Skeletonization Algorithm Research Articles

Research on Automatic Generation of Park Road Network Based on Skeleton Algorithm

Research on Automatic Generation of Park Road Network Based on Skeleton Algorithm

Periconceptional maternal intake of ultra-processed foods, energy and macronutrients the impact on imaging markers of early utero-placental vascular development: The rotterdam periconception cohort

Background & AimsThe quantity and quality of maternal nutrition in the periconception period is an important determinant for embryonic and foetal development and subsequent pregnancy course and outcome. The intake of ultra-processed foods (UPF) has increased worldwide and adverse health outcomes have been reported. However, the impact of UPF intake on the placenta, essential for prenatal nourishment, is unknown. Therefore, we aim to investigate associations between the periconceptional maternal intake of UPF, energy and related macronutrients, and first-trimester utero-placental vascular development. MethodsWe included 214 ongoing pregnancies in the Virtual Placenta study, a subcohort of the Rotterdam periconception cohort. At enrollment, participants filled out a food frequency questionnaire from which we calculated the average daily energy from UPF, total energy intake and macronutrient intake from UPF. At 7-9-11 weeks of gestation, we performed sequential three-dimensional power Doppler ultrasounds of the first-trimester utero-placental vasculature. Virtual Organ Computer-aided AnaLysis (VOCAL) software, Virtual Reality segmentation and a skeletonization algorithm were applied to measure placental volume (PV), utero-placental vascular volume (uPVV) and generate the utero-placental vascular skeleton (uPVS). Absolute vascular morphology was quantified by assigning a morphologic characteristic to each voxel in the uPVS (end-, bifurcation-, crossing- or vessel point) and used to calculate density of vascular branching. Linear mixed models adjusted for confounders were used to investigate associations between maternal intake of UPF, total energy and macronutrients from UPF and PV, uPVV and uPVS characteristics ResultsEnergy intake from UPF and total energy intake were not consistently associated with imaging markers of utero-placental vascular development. . Higher carbohydrate intake of 10g/day from UPF was associated with increased uPVS trajectories (end points (β=0.34, 95%CI=0.07;0.61), bifurcation points (β=0.38, 95%CI=0.05;0.70), vessel points (β=0.957, 95%CI=0.21;1.71). No associations were observed with PV. ConclusionsAgainst our hypothesis, periconceptional maternal intake of UPF and total energy were not convincingly associated with impaired first-trimester utero-placental vascular development. Remarkably, the increased intake of carbohydrates from UPF, which is often considered ‘unhealthy’, is positively associated with first-trimester utero-placental vascular development. Given the complexity of diet, further research should elucidate what underlies these findings to be able to interpret how nutrition may impact utero-placental vascular development in early pregnancy. Clinical trial numberThis study is registered at the Dutch Trial Register (NTR6854).

Detection of maize stem diameter by using RGB-D cameras' depth information under selected field condition.

Stem diameter is a critical phenotypic parameter for maize, integral to yield prediction and lodging resistance assessment. Traditionally, the quantification of this parameter through manual measurement has been the norm, notwithstanding its tedious and laborious nature. To address these challenges, this study introduces a non-invasive field-based system utilizing depth information from RGB-D cameras to measure maize stem diameter. This technology offers a practical solution for conducting rapid and non-destructive phenotyping. Firstly, RGB images, depth images, and 3D point clouds of maize stems were captured using an RGB-D camera, and precise alignment between the RGB and depth images was achieved. Subsequently, the contours of maize stems were delineated using 2D image processing techniques, followed by the extraction of the stem's skeletal structure employing a thinning-based skeletonization algorithm. Furthermore, within the areas of interest on the maize stems, horizontal lines were constructed using points on the skeletal structure, resulting in 2D pixel coordinates at the intersections of these horizontal lines with the maize stem contours. Subsequently, a back-projection transformation from 2D pixel coordinates to 3D world coordinates was achieved by combining the depth data with the camera's intrinsic parameters. The 3D world coordinates were then precisely mapped onto the 3D point cloud using rigid transformation techniques. Finally, the maize stem diameter was sensed and determined by calculating the Euclidean distance between pairs of 3D world coordinate points. The method demonstrated a Mean Absolute Percentage Error (MAPE) of 3.01%, a Mean Absolute Error (MAE) of 0.75mm, a Root Mean Square Error (RMSE) of 1.07mm, and a coefficient of determination (R²) of 0.96, ensuring accurate measurement of maize stem diameter. This research not only provides a new method of precise and efficient crop phenotypic analysis but also offers theoretical knowledge for the advancement of precision agriculture.

