Region-based Convolutional Neural Network Model Research Articles

Manipulation of a Complex Object Using Dual-Arm Robot with Mask R-CNN and Grasping Strategy

Hot forging is one of the common manufacturing processes for producing brass workpieces. However forging produces flash which is a thin metal part around the desired part formed with an excessive material. Using robots with vision system to manipulate this workpiece has encountered several challenging issues, e.g. the uncertain shape of flash, color, reflection of brass surface, different lighting condition, and the uncertainty surrounding the position and orientation of the workpiece. In this research, Mask region-based convolutional neural network together with image processing is used to resolve these issues. The depth camera can provide images for visual detection. Machine learning Mask region-based convolutional neural network model was trained with color images and the position of the object is determined by the depth image. A dual arm 7 degree of freedom collaborative robot with proposed grasping strategy is used to grasp the workpiece that can be in inappropriate position and pose. Eventually, experiments were conducted to assess the visual detection process and the grasp planning of the robot.

Fusion of Mask RCNN and multi-augmentation for rice recognition

In the real world, the appearance of similar rice varieties depends on various factors such as resolution, angle, lighting conditions, and perspective. Additionally, complex environmental factors and characteristics of each rice type, such as enhanced light intensity, cross-polarization, and shading, rice background color, and image similarity, play a role. This indicates that the data augmentation process may enhance the accuracy of crop identification, particularly in the context of self-supervised machine learning. The aim of this research is to develop a precise rice segmentation method based on the improved Mask R-CNN (Region-based Convolutional Neural Network) with multitask data augmentation. The Mask R-CNN model is enhanced by incorporating multitask input to improve feature extraction for rice. Experimental results demonstrate that the improved Mask R-CNN model can accurately segment various rice types under different conditions, such as different background colors and varying sizes of rice grains. The achieved precision, recall, F1 score, and segmentation mean Average Precision (mAP) are 95.5%, 96.3%, 95.9%, and 0.924, respectively. The average runtime on the test set is 0.35 seconds per image. Our method outperforms two comparative approaches, showcasing its ability to accurately segment rice in the market deployment phase with near real-time performance. This study establishes the foundation for the accurate detection of valuable agricultural products.

Deep Learning-Assisted Quantitative Measurement of Thoracolumbar Fracture Features on Lateral Radiographs.

This study aimed to develop and validate a deep learning (DL) algorithm for the quantitative measurement of thoracolumbar (TL) fracture features, and to evaluate its efficacy across varying levels of clinical expertise. Using the pretrained Mask Region-Based Convolutional Neural Networks model, originally developed for vertebral body segmentation and fracture detection, we fine-tuned the model and added a new module for measuring fracture metrics-compression rate (CR), Cobb angle (CA), Gardner angle (GA), and sagittal index (SI)-from lumbar spine lateral radiographs. These metrics were derived from six-point labeling by 3 radiologists, forming the ground truth (GT). Training utilized 1,000 nonfractured and 318 fractured radiographs, while validations employed 213 internal and 200 external fractured radiographs. The accuracy of the DL algorithm in quantifying fracture features was evaluated against GT using the intraclass correlation coefficient. Additionally, 4 readers with varying expertise levels, including trainees and an attending spine surgeon, performed measurements with and without DL assistance, and their results were compared to GT and the DL model. The DL algorithm demonstrated good to excellent agreement with GT for CR, CA, GA, and SI in both internal (0.860, 0.944, 0.932, and 0.779, respectively) and external (0.836, 0.940, 0.916, and 0.815, respectively) validations. DL-assisted measurements significantly improved most measurement values, particularly for trainees. The DL algorithm was validated as an accurate tool for quantifying TL fracture features using radiographs. DL-assisted measurement is expected to expedite the diagnostic process and enhance reliability, particularly benefiting less experienced clinicians.

