDKAL1 (CDK5 regulatory subunit associated protein 1 like 1) and POMC (proopiomelanocortin) linked to weight variation and some putative genes associated with high-altitude adaptability and immune. This study uncovers genetic diversity and inbreeding levels within local goat breeds in Yunnan province, China. The identification of specific genes associated with economic traits and adaptability provides actionable insights for utilization and conservation efforts.

A Simple Preprocessing Method Enhances Machine Learning Application to EEG Data for Differential Diagnosis of Autism

A new pre-processing approach of EEG data to detect topological EEG features has been applied to a continuous segment of artifact-free EEG data lasting 10 minutes in ASCII format derived from 50 ASD children and 50 children with other Neuro-Psychiatric Disorders (NPD), matched for age and male/female ratios. Each EEG is transformed in a triangular matrix of 171 values expressing all reciprocal Manhattan distances among the 19 electrodes of to the international 10-20 system. From this matrix, the minimum spanning tree (MST) is calculated. Electrode identification serial codes sorted according to the decreasing number of links in MST, and the number of links in MST are taken as input vectors for machine learning systems. Machine learning systems have been applied to build up a predictive model to distinguish between the two diagnostic classes (autism vs NPD) following a rigorous validation protocol. The best machine learning system (KNN algorithm) obtained a global accuracy of 93.2% (92.37 % sensitivity and 94.03 % specificity) in differentiating ASD subjects from NPD subjects. The results obtained in this study suggest that, thanks to the new pre-processing method introduced, there is the possibility to discriminate subjects with autism from subjects affected by other psychiatric disorders with a modest computational time reducing the information to 38 figures.

Study on the effect of the application of metallic alloys on continuous casting machine segment drive roller surface

Study on the effect of the application of metallic alloys on continuous casting machine segment drive roller surface

CylinGCN: Cylindrical structures segmentation in 3D biomedical optical imaging by a contour-based graph convolutional network

Cylindrical organs, e.g., blood vessels, airways, and intestines, are ubiquitous structures in biomedical optical imaging analysis. Image segmentation of these structures serves as a vital step in tissue physiology analysis. Traditional model-driven segmentation methods seek to fit the structure by constructing a corresponding topological geometry based on domain knowledge. Classification-based deep learning methods neglect the geometric features of the cylindrical structure and therefore cannot ensure the continuity of the segmentation surface. In this paper, by treating the cylindrical structures as a 3D graph, we introduce a novel contour-based graph neural network for 3D cylindrical structure segmentation in biomedical optical imaging. Our proposed method, which we named CylinGCN, adopts a novel learnable framework that extracts semantic features and complex topological relationships in the 3D volumetric data to achieve continuous and effective 3D segmentation. Our CylinGCN consists of a multiscale 3D semantic feature extractor for extracting inter-frame multiscale semantic features, and a residual graph convolutional network (GCN) contour generator that combines the semantic features and cylindrical topological priors to generate segmentation contours. We tested the CylinGCN framework on two types of optical tomographic imaging data, small animal whole body photoacoustic tomography (PAT) and endoscopic airway optical coherence tomography (OCT), and the results show that CylinGCN achieves state-of-the-art performance. Code will be released at https://github.com/lzc-smu/CylinGCN.git.

Corrosivity study in Usiminas continuous casting machines segments

Corrosivity study in Usiminas continuous casting machines segments

Vertical tearing of subducting plates controlled by geometry and rheology of oceanic plates

Lateral non-uniform subduction is impacted by continuous plate segmentation owing to vertical tearing of the subducting plate. However, the dynamics and physical controls of vertical tearing remain controversial. Here, we employed 3D numerical models to investigate the effects of trench geometry (offset by a transform boundary) and plate rheology (plate age and the magnitude of brittle/plastic strain weakening) on the evolution of shear stress-controlled vertical tearing within a homogenous subducting oceanic plate. Numerical results suggest that the trench offset geometry could result in self-sustained vertical tearing as a narrow shear zone within the intact subducting oceanic plate, and that this process of tearing could operate throughout the entire subduction process. Further, the critical trench offset length for the maturation of vertical tearing is impacted by plate rheology. Comparison between numerical modelling results and natural observations suggests that vertical tearing attributed to trench offset geometry is broadly developed in modern subduction and collision systems worldwide.