Automated Pavement Condition Index Assessment with Deep Learning and Image Analysis: An End-to-End Approach.

The degradation of road pavements due to environmental factors is a pressing issue in infrastructure maintenance, necessitating precise identification of pavement distresses. The pavement condition index (PCI) serves as a critical metric for evaluating pavement conditions, essential for effective budget allocation and performance tracking. Traditional manual PCI assessment methods are limited by labor intensity, subjectivity, and susceptibility to human error. Addressing these challenges, this paper presents a novel, end-to-end automated method for PCI calculation, integrating deep learning and image processing technologies. The first stage employs a deep learning algorithm for accurate detection of pavement cracks, followed by the application of a segmentation-based skeleton algorithm in image processing to estimate crack width precisely. This integrated approach enhances the assessment process, providing a more comprehensive evaluation of pavement integrity. The validation results demonstrate a 95% accuracy in crack detection and 90% accuracy in crack width estimation. Leveraging these results, the automated PCI rating is achieved, aligned with standards, showcasing significant improvements in the efficiency and reliability of PCI evaluations. This method offers advancements in pavement maintenance strategies and potential applications in broader road infrastructure management.

Abstract 3516: AI-derived features of 3D glandular networks are associated with likelihood of biochemical recurrence post radical prostatectomy

Abstract Introduction: Prostate cancer is the most common cancer in American men and is mainly diagnosed from two-dimensional histology sections. Recently, three-dimensional (3D) prostate histopathology has emerged as a valuable strategy to enhance the understanding of the disease by increasing microscopic sampling of specimens. This strategy enhances the microscopic examination of specimens by facilitating the analysis of the volumetric shape of cells and glands, as well as the pathway structures through entire biopsies and other morphological features. These aspects hold the potential to be linked with biochemical recurrence (BCR) outcomes. This study highlights the potential of features derived from the 3D structure of glands in samples of patients with prostate cancer for identifying patients who are at higher risk of BCR. Methods: Radical prostatectomy specimens were collected from 50 patients along with 5-year postoperative BCR follow-up from the University of Washington. A total of 118 ex vivo whole-biopsy 3D pathology images were obtained from RP by an open-top light-sheet microscopy platform. Images were converted into a synthetic CK8 immunofluorescence dataset using an image-sequence translation model. From these 3D images, a glandular structure mask was generated by a computer-vision algorithm. The 3D structure of prostate glands was modeled through a Tree-structure Extraction Algorithm for Accurate and Robust Skeletons (TEASER), which provides a simplification of the length of the gland, its shape, and its pathway. 91 features were then extracted from the glandular skeleton architecture. A 3-fold cross-validation approach was used in which patients were divided into training (D1) and test (D2) sets of 33 and 17 patients, respectively. Fisher’s score, used to select features based on their correlation to the outcome, was applied to D1 to identify the top 5 features most correlated with BCR. We then utilized these features to train a K-Nearest Neighbors classifier on D1, aimed at predicting whether a patient would experience BCR. The model’s performance was evaluated on D2 and averaged across all folds. Results: The classifier was able to accurately differentiate between BCR+ and BCR- patients, achieving an average ROCAUC of 0.73 (±0.04) in D2. Significant differences in 3D skeletal features were observed between BCR+ and BCR- groups (p<0.05). Among the five selected features, three were related to the length and diameter of the gland, with the longest glands observed in BCR- patients. The remaining two features pertained to gland curvature and tortuosity, revealing more twisted glands in BCR+ patients. Conclusion: Features derived from a skeleton model of the 3D glandular structure of biopsies showed promise in identifying patients with prostate cancer who are at a higher risk of BCR. Additional, multi-site independent validation of these findings is warranted. Citation Format: Jennifer Salguero, Hisham Abdeltawab, Kamal Hammouda, German Corredor, Sarah Hawley, Pushkar Mutha, Rohan Dhamdhere, Sebastian Medina, Tilak Pathak, Kevin Bishop, Rob Serafin, Tuomas Mirtti, Priti Lal, Lawrence True, Jonathan Liu, Anant Madabhushi. AI-derived features of 3D glandular networks are associated with likelihood of biochemical recurrence post radical prostatectomy [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 3516.