Feasibility of predicting functional decline in the elderly through key posture information during sit-to-stand movement

BackgroundEarly detection of functional decline in the elderly in day care centres facilitates timely implementation of preventive and treatment measures. Research questionWhether or not a predictive model can be developed by applying image recognition to analyze elderly individuals' posture during the sit-to-stand (STS) manoeuvre. MethodsWe enrolled sixty-six participants (24 males and 42 females) in an observational study design. To estimate posture key point information, we employed a region-based convolutional neural network model and utilized nine key points and their coordinates to calculate seven eigenvalues (X1-X7) that represented the motion curve features during the STS manoeuvre. One-way analysis of variance was performed to evaluate four STS strategies and four types of compensation strategies for three groups with different capacities (college students, community-dwelling elderly, and day care center elderly). Finally, a machine learning predictive model was established. ResultsSignificant differences (p < 0.05) were observed in all eigenvalues except X2 (momentum transfer phase, p = 0.168) between participant groups; significant differences (p < 0.05) were observed in all eigenvalues except X2 (p = 0.219) and X3 (hip-rising phase, p = 0.286) between STS patterns; significant differences (p < 0.05) were observed in all eigenvalues except X2 (p = 0.842) and X3 (p = 0.074) between compensation strategies. The motion curve eigenvalues of the seven posture key points were used to build a machine learning model with 85% accuracy in capacity detection, 70% accuracy in pattern detection, and 85% accuracy in compensation strategy detection. SignificanceThis study preliminarily demonstrates that eigenvalues can be used to detect STS patterns and compensation strategies adopted by individuals with different capacities. Our machine learning model has excellent predictive accuracy and may be used to develop inexpensive and effective systems to help caregivers to continuously monitor STS patterns and compensation strategies of elderly individuals as warning signs of functional decline.

Optimization of Targeted Differential Interferometric Measurements for Wellpads Detected by Mask Region-Based Convolutional Neural Network in the Tengiz Oilfield of the Caspian Sea Coast

Many previous studies have primarily focused on the use of deep learning for interferometric processing or separate recognition purposes rather than targeted measurements of detected wellpads. The present study centered around the integration of deep learning recognition and interferometric measurements for Tengiz oilfield wellpads. This study proposes the optimization, automation, and acceleration of targeted ground deformation wellpad monitoring. Mask Region-based Convolutional Neural Network (R-CNN)-based deep learning wellpad recognition and consequent Small Baseline Subset Synthetic Aperture Radar Interferometry (SBAS-InSAR) analyses were used for the assessment of ground deformation in the wellpads. The Mask R-CNN technique allowed us to detect 159 wells with a confidence level of more than 95%. The Mask R-CNN model achieved a precision value of 0.71 and a recall value of 0.91. SBAS-InSAR interferometric measurements identified 13 wells for Sentinel-1 (SNT1), 8 wells for COSMO-SkyMed (CSK), and 20 wells for TerraSAR-X (TSX) located within the −54–−40 mm/y class of vertical displacement (VD) velocity. Regression analyses for the annual deformation velocities and cumulative displacements (CD) of wells derived from SNT1, CSK, and TSX satellite missions showed a good agreement with R2 &gt; 95. The predictions for cumulative displacements showed that the vertical subsidence processes will continue and reach −339 mm on 31 December 2023, with increasing spatial coverage and the potential to impact a higher number of wells. The hydrological analyses in the Tengiz oilfield clearly demonstrated that water flow has been moving towards the detected hotspot of subsidence and that its accumulation will increase with increasing subsidence. This detected subsidence hotspot was observed at a crossing with a seismic fault that might always be subject to reactivation. The role of this seismic fault should also be investigated as one of the ground deformation-controlling factors, even though this area is not considered seismically active. The primary practical and scientific values of these studies were identified for the operational risk assessment and maintenance needs of oilfield and gas field operators.

A timely and accurate approach to nearshore oil spill monitoring using deep learning and GIS

Oil spill accidents are a key contributor to marine pollution worldwide. Therefore, timely and effective oil spill detection is crucial for reducing marine pollution and enhancing environmental protection. Against this backdrop, this study explored two methods for performing nearshore on-site oil spill detection and segmentation, namely the U-net and Mask region-based convolutional neural network (R-CNN) methods. The U-net and Mask R-CNN models were revealed to exhibit acceptable and favorable performance, achieving overall accuracy of 77.01 % and 89.02 %, respectively. Subsequently, a verification system based on the Geographic Information System (GIS) was developed to improve the performance of the deep-learning model. With the integration of the verification system, the Mask R-CNN model achieved higher overall accuracy of 90.78 %. The feasibility of applying deep-learning methods to nearshore on-site oil spill monitoring was demonstrated through this study. In addition, the integration of the GIS not only assisted in the provision of oil spill information but also in the improvement of the deep-learning models. The timely, accurate, and effective method for nearshore on-site oil spill monitoring that this study explored can be applied to considerably improve traditional on-site oil spill monitoring, which has received limited academic attention in the last two decades.

