Search

Using deep learning or artificial intelligence (AI) in research with animals is a new interdisciplinary area of research. In this study, we have explored the potential of thermal imaging and AI in pig research. Thermal cameras play a vital role in obtaining and collecting a large amount of data, and AI has the capabilities of processing and extracting valuable information from these data. The amount of data collected using thermal imaging is huge, and automation techniques are therefore crucial to find a meaningful interpretation of the changes in temperature. In this paper, we present a complete pipeline to extract temperature automatically from a selected Region of Interest (ROI). This system consists of three stages: the first one checks whether the ROI is completely visible to observe the thermal temperature, and then the second stage uses an encoder–decoder structure of a convolution neural network to segment the ROI, if the condition was met at stage one. In the last stage, the maximum temperature is extracted and saved in an external file. The segmentation model showed good performance, with a mean Pixel Class accuracy of 92.3%, and a mean Intersection over Union of 87.1%. The extracted temperature observed by the model entirely matched the manually observed temperature. The system showed reliable results to be used independently without human intervention to determine the temperature in the selected ROI in pigs. [ABSTRACT FROM AUTHOR]

With the advances in technology, assistive medical systems are emerging with rapid growth and helping healthcare professionals. The proactive diagnosis of diseases with artificial intelligence (AI) and its aligned technologies has been an exciting research area in the last decade. Doctors usually detect tuberculosis (TB) by checking the lungs' X-rays. Classification using deep learning algorithms is successfully able to achieve accuracy almost similar to a doctor in detecting TB. It is found that the probability of detecting TB increases if classification algorithms are implemented on segmented lungs instead of the whole X-ray. The paper's novelty lies in detailed analysis and discussion of U-Net + + results and implementation of U-Net + + in lung segmentation using X-ray. A thorough comparison of U-Net + + with three other benchmark segmentation architectures and segmentation in diagnosing TB or other pulmonary lung diseases is also made in this paper. To the best of our knowledge, no prior research tried to implement U-Net + + for lung segmentation. Most of the papers did not even use segmentation before classification, which causes data leakage. Very few used segmentations before classification, but they only used U-Net, which U-Net + + can easily replace because accuracy and mean_iou of U-Net + + are greater than U-Net accuracy and mean_iou , discussed in results, which can minimize data leakage. The authors achieved more than 98% lung segmentation accuracy and mean_iou 0.95 using U-Net + + , and the efficacy of such comparative analysis is validated. [ABSTRACT FROM AUTHOR]

Road scene segmentation is an integral part of the Intelligent Transport System (ITS) for precise interpretation of the environment and safer vehicle navigation. Traditional segmentation methods have faced difficulties in meeting the requirements of unstructured and complex image segmentation. Therefore, the Deep-Neural Network (DNN) plays a significant role in effectively segmenting images with multiple classes in an unstructured environment. In this work, semantic segmentation models such as U-net, LinkNet, FPN, and PSPNet are updated to use classification networks such as VGG19, Resnet50, Efficientb7, MobilenetV2, and Inception V3 as pre-trained backbone architectures, and the performance of each updated model is compared with the unstructured Indian Driving-Lite (IDD-Lite) dataset. In order to improve segmentation performance, a stacking ensemble approach is proposed to combine the predictions of a semantic segmentation model across different backbone architectures using a simple grid search method. Thus, four ensemble models are formed and analyzed on the IDD-Lite dataset. The two metrics Intersection over Union (IoU or Jaccard index) and Dice coefficient (F1 score) are used to assess the segmentation performance of each ensemble model. The results show that an ensemble of U-net with different backbone architectures is more efficient than other ensemble models. This model has achieved 73.12% and 76.67%, respectively, in IoU and F1 scores. [ABSTRACT FROM AUTHOR]

Cracks in concrete structures are one of the most important indicators of structural damage, and it is a necessity to detect and measure cracks for ensuring safety and integrity of concrete structures. The widely practised approach in inspecting the structures is by performing visual inspections followed by manual estimation of crack widths. This approach is not only time-consuming, laborious, and time-intensive but also prone to subjective errors and inefficient. To address these issues, we propose a novel deep learning framework for detecting cracks and then estimating crack widths in concrete surface images. Our framework handles both small- and large-sized images and provides a prediction of crack width at locations specified by the user. The proposed framework uses Attention Recurrent Residual U-Net (Attention R2U-Net) with Random Forest regressor to predict crack width with the mean prediction error of ±0.31 mm for crack widths varying from 0 to 8.95 mm and produces the lowest absolute maximum error of 1.3 mm. Our model has a coefficient of determination (R 2) of 0.91, showing a non-linear mapping function with low prediction errors. We compare our model with a combination of four other segmentation models and regression models. Our proposed model has superior performance compared to other models, and one can easily adopt our framework to a variety of Structural Health Monitoring applications using Internet of Things sensors. [ABSTRACT FROM AUTHOR]