Morphologic development of the first-trimester utero-placental vasculature is positively associated with embryonic and fetal growth: the Rotterdam Periconception Cohort.

Is morphologic development of the first-trimester utero-placental vasculature associated with embryonic growth and development, fetal growth, and birth weight percentiles? Using the utero-placental vascular skeleton (uPVS) as a new imaging marker, this study reveals morphologic development of the first-trimester utero-placental vasculature is positively associated with embryonic growth and development, fetal growth, and birth weight percentiles. First-trimester development of the utero-placental vasculature is associated with placental function, which subsequently impacts embryonic and fetal ability to reach their full growth potential. The attribution of morphologic variations in the utero-placental vascular development, including the vascular structure and branching density, on prenatal growth remains unknown. This study was conducted in the VIRTUAL Placental study, a subcohort of 214 ongoing pregnancies, embedded in the prospective observational Rotterdam Periconception Cohort (Predict study). Women were included before 10 weeks gestational age (GA) at a tertiary referral hospital in The Netherlands between January 2017 and March 2018. We obtained three-dimensional power Doppler volumes of the gestational sac including the embryo and the placenta at 7, 9, and 11 weeks of gestation. Virtual Reality-based segmentation and a recently developed skeletonization algorithm were applied to the power Doppler volumes to generate the uPVS and to measure utero-placental vascular volume (uPVV). Absolute vascular morphology was quantified by assigning a morphologic characteristic to each voxel in the uPVS (i.e. end-, bifurcation-crossing-, or vessel point). Additionally, total vascular length (mm) was calculated. The ratios of the uPVS characteristics to the uPVV were calculated to determine the density of vascular branching. Embryonic growth was estimated by crown-rump length and embryonic volume. Embryonic development was estimated by Carnegie stages. Fetal growth was measured by estimated fetal weight in the second and third trimester and birth weight percentiles. Linear mixed models were used to estimate trajectories of longitudinal measurements. Linear regression analysis with adjustments for confounders was used to evaluate associations between trajectories of the uPVS and prenatal growth. Groups were stratified for conception method (natural/IVF-ICSI conceptions), fetal sex (male/female), and the occurrence of placenta-related complications (yes/no). Increased absolute vascular morphologic development, estimated by positive random intercepts of the uPVS characteristics, is associated with increased embryonic growth, reflected by crown-rump length (endpoints β = 0.017, 95% CI [0.009; 0.025], bifurcation points β = 0.012, 95% CI [0.006; 0.018], crossing points β = 0.017, 95% CI [0.008; 0.025], vessel points β = 0.01, 95% CI [0.002; 0.008], and total vascular length β = 0.007, 95% CI [0.003; 0.010], and similarly with embryonic volume and Carnegie stage, all P-values ≤ 0.01. Density of vascular branching was negatively associated with estimated fetal weight in the third trimester (endpoints: uPVV β = -94.972, 95% CI [-185.245; -3.698], bifurcation points: uPVV β = -192.601 95% CI [-360.532; -24.670]) and birth weight percentiles (endpoints: uPVV β = -20.727, 95% CI [-32.771; -8.683], bifurcation points: uPVV β -51.097 95% CI [-72.257; -29.937], and crossing points: uPVV β = -48.604 95% CI [-74.246; -22.961])), all P-values < 0.05. After stratification, the associations were observed in natural conceptions specifically. Although the results of this prospective observational study clearly demonstrate associations between first-trimester utero-placental vascular morphologic development and prenatal growth, further research is required before we can draw firm conclusions about a causal relationship. Our findings support the hypothesis that morphologic variations in utero-placental vascular development play a role in the vascular mechanisms involved in embryonic and fetal growth and development. Application of the uPVS could benefit our understanding of the pathophysiology underlying placenta-related complications. Future research should focus on the clinical applicability of the uPVS as an imaging marker for the early detection of fetal growth restriction. This research was funded by the Department of Obstetrics and Gynecology of the Erasmus MC, University Medical Centre, Rotterdam, The Netherlands. There are no conflicts of interest. Registered at the Dutch Trial Register (NTR6854).