Detection of the Corn Kernel Breakage Rate Based on an Improved Mask Region-Based Convolutional Neural Network

Real-time knowledge of kernel breakage during corn harvesting plays a significant role in the adjustment of operational parameters of corn kernel harvesters. (1) Transfer learning by initializing the DenseNet121 network with pre-trained weights for training and validating a dataset of corn kernels was adopted. Additionally, the feature extraction capability of DenseNet121 was improved by incorporating the attention mechanism of a Convolutional Block Attention Module (CBAM) and a Feature Pyramid Network (FPN) structure. (2) The quality of intact and broken corn kernels and their pixels were found to be coupled, and a linear regression model was established using the least squares method. The results of the test showed that: (1) The MAPb50 and MAPm50 of the improved Mask Region-based Convolutional Neural Network (RCNN) model were 97.62% and 98.70%, in comparison to the original Mask Region-based Convolutional Neural Network (RCNN) model, which were improved by 0.34% and 0.37%, respectively; the backbone FLOPs and Params were 3.09 GMac and 9.31 M, and the feature extraction time was 206 ms; compared to the original backbone, these were reduced by 3.87 GMac and 17.32 M, respectively. The training of the obtained prediction weights for the detection of a picture of the corn kernel took 76 ms, so compared to the Mask RCNN model, it was reduced by 375 ms; based on the concept of transfer learning, the improved Mask RCNN model converged twice as quickly with the loss function using pre-training weights than the loss function without pre-training weights during training. (2) The coefficients of determination R2 of the two models, when the regression models of the pixels and the quality of intact and broken corn kernels were analyzed, were 0.958 and 0.992, respectively. These findings indicate a strong correlation between the pixel characteristics and the quality of corn kernels. The improved Mask RCNN model was used to segment mask pixels to calculate the corn kernel breakage rate. The verified error between the machine vision and the real breakage rate ranged from −0.72% to 0.65%, and the detection time of the corn kernel breakage rate was only 76 ms, which could meet the requirements for real-time detection. According to the test results, the improved Mask RCNN method had the advantages of a fast detection speed and high accuracy, and can be used as a data basis for adjusting the operation parameters of corn kernel harvesters.

Improvement of Road Instance Segmentation Algorithm Based on the Modified Mask R-CNN

Although the Mask region-based convolutional neural network (R-CNN) model possessed a dominant position for complex and variable road scene segmentation, some problems still existed, including insufficient feature expressive ability and low segmentation accuracy. To address these problems, a novel road scene segmentation algorithm based on the modified Mask R-CNN was proposed. The multi-scale backbone network, Res2Net, was utilized to replace the ResNet network, and aimed to improve the feature extraction capability. The soft non-maximum suppression algorithm with attenuation function (soft-NMS) was adopted to improve detection efficiency in the case of a higher overlap rate. The comparison analyses of partition accuracy for various models were performed on the adopted Cityscapes dataset. The results demonstrated that the modified Mask R-CNN effectively increased the segmentation accuracy, especially for small and highly overlapping objects. The adopted Res2Net and soft-NMS can effectively enhance the feature extraction and improve segmentation performance. The average accuracy of the modified Mask R-CNN model reached up to 0.321, and was 0.054 higher than Mask R-CNN. This work provides important guidance to design a more efficient road scene instance segmentation algorithm for further promoting the actual application in automatic driving systems.