The present study is the first to examine account-based tracking data of Portuguese online lottery players comprising the gambling activity of all active players over a one-year period (N = 154,585). The main research goal was the identification of groups or segments of players by their engagement levels (high, neutral, low) and to assess preferences in product category with the use of CHAID (Chi-Square Automatic Interaction Detection) segmentation models, based on expenditure and sociodemographic variables. Findings showed that (1) age was found to be the most influential differentiating variable in player segmentation and had a positive correlation with expenditures and wagers, (2) gender was the second most influential variable (males represented 78.7% of players), (3) education the third most influential variable and had a negative correlation with expenditure, and (4) region was the least relevant variable. The models generated several players segments that engaged in different games. Older males (54–64 years; ≥ 65 years) were the most engaged overall. Younger males (18–34 years) were the least engaged but showed preferences for lotto as did females (35–49 years). Lower educated males and older males (49 years+) with a high school education were the most engaged in instant lottery games. These findings show that Portuguese lottery players can be grouped into several segments with distinct demographic characteristics and corresponding engagement levels. These findings help support more effective marketing segmentation and will help in the targeting of responsible gambling approaches. [ABSTRACT FROM AUTHOR]

Road scene segmentation is an integral part of the Intelligent Transport System (ITS) for precise interpretation of the environment and safer vehicle navigation. Traditional segmentation methods have faced difficulties in meeting the requirements of unstructured and complex image segmentation. Therefore, the Deep-Neural Network (DNN) plays a significant role in effectively segmenting images with multiple classes in an unstructured environment. In this work, semantic segmentation models such as U-net, LinkNet, FPN, and PSPNet are updated to use classification networks such as VGG19, Resnet50, Efficientb7, MobilenetV2, and Inception V3 as pre-trained backbone architectures, and the performance of each updated model is compared with the unstructured Indian Driving-Lite (IDD-Lite) dataset. In order to improve segmentation performance, a stacking ensemble approach is proposed to combine the predictions of a semantic segmentation model across different backbone architectures using a simple grid search method. Thus, four ensemble models are formed and analyzed on the IDD-Lite dataset. The two metrics Intersection over Union (IoU or Jaccard index) and Dice coefficient (F1 score) are used to assess the segmentation performance of each ensemble model. The results show that an ensemble of U-net with different backbone architectures is more efficient than other ensemble models. This model has achieved 73.12% and 76.67%, respectively, in IoU and F1 scores.

In this paper we reformulate a class of non-linear variational models for global and selective image segmentation and obtain convergent multigrid solutions. In contrast, non-linear multigrid schemes do not converge for these problems with strong non-linearity and non-smoothness (jumps). Our new approach is to reformulate the non-linear models, using splitting techniques, to generate linear models in a higher dimension which are easier to solve and amenable to the linear multigrid framework. Although splitting techniques are well studied in isolation, direct application of a splitting idea is not sufficient and it is the combination of two splitting approaches and linear multigrid theory approaches which results in a highly effective multigrid algorithm. Numerical results demonstrate the fast convergence of the new multigrid methods. [ABSTRACT FROM AUTHOR]

Abstract Background Cell-scaffold contact measurements are derived from pairs of co-registered volumetric fluorescent confocal laser scanning microscopy (CLSM) images (z-stacks) of stained cells and three types of scaffolds (i.e., spun coat, large microfiber, and medium microfiber). Our analysis of the acquired terabyte-sized collection is motivated by the need to understand the nature of the shape dimensionality (1D vs 2D vs 3D) of cell-scaffold interactions relevant to tissue engineers that grow cells on biomaterial scaffolds. Results We designed five statistical and three geometrical contact models, and then down-selected them to one from each category using a validation approach based on physically orthogonal measurements to CLSM. The two selected models were applied to 414 z-stacks with three scaffold types and all contact results were visually verified. A planar geometrical model for the spun coat scaffold type was validated from atomic force microscopy images by computing surface roughness of 52.35 nm ±31.76 nm which was 2 to 8 times smaller than the CLSM resolution. A cylindrical model for fiber scaffolds was validated from multi-view 2D scanning electron microscopy (SEM) images. The fiber scaffold segmentation error was assessed by comparing fiber diameters from SEM and CLSM to be between 0.46% to 3.8% of the SEM reference values. For contact verification, we constructed a web-based visual verification system with 414 pairs of images with cells and their segmentation results, and with 4968 movies with animated cell, scaffold, and contact overlays. Based on visual verification by three experts, we report the accuracy of cell segmentation to be 96.4% with 94.3% precision, and the accuracy of cell-scaffold contact for a statistical model to be 62.6% with 76.7% precision and for a geometrical model to be 93.5% with 87.6% precision. Conclusions The novelty of our approach lies in (1) representing cell-scaffold contact sites with statistical intensity and geometrical shape models, (2) designing a methodology for validating 3D geometrical contact models and (3) devising a mechanism for visual verification of hundreds of 3D measurements. The raw and processed data are publicly available from https://isg.nist.gov/deepzoomweb/data/ together with the web -based verification system.