Reconstructing microvascular network skeletons from 3D images: What is the ground truth?

Structural changes to microvascular networks are increasingly highlighted as markers of pathogenesis in a wide range of disease, e.g. Alzheimer’s disease, vascular dementia and tumour growth. This has motivated the development of dedicated 3D imaging techniques, alongside the creation of computational modelling frameworks capable of using 3D reconstructed networks to simulate functional behaviours such as blood flow or transport processes. Extraction of 3D networks from imaging data broadly consists of two image processing steps: segmentation followed by skeletonisation. Much research effort has been devoted to segmentation field, and there are standard and widely-applied methodologies for creating and assessing gold standards or ground truths produced by manual annotation or automated algorithms.The Skeletonisation field, however, lacks widely applied, simple to compute metrics for the validation or optimisation of the numerous algorithms that exist to extract skeletons from binary images. This is particularly problematic as 3D imaging datasets increase in size and visual inspection becomes an insufficient validation approach. In this work, we first demonstrate the extent of the problem by applying 4 widely-used skeletonisation algorithms to 3 different imaging datasets. In doing so we show significant variability between reconstructed skeletons of the same segmented imaging dataset. Moreover, we show that such a structural variability propagates to simulated metrics such as blood flow. To mitigate this variability we introduce a new, fast and easy to compute super metric that compares the volume, connectivity, medialness, bifurcation point identification and homology of the reconstructed skeletons to the original segmented data. We then show that such a metric can be used to select the best performing skeletonisation algorithm for a given dataset, as well as to optimise its parameters. Finally, we demonstrate that the super metric can also be used to quickly identify how a particular skeletonisation algorithm could be improved, becoming a powerful tool in understanding the complex implication of small structural changes in a network.

Identification of wheel track in the wheat field

Agriculture machinery navigating along permanent traffic lanes in the farmland may avoid causing extensive soil compaction. However, the permanent traffic lanes are frequently covered up or eliminated by following tillage practices. It is necessary to identify the wheel tracks designed as permanent traffic lanes in order to ensure the agriculture machinery travels along the designated wheel tracks when cultivating the field. This study proposed an identification method of wheel tracks based on the morphological characteristics of wheel tracks and the environmental conditions around the wheel tracks in the wheat fields. The proposed method first utilized the maximum interclass variance to identify the contours of the main part of the wheel track and the shadow regions around the wheel track’s edges. The main part of the wheel tracks was then separated from interference pixels by moving the centerline of the main part of the wheel track, which was derived by skeleton algorithm and curve fitting, towards the right or left edge of the wheel track at a specific distance. In a morphological opening operation, specific linear and circular structural elements were used to segment the shadow regions along the edge of the wheel track. The remaining wheel track was finally recognized by computing the complement of the region identified. After achieving the segmentation of wheel tracks, many reference points near the outside of the wheel track edge in the original image were chosen as fiducial points for evaluating the differences between the actual value and the recognized wheel track edge. The evaluation was based on computing the root mean squared error (RMSE) and the mean absolute error (MAE) of coordinates of reference points and recognized wheel track edge. The results showed that the largest RMSE and MAE were 24.01 pixels (0.0045 m) and 17.32 pixels (0.0032 m), respectively. The low values of RMSE and MAE reveal that the accuracy of the algorithm developed in this study is high, and using this algorithm may segment the wheel track in the wheat field accurately.