A Method for Detecting Key Points of Transferring Barrel Valve by Integrating Keypoint R-CNN and MobileNetV3

Industrial robots need to accurately identify the position and rotation angle of the handwheel of chemical raw material barrel valves during the process of opening and closing, in order to avoid interference between the robot gripper and the handwheel. This paper proposes a handwheel keypoint detection algorithm for fast and accurate acquisition of handwheel position and rotation pose. The algorithm is based on the Keypoint R-CNN (Region-based Convolutional Neural Network) keypoint detection model, which integrates the lightweight mobile network MobileNetV3, the Coordinate Attention module, and improved BiFPN (Bi-directional Feature Pyramid Network) structure to improve the detection speed of the model, enhance the feature extraction performance of the handwheel, and improve the expression capability of small targets at keypoint locations. Experimental results on a self-built handwheel dataset demonstrate that the proposed algorithm outperforms the Keypoint R-CNN model in terms of detection speed and accuracy, with a speed improvement of 54.6%. The detection accuracy and keypoint detection accuracy reach 93.3% and 98.7%, respectively, meeting the requirements of the application scenario and enabling accurate control of the robot’s rotation of the valve handwheel.

Improved Faster Region-Based Convolutional Neural Networks (R-CNN) Model Based on Split Attention for the Detection of Safflower Filaments in Natural Environments

The accurate acquisition of safflower filament information is the prerequisite for robotic picking operations. To detect safflower filaments accurately in different illumination, branch and leaf occlusion, and weather conditions, an improved Faster R-CNN model for filaments was proposed. Due to the characteristics of safflower filaments being dense and small in the safflower images, the model selected ResNeSt-101 with residual network structure as the backbone feature extraction network to enhance the expressive power of extracted features. Then, using Region of Interest (ROI) Align improved ROI Pooling to reduce the feature errors caused by double quantization. In addition, employing the partitioning around medoids (PAM) clustering was chosen to optimize the scale and number of initial anchors of the network to improve the detection accuracy of small-sized safflower filaments. The test results showed that the mean Average Precision (mAP) of the improved Faster R-CNN reached 91.49%. Comparing with Faster R-CNN, YOLOv3, YOLOv4, YOLOv5, and YOLOv6, the improved Faster R-CNN increased the mAP by 9.52%, 2.49%, 5.95%, 3.56%, and 1.47%, respectively. The mAP of safflower filaments detection was higher than 91% on a sunny, cloudy, and overcast day, in sunlight, backlight, branch and leaf occlusion, and dense occlusion. The improved Faster R-CNN can accurately realize the detection of safflower filaments in natural environments. It can provide technical support for the recognition of small-sized crops.

Automatic borehole fracture detection and characterization with tailored Faster R-CNN and simplified Hough transform

Understanding the geological information of fractures from borehole images is essential for conducting a thorough geological analysis in engineering. Recently, the convolutional neural network (CNN) has been employed to extract high-level features of fractures from images. We propose an improved Faster R-CNN (region-based CNN) model for fracture detection, which is suitably equipped with a tailored region proposal network f-RPN for borehole fractures. The anchors generated by f-RPN are specifically designed with a fixed width and varying heights, and only the anchors with the maximum intersection-over-union values are labeled as positive. By taking advantage of the cyclic feature of borehole images, the images are expanded in four orientations to increase the amount and diversity of the model input. The detection results of the model can further be utilized to simplify the Hough transform for characterizing dip angles and dip directions of fractures. The proposed method is validated with borehole images from multiple projects in hydraulic engineering. Experiments reveal that the average detection accuracy is approximately 91.5%, which exhibits an improvement of 11.06% compared to the original Faster R-CNN. Further analysis indicates that both f-RPN and the multi-orientation expansion contribute to improving the prediction accuracy. It is also shown that this method demonstrates robustness against blurred image background and outperforms traditional image processing methods in the presence of noise. The fracture characterization by the simplified Hough transform turns out to be computationally efficient and accurate, with the absolute errors less than 5° compared to the manual identification results.

Moving Object Detection using Intelligent Spectaculars

Abstract: Moving Object detection systems are used to identify and locate objects in images or videos. When used spectacles, an object detection system can allow the user to see information about the objects in their field of view. This can be useful in a variety of applications, such as helping blind people navigate their environment, or providing augmented reality information to workers. Region-based Convolutional Neural Networks (RCNN) is a type of machine learning model that is commonly used for object detection. The RCNN model first uses a convolutional neural network (CNN) to extract features from the input image, and then applies a region proposal algorithm to identify potential object regions in the image. These regions are then fed into a second CNN, which classifies the regions as objects or background. The RCNN model has been shown to be effective at detecting a wide range of objects in images and videos.