Background: Cell-scaffold contact measurements are derived from pairs of co-registered volumetric fluorescent confocal laser scanning microscopy (CLSM) images (z-stacks) of stained cells and three types of scaffolds (i.e., spun coat, large microfiber, and medium microfiber). Our analysis of the acquired terabyte-sized collection is motivated by the need to understand the nature of the shape dimensionality (1D vs 2D vs 3D) of cell-scaffold interactions relevant to tissue engineers that grow cells on biomaterial scaffolds. Results: We designed five statistical and three geometrical contact models, and then down-selected them to one from each category using a validation approach based on physically orthogonal measurements to CLSM. The two selected models were applied to 414 z-stacks with three scaffold types and all contact results were visually verified. A planar geometrical model for the spun coat scaffold type was validated from atomic force microscopy images by computing surface roughness of 52.35 nm ±31.76 nm which was 2 to 8 times smaller than the CLSM resolution. A cylindrical model for fiber scaffolds was validated from multi-view 2D scanning electron microscopy (SEM) images. The fiber scaffold segmentation error was assessed by comparing fiber diameters from SEM and CLSM to be between 0.46% to 3.8% of the SEM reference values. For contact verification, we constructed a web-based visual verification system with 414 pairs of images with cells and their segmentation results, and with 4968 movies with animated cell, scaffold, and contact overlays. Based on visual verification by three experts, we report the accuracy of cell segmentation to be 96.4% with 94.3% precision, and the accuracy of cell-scaffold contact for a statistical model to be 62.6% with 76.7% precision and for a geometrical model to be 93.5% with 87.6% precision. Conclusions: The novelty of our approach lies in (1) representing cell-scaffold contact sites with statistical intensity and geometrical shape models, (2) designing a methodology for validating 3D geometrical contact models and (3) devising a mechanism for visual verification of hundreds of 3D measurements. The raw and processed data are publicly available from https://isg.nist.gov/deepzoomweb/data/together with the web-based verification system. [ABSTRACT FROM AUTHOR]

Background The mismatch between dorsal and ventral trunk features along the millipede trunk was long a subject of controversy, largely resting on alternative interpretations of segmentation. Most models of arthropod segmentation presuppose a strict sequential antero-posterior specification of trunk segments, whereas alternative models involve the early delineation of a limited number of 'primary segments' followed by their sequential stereotypic subdivision into 2n definitive segments. The 'primary segments' should be intended as units identified by molecular markers, rather than as overt morphological entities. Two predictions were suggested to test the plausibility of multiple-duplication models of segmentation: first, a specific pattern of evolvability of segment number in those arthropod clades in which, segment number is not fixed (e.g., epimorphic centipedes and millipedes); second, the occurrence of discrete multisegmental patterns due to early, initially contiguous positional markers. Results We describe a unique case of a homeotic millipede with 6 extra pairs of ectopic gonopods replacing walking legs on rings 8 (leg-pairs 10-11), 15 (leg-pairs 24-25) and 16 (leg-pairs 26- 27); we discuss the segmental distribution of these appendages in the framework of alternative models of segmentation and present an interpretation of the origin of the distribution of the additional gonopods. The anterior set of contiguous gonopods (those normally occurring on ring 7 plus the first set of ectopic ones on ring 8) is reiterated by the posterior set (on rings 15-16) after exactly 16 leg positions along the AP body axis. This suggests that a body section including 16 leg pairs could be a module deriving from 4 cycles of regular binary splitting of an embryonic 'primary segment.' Conclusions A very likely early determination of the sites of the future metamorphosis of walking legs into gonopods and a segmentation process according to the multiplicative model may provide a detailed explanation for the distribution of the extra gonopods in the homeotic specimen. The hypothesized steps of segmentation are similar in both a normal and the studied homeotic specimen. The difference between them would consist in the size of the embryonic trunk region endowed with a positional marker whose presence will later determine the replacement of walking legs by gonopods. [ABSTRACT FROM AUTHOR]