Skeleton Ground Truth Extraction: Methodology, Annotation Tool and Benchmarks

AbstractSkeleton Ground Truth (GT) is critical to the success of supervised skeleton extraction methods, especially with the popularity of deep learning techniques. Furthermore, we see skeleton GTs used not only for training skeleton detectors with Convolutional Neural Networks (CNN), but also for evaluating skeleton-related pruning and matching algorithms. However, most existing shape and image datasets suffer from the lack of skeleton GT and inconsistency of GT standards. As a result, it is difficult to evaluate and reproduce CNN-based skeleton detectors and algorithms on a fair basis. In this paper, we present a heuristic strategy for object skeleton GT extraction in binary shapes and natural images. Our strategy is built on an extended theory of diagnosticity hypothesis, which enables encoding human-in-the-loop GT extraction based on clues from the target’s context, simplicity, and completeness. Using this strategy, we developed a tool, SkeView, to generate skeleton GT of 17 existing shape and image datasets. The GTs are then structurally evaluated with representative methods to build viable baselines for fair comparisons. Experiments demonstrate that GTs generated by our strategy yield promising quality with respect to standard consistency, and also provide a balance between simplicity and completeness.

Development and assessment of a methodology for abstraction of topology optimization results to enable the substitution of optimized beams

Improving mechanical topology optimization (TO) results by substituting biomimetic beams is one possibility to achieve designs of mechanical components that are highly sustainable and show good mechanical performance. Because of their geometric complexity, such designs were found to be well-suited for production by laser additive manufacturing. One obstacle of incorporating biomimetics beams in TO designs is the lack of detailed design methodologies. Röver et al. [“Methodology for integrating biomimetic beams in abstracted topology optimization results,” in Proceedings of the ASME 2022 International Mechanical Engineering Congress and Exposition. Volume 4: Biomedical and Biotechnology; Design, Systems, and Complexity Columbus, OH, 30 October–3 November (ASME, New York, 2022)] proposed a corresponding design concept. Building on their concept, we present in this work a detailed methodology for abstraction of TO results to a design consisting of ball nodes and cylindrical beams. Using such an auxiliary design, the internal forces and moments of the beams can be evaluated to allow for the substitution of suitable biomimetic beams to generate biomimetic component designs in a next step. We present a skeletonization algorithm based on the potential field approach. Using the skeletonization and an additional analysis of the dimensions of the beams in the TO result, the algorithm develops an auxiliary design of the original TO result. The final algorithm was applied to three common TO results to obtain one auxiliary component design each. The developed algorithm was found to generate abstractions that were well-suited for use in the methodology proposed in Röver et al. [“Methodology for integrating biomimetic beams in abstracted topology optimization results,” in Proceedings of the ASME 2022 International Mechanical Engineering Congress and Exposition. Volume 4: Biomedical and Biotechnology; Design, Systems, and Complexity Columbus, OH, 30 October–3 November (ASME, New York, 2022)], because internal forces and moments in the abstracted beams could be evaluated with less effort. Therefore, our work contributes to a detailed design methodology for biomimetic mechanical components in the field of design for additive manufacturing.

Skeleting of Low-Contrast Noisy Halftone Images

The problem of forming the skeletons of halftone images with two-mode brightness histograms under conditions of changing contrast and noise is considered. On such histograms, one mode corresponds to the objects, and the other to the background. Thanks to this feature, images are relatively easy to binarize and then skeletonize. The skeleton of a region uniform in brightness is a set of thin (limited by one-pixel) connected lines enclosed within this region and compactly describing its structure. Under conditions of high contrast and low noise on the original halftone image, binary skeletonization algorithms are widely used. They are relatively simple and can be resistant to multiplicative noise that appears at the boundaries of the regions after binarization. However, when the contrast is reduced and the noise of the original halftone image is increased, the skeletons formed by such algorithms are destroyed under the influence of additive noise, which manifests itself in the depth of the regions of the skeletonized binary image. To reduce skeletonization errors in such cases, algorithms based on preliminary low-pass filtering of the original grayscale image are used. To increase the stability of the skeletons of halftone images with a two-mode brightness histogram to noise, the article proposes a skeletonization model that takes into account the presence of multiplicative and additive noise components in a binary skeletonized image. Taking this model into account, a skeletonization algorithm has been developed, which takes into account the distortions in the shapes of the areas of the skeletonized binary image as a result of low-frequency filtering of the original halftone image and allows to reduce errors in the skeletonization of halftone images.