Affective Feedback Synthesis Towards Multimodal Text and Image Data

In this article, we have defined a novel task of affective feedback synthesis that generates feedback for input text and corresponding images in a way similar to humans responding to multimodal data. A feedback synthesis system has been proposed and trained using ground-truth human comments along with image–text input. We have also constructed a large-scale dataset consisting of images, text, Twitter user comments, and the number of likes for the comments by crawling news articles through Twitter feeds. The proposed system extracts textual features using a transformer-based textual encoder. The visual features have been extracted using a Faster region-based convolutional neural networks model. The textual and visual features have been concatenated to construct multimodal features that the decoder uses to synthesize the feedback. We have compared the results of the proposed system with baseline models using quantitative and qualitative measures. The synthesized feedbacks have been analyzed using automatic and human evaluation. They have been found to be semantically similar to the ground-truth comments and relevant to the given text–image input.

Research into Heat Stress Behavior Recognition and Evaluation Index for Yellow-Feathered Broilers, Based on Improved Cascade Region-Based Convolutional Neural Network

The heat stress response of broilers will adversely affect the large-scale and welfare of the breeding of broilers. In order to detect the heat stress state of broilers in time, make reasonable adjustments, and reduce losses, this paper proposed an improved Cascade R-CNN (Region-based Convolutional Neural Networks) model based on visual technology to identify the behavior of yellow-feathered broilers. The improvement of the model solved the problem of the behavior recognition not being accurate enough when broilers were gathered. The influence of different iterations on the model recognition effect was compared, and the optimal model was selected. The final average accuracy reached 88.4%. The behavioral image data with temperature and humidity data were combined, and the heat stress evaluation model was optimized using the PLSR (partial least squares regression) method. The behavior recognition results and optimization equations were verified, and the test accuracy reached 85.8%. This proves the feasibility of the heat stress evaluation optimization equation, which can be used for reasonably regulating the broiler chamber.

Classification of normal and abnormal fetal heart ultrasound images and identification of ventricular septal defects based on deep learning.

Congenital heart defects (CHDs) are the most common birth defects. Recently, artificial intelligence (AI) was used to assist in CHD diagnosis. No comparison has been made among the various types of algorithms that can assist in the prenatal diagnosis. Normal and abnormal fetal ultrasound heart images, including five standard views, were collected according to the International Society of Ultrasound in Obstetrics and Gynecology (ISUOG) Practice guidelines. You Only Look Once version 5 (YOLOv5) models were trained and tested. An excellent model was screened out after comparing YOLOv5 with other classic detection methods. On the training set, YOLOv5n performed slightly better than the others. On the validation set, YOLOv5n attained the highest overall accuracy (90.67 %). On the CHD test set, YOLOv5n, which only needed 0.007 s to recognize each image, had the highest overall accuracy (82.93 %), and YOLOv5l achieved the best accuracy on the abnormal dataset (71.93 %). On the VSD test set, YOLOv5l had the best performance, with a 92.79 % overall accuracy rate and 92.59 % accuracy on the abnormal dataset. The YOLOv5 models achieved better performance than the Fast region-based convolutional neural network (RCNN) & ResNet50 model and the Fast RCNN & MobileNetv2 model on the CHD test set (p<0.05) and VSD test set (p<0.01). YOLOv5 models are able to accurately distinguish normal and abnormal fetal heart ultrasound images, especially with respect to the identification of VSD, which have the potential to assist ultrasound in prenatal diagnosis.