Segmentation and thickness calculation of glomerular basement membrane using RADS-Net in glomerular microscopic images

Glomerular ultrastructure needs to be observed under transmission electron microscopy during pathological diagnosis of renal biopsies. Evaluating the glomerular basement membranes (GBM) is necessary for the renal pathological diagnosis. There is little work that performs well for full-flow automated analysis of GBM in whole-glomerular electron microscopy images, including GBM segmentation, location, and thickness calculation. To address this problem, we propose a network architecture, RADS-Net, whose segmentation module combines the advantages of ViT and CNN to achieve better performance in GBM and electron dense deposit (EDD) segmentation tasks. It also contains an uncertainty evaluation module and a location module to exclude unsuitable regions for measurement and extract accurate GBM contours. We propose a GBM thickness calculation method based on statistical learning method, skeletonization algorithm, and improved region growth method. Systematic experiments demonstrate that our proposed segmentation network achieves superior accuracy compared to other competing methods. Additionally, the computed GBM thickness, after an automatic screening of suitable fields of view, meets the requirements for pathology evaluation.

Continuous identification of the tea shoot tip and accurate positioning of picking points for a harvesting from standard plantations.

To address the current problems of large positioning error, low picking efficiency, and high cost of tea shoot picking, a continuous and precise harvesting scheme for tea shoots based on a two-dimensional (2D) perspective is designed in this study. A high-speed harvesting method for tea shoots in a standardized tea plantation assembly line type was proposed. First, a 2D view recognition model of tea shoot tips in a multi-disturbance environment was constructed, and accurate picking point coordinates were determined by combining a skeleton algorithm and curve growth. To avoid the losses of recognition accuracy caused by the mistaken clamping of blades and vibrations during harvester operations, accurate control of the harvester was realized by combining path planning and the S-curve speed control function. The recognition accuracy for the verification set of the recognition model was 99.9%, and the mean average precision (0.5:0.95) value was 0.97. The test results show that the error between the actual picking point position and the position determined by the model was within ± 3 mm, and the picking success rate was 83.6%. Therefore, we can realize fast and accurate picking of tea shoots and lay the foundation for continuous tea picking in the future by simplifying the identification and picking process.

Classifying Poor Postures of the Neck and Spine in Computer Work by Using Image and Skeleton Analysis

When using a desktop computer, people tend to adopt postures that are detrimental to their bodies, such as text neck and the L-posture of leaning forward with their buttocks out and their shoulders against the backrest of the chair. These two postures cause chronic problems by bending the cervical and thoracic spines and can have detrimental effects on the body. While there have been many studies on text neck posture, there were limited studies on classifying these two postures together, and there are limitations to the accuracy of their classification. To address these limitations, we propose an algorithm for classifying good posture, text neck posture, and L-posture, the latter two of which may negatively affect the body when using a desktop computer. The proposed algorithm utilizes a skeleton algorithm to calculate angles from images of the user’s lateral posture, and then classifies the three postures based on the angle values. If there is sufficient space next to the computer, the method can be implemented anywhere, and classification can be performed at low cost. The experimental results showed a high accuracy rate of 97.06% and an F1-score of 95.23%; the L posture was classified with 100% accuracy.