Leguminous seeds detection based on convolutional neural networks: Comparison of Faster R-CNN and YOLOv4 on a small custom dataset

This paper help with leguminous seeds detection and smart farming. There are hundreds of kinds of seeds and it can be very difficult to distinguish between them. Botanists and those who study plants, however, can identify the type of seed at a glance. As far as we know, this is the first work to consider leguminous seeds images with different backgrounds and different sizes and crowding. Machine learning is used to automatically classify and locate 11 different seed types. We chose Leguminous seeds from 11 types to be the objects of this study. Those types are of different colors, sizes, and shapes to add variety and complexity to our research. The images dataset of the leguminous seeds was manually collected, annotated, and then split randomly into three sub-datasets train, validation, and test (predictions), with a ratio of 80%, 10%, and 10% respectively. The images considered the variability between different leguminous seed types. The images were captured on five different backgrounds: white A4 paper, black pad, dark blue pad, dark green pad, and green pad. Different heights and shooting angles were considered. The crowdedness of the seeds also varied randomly between 1 and 50 seeds per image. Different combinations and arrangements between the 11 types were considered. Two different image-capturing devices were used: a SAMSUNG smartphone camera and a Canon digital camera. A total of 828 images were obtained, including 9801 seed objects (labels). The dataset contained images of different backgrounds, heights, angles, crowdedness, arrangements, and combinations. The TensorFlow framework was used to construct the Faster Region-based Convolutional Neural Network (R-CNN) model and CSPDarknet53 is used as the backbone for YOLOv4 based on DenseNet designed to connect layers in convolutional neural. Using the transfer learning method, we optimized the seed detection models. The currently dominant object detection methods, Faster R-CNN, and YOLOv4 performances were compared experimentally. The mAP (mean average precision) of the Faster R-CNN and YOLOv4 models were 84.56% and 98.52% respectively. YOLOv4 had a significant advantage in detection speed over Faster R-CNN which makes it suitable for real-time identification as well where high accuracy and low false positives are needed. The results showed that YOLOv4 had better accuracy, and detection ability, as well as faster detection speed beating Faster R-CNN by a large margin. The model can be effectively applied under a variety of backgrounds, image sizes, seed sizes, shooting angles, and shooting heights, as well as different levels of seed crowding. It constitutes an effective and efficient method for detecting different leguminous seeds in complex scenarios. This study provides a reference for further seed testing and enumeration applications.

Computer-aided automated diminutive colonic polyp detection in colonoscopy by using deep machine learning system; first indigenous algorithm developed in India.

Colonic polyps can be detected and resected during a colonoscopy before cancer development. However, about 1/4th of the polyps could be misseddue to their small size, location or human errors. An artificial intelligence (AI) system can improve polyp detection and reduce colorectal cancer incidence. We are developing an indigenous AI systemto detect diminutive polyps in real-life scenarios that can be compatible with any high-definition colonoscopy and endoscopic video-capture software. We trained a masked region-based convolutional neural network model to detect and localize colonic polyps. Three independent datasets of colonoscopy videos comprising 1,039 image frames were used and divided into a training dataset of 688 frames and a testing dataset of 351 frames. Of 1,039 image frames, 231 were from real-life colonoscopy videos from our centre. The rest were from publicly available image frames already modified to be directly utilizable for developing the AI system. The image frames of the testing dataset were also augmented by rotating and zooming the images to replicate real-life distortions of images seen during colonoscopy. The AI system was trained to localize the polyp by creating a 'bounding box'. It was then applied to the testing dataset to test its accuracy in detecting polyps automatically. The AI system achieved a mean average precision (equivalent to specificity) of 88.63% for automatic polyp detection. All polyps in the testing were identified by AI, i.e., nofalse-negativeresult in the testing dataset (sensitivity of 100%). The mean polyp size in the study was 5 (± 4) mm. The mean processing time per image frame was 96.4minutes. This AI system, when applied to real-life colonoscopy images, having wide variations in bowel preparation and small polyp size, can detect colonic polyps with a high degree of accuracy.