Evaluation of Skeletonization Techniques for 2D Binary Images

In the realm of modern image processing, the emphasis often lies on engineering-based approaches rather than scientific solutions to address diverse practical problems. One prevalent task within this domain involves the skeletonization of binary images. Skeletonization is a powerful process for extracting the skeleton of objects located in digital binary images. This process is widely employed for automating many tasks in numerous fields such as pattern recognition, robot vision, animation, and image analysis. The existing skeletonization techniques are mainly based on three approaches: boundary erosion, distance coding, and Voronoi diagram for identifying an approximate skeleton. In this work, we present an empirical evaluation of a set of well-known techniques and report our findings. We specifically deal with computing skeletons in 2d binary images by selecting different approaches and evaluating their effectiveness. Visual evaluation is the primary method used to showcase the performance of selected skeletonization algorithms. Due to the absence of a definitive definition for the "true" skeleton of a digital object, accurately assessing the effectiveness of skeletonization algorithms poses a significant research challenge. Although researchers have attempted quantitative assessments, these measures are typically customized for specific domains and may not be suitable for our current work. The experimental results shown in this work illustrate the performance of the three main approaches in applying skeletonization with respect to different perspectives.

Edge-Supervised Linear Object Skeletonization for High-Speed Camera.

This paper presents a high-speed skeletonization algorithm for detecting the skeletons of linear objects from their binary images. The primary objective of our research is to achieve rapid extraction of the skeletons from binary images while maintaining accuracy for high-speed cameras. The proposed algorithm uses edge supervision and a branch detector to efficiently search inside the object, avoiding unnecessary computation on irrelevant pixels outside the object. Additionally, our algorithm addresses the challenge of self-intersections in linear objects with a branch detection module, which detects existing intersections and initializes new searches on emerging branches when necessary. Experiments on various binary images, such as numbers, ropes, and iron wires, demonstrated the reliability, accuracy, and efficiency of our approach. We compared the performance of our method with existing skeletonization techniques, showing its superiority in terms of speed, especially for larger image sizes.

Computed-tomography-based discrete fracture-matrix modeling: An integrated framework for deriving fracture networks

Investigating the topology of natural fractures and associated flow properties allow for better understanding of fluid flow patterns and mechanical response in fractured reservoirs, key points in the energy supply through geothermal systems, or in the reduction of net CO2 emissions through carbon capture and storage (CCS) practices. However, limited information on the fracture networks is available from seismic data, which poses as a challenge in the accurate modeling of fractured reservoirs. This work proposes an integrated framework for obtaining fracture networks with real aperture distributions, based on X-ray computed tomography (CT) images of core plugs from naturally fractured rock formations. We adapt and employ segmentation, skeletonization and geometry generation algorithms that are available in open-source libraries and licensed software. In addition, we investigate the contributions of fractures and porous matrix to the equivalent permeability tensor through fluid flow simulations considering three different scenarios for the matrix permeability. This work is novel in the coordination of the steps involved in the generation of discrete fracture-matrix (DFM) models from CT-scans of core plugs.

3D Visualization, Skeletonization and Branching Analysis of Blood Vessels in Angiogenesis.

Angiogenesis is the process of new blood vessels growing from existing vasculature. Visualizing them as a three-dimensional (3D) model is a challenging, yet relevant, task as it would be of great help to researchers, pathologists, and medical doctors. A branching analysis on the 3D model would further facilitate research and diagnostic purposes. In this paper, a pipeline of vision algorithms is elaborated to visualize and analyze blood vessels in 3D from formalin-fixed paraffin-embedded (FFPE) granulation tissue sections with two different staining methods. First, a U-net neural network is used to segment blood vessels from the tissues. Second, image registration is used to align the consecutive images. Coarse registration using an image-intensity optimization technique, followed by finetuning using a neural network based on Spatial Transformers, results in an excellent alignment of images. Lastly, the corresponding segmented masks depicting the blood vessels are aligned and interpolated using the results of the image registration, resulting in a visualized 3D model. Additionally, a skeletonization algorithm is used to analyze the branching characteristics of the 3D vascular model. In summary, computer vision and deep learning is used to reconstruct, visualize and analyze a 3D vascular model from a set of parallel tissue samples. Our technique opens innovative perspectives in the pathophysiological understanding of vascular morphogenesis under different pathophysiological conditions and its potential diagnostic role.