Crack detection based on deep learning: a method for evaluating the object detection networks considering the random fractal of crack

The visual test for cracks is a tough work for human eyes. With the help of cameras, methods for object detection based on deep learning can detect the cracks automatically from digital images. However, an anomaly is observed that the models for crack detection are obviously underestimated by the widely used mean Average Precision (mAP) standard. In this study, by theoretical analyses about the topology of cracks, we attribute the failure of mAP standard to the random fractal feature. More deeply, this issue is due to the strict box matching during the calculation. To solve this problem, we construct a practical fractal-available evaluation method adopting the idea of covering box matching. Several metrics including Extended Recall (XR), Extended Precision (XP), and Extended F-score (Fext) are defined for scoring the crack detectors. Then, we conduct experiments on several mainstream models for object detection. Results show two advantages of the proposed evaluation method: (1) the proposed metrics evaluate cases better; (2) the proposed evaluation is non-maximum suppression-unrelated. Moreover, we present a case study to optimize a faster region-based convolutional neural networks model for crack detection adopting the proposed metrics. Recall (represented by the XR) of our best model achieves an industrial-level at 95.8%. These results denote that the proposed evaluation method removes the obstacle in evaluating the methods for object detection for cracks, which show great potential for automatic industrial inspections.

Computer vision system using deep learning to predict rib and loin yield in the fish Colossoma macropomum.

Computer vision system (CVSs) are effective tools that enable large-scale phenotyping with a low-cost and non-invasive method, which avoids animal stress. Economically important traits, such as rib and loin yield, are difficult to measure; therefore, the use of CVS is crucial to accurately predict several measures to allow their inclusion in breeding goals by indirect predictors. Therefore, this study aimed (1) to validate CVS by a deep learning approach and to automatically predict morphometric measurements in tambaqui and (2) to estimate genetic parameters for growth traits and body yield. Data from 365 individuals belonging to 11 full-sib families were evaluated. Seven growth traits were measured. After biometrics, each fish was processed in the following body regions: head, rib, loin, R + L (rib + loin). For deep learning image segmentation, we adopted a method based on the instance segmentation of the Mask R-CNN (Region-based Convolutional Neural Networks) model. Pearson's correlation values between measurements predicted manually and automatically by the CVS were high and positive. Regarding the classification performance, visible differences were detected in only about 3% of the images. Heritability estimates for growth and body yield traits ranged from low to high. The genetic correlations between the percentage of body parts and morphometric characteristics were favorable and highly correlated, except for percentage head, whose correlations were unfavorable. In conclusion, the CVS validated in this image dataset proved to be resilient and can be used for large-scale phenotyping in tambaqui. The weight of the rib and loin are traits under moderate genetic control and should respond to selection. In addition, standard length and pelvis length can be used as an efficient and indirect selection criterion for body yield in this tambaqui population.

Detecting pediatric wrist fractures using deep-learning-based object detection.

Missed fractures are the leading cause of diagnostic error in the emergency department, and fractures of pediatric bones, particularly subtle wrist fractures, can be misidentified because of their varying characteristics and responses to injury. This study evaluated the utility of an object detection deep learning framework for classifying pediatric wrist fractures as positive or negative for fracture, including subtle buckle fractures of the distal radius, and evaluated the performance of this algorithm as augmentation to trainee radiograph interpretation. We obtained 395 posteroanterior wrist radiographs from unique pediatric patients (65% positive for fracture, 30% positive for distal radial buckle fracture) and divided them into train (n = 229), tune (n = 41) and test (n = 125) sets. We trained a Faster R-CNN (region-based convolutional neural network) deep learning object-detection model. Two pediatric and two radiology residents evaluated radiographs initially without the artificial intelligence (AI) assistance, and then subsequently with access to the bounding box generated by the Faster R-CNN model. The Faster R-CNN model demonstrated an area under the curve (AUC) of 0.92 (95% confidence interval [CI] 0.87-0.97), accuracy of 88% (n = 110/125; 95% CI 81-93%), sensitivity of 88% (n = 70/80; 95% CI 78-94%) and specificity of 89% (n = 40/45, 95% CI 76-96%) in identifying any fracture and identified 90% of buckle fractures (n = 35/39, 95% CI 76-97%). Access to Faster R-CNN model predictions significantly improved average resident accuracy from 80 to 93% in detecting any fracture (P < 0.001) and from 69 to 92% in detecting buckle fracture (P < 0.001). After accessing AI predictions, residents significantly outperformed AI in cases of disagreement (73% resident correct vs. 27% AI, P = 0.002). An object-detection-based deep learning approach trained with only a few hundred examples identified radiographs containing pediatric wrist fractures with high accuracy. Access to model predictions significantly improved resident accuracy in diagnosing these fractures.