Positional assessment of lower third molar and mandibular canal using explainable artificial intelligence

ObjectiveThe aim of this study is to automatically assess the positional relationship between lower third molars (M3i) and the mandibular canal (MC) based on the panoramic radiograph(s) (PR(s)). Material and methodsA total of 1444 M3s were manually annotated and labeled on 863 PRs as a reference. A deep-learning approach, based on MobileNet-V2 combination with a skeletonization algorithm and a signed distance method, was trained and validated on 733 PRs with 1227 M3s to classify the positional relationship between M3i and MC into three categories. Subsequently, the trained algorithm was applied to a test set consisting of 130 PRs (217 M3s). Accuracy, precision, sensitivity, specificity, negative predictive value, and F1-score were calculated. ResultsThe proposed method achieved a weighted accuracy of 0.951, precision of 0.943, sensitivity of 0.941, specificity of 0.800, negative predictive value of 0.865 and an F1-score of 0.938. ConclusionAI-enhanced assessment of PRs can objectively, accurately, and reproducibly determine the positional relationship between M3i and MC. Clinical significanceThe use of such an explainable AI system can assist clinicians in the intuitive positional assessment of lower third molars and mandibular canals. Further research is required to automatically assess the risk of alveolar nerve injury on panoramic radiographs.

Assessment of First-Trimester Utero-Placental Vascular Morphology by 3D Power Doppler Ultrasound Image Analysis Using a Skeletonization Algorithm: The Rotterdam Periconception Cohort

ABSTRACT Abnormal uteroplacental vascular development is one of the primary causes of the major disorders of pregnancy. Identification of this aberrant development in the preconception stage and in the first trimester, when measures to prevent complications are most effective, remains a challenge. Recent advances in offline imaging processing and Doppler techniques have allowed ultrasonographic analysis of placental volume (PV) and uteroplacental vascular volume (uPVV) in the first trimester. The ability of ultrasound to assess specific vascular morphology is not fully understood. The aim of this study was to investigate morphologic development of the first-trimester human uteroplacental vasculature in vivo using 3-dimensional power Doppler ultrasound and advanced image processing to generate the uteroplacental vascular skeleton (uPVS). Data were obtained from the VIRTUAL Placenta study, an ongoing observational study including adult women carrying a singleton pregnancy at less than 10-week gestational age (GA). At least 2 study visits were scheduled in the first trimester, at 7, 9, and 11 weeks' GA. The image quality of ultrasound scans was scored on a 4-point scale ranging between 0 (optimal) and 3 (unusable), and PV was measured using VOCAL software. The uPVV was measured using a virtual reality desktop system with the V-scope volume-rendering application. A skeletonization algorithm was then applied to the uPVV segmentation to generate the uPVS, and 7 morphological uPVS characteristics were used to determine the density of vascular branching within the placenta. A total of 214 women from the VIRTUAL cohort were eligible for inclusion, and 81% of all ultrasound data were of sufficient quality for generation of the uPVS. The distribution of endpoints, vessel points, bifurcation points, and crossing points in the uPVS and the distribution of uPVS characteristics per cm3 uPVV show consistent morphologic patterns throughout the first trimester. Moderate to strong positive correlations between uPVS characteristics and PV and strong positive correlations between uPVS characteristics and uPVV were identified. All uPVS characteristics increased significantly through the first trimester, except for uPVS average length, which remained constant between 7, 9, and 11 weeks' GA. When stratifying for placenta-related complications among the cohort, no significant differences in the uPVS characteristics were seen at 7 weeks' GA (n = 94). However, at 9 weeks (n = 170), there were significantly fewer number of vessel points (P = 0.040), bifurcation points (P = 0.050), crossing points (P = 0.020), and a shorter total network length (P = 0.023) among pregnancies with placenta-related complications. At 11 weeks' GA (n = 129), uPVS average vascular thickness was significantly lower in pregnancies with placenta-related complications (P = 0.007), but no other uPVS differences were observed. At 11 weeks' GA, pregnancies with placenta-related complications were found to have an increased density of vascular branching in the uPVV. The results of this study demonstrate a quantitative morphologic analysis of the human uteroplacental vasculature in the first trimester of pregnancy and key differences in the vascular morphology between pregnancies with and without placenta-related complications.