Deep Multiview Image Fusion for Soybean Yield Estimation in Breeding Applications.
Ontology highlight
ABSTRACT:
SUBMITTER: Riera LG
PROVIDER: S-EPMC8240512 | biostudies-literature |
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:The contrast and resolution of images obtained with optical microscopes can be improved by deconvolution and computational fusion of multiple views of the same sample, but these methods are computationally expensive for large datasets. Here we describe theoretical and practical advances in algorithm and software design that result in image processing times that are tenfold to several thousand fold faster than with previous methods. First, we show that an 'unmatched back projector' accelerates deconvolution relative to the classic Richardson-Lucy algorithm by at least tenfold. Second, three-dimensional image-based registration with a graphics processing unit enhances processing speed 10- to 100-fold over CPU processing. Third, deep learning can provide further acceleration, particularly for deconvolution with spatially varying point spread functions. We illustrate our methods from the subcellular to millimeter spatial scale on diverse samples, including single cells, embryos and cleared tissue. Finally, we show performance enhancement on recently developed microscopes that have improved spatial resolution, including dual-view cleared-tissue light-sheet microscopes and reflective lattice light-sheet microscopes.
| S-EPMC7642198 | biostudies-literature
Project description:Epigenetic variation has been associated with a wide range of adaptive phenotypes in plants, but there exist few direct means for exploiting this variation. RNAi suppression of the plant-specific gene, MutS HOMOLOG1 (MSH1), in multiple plant species produces a range of developmental changes accompanied by modulation of defence, phytohormone and abiotic stress response pathways along with methylome repatterning. This msh1-conditioned developmental reprogramming is retained independent of transgene segregation, giving rise to transgene-null 'memory' effects. An isogenic memory line crossed to wild type produces progeny families displaying increased variation in adaptive traits that respond to selection. This study investigates amenability of the MSH1 system for inducing agronomically valuable epigenetic variation in soybean. We developed MSH1 epi-populations by crossing with msh1-acquired soybean memory lines. Derived soybean epi-lines showed increase in variance for multiple yield-related traits including pods per plant, seed weight and maturity time in both glasshouse and field trials. Selected epi-F2:4 and epi-F2:5 lines showed an increase in seed yield over wild type. By epi-F2:6, we observed a return of MSH1-derived enhanced growth back to wild-type levels. Epi-populations also showed evidence of reduced epitype-by-environment (e × E) interaction, indicating higher yield stability. Transcript profiling of epi-lines identified putative signatures of enhanced growth behaviour across generations. Genes related to cell cycle, abscisic acid biosynthesis and auxin response, particularly SMALL AUXIN UP RNAs (SAURs), were differentially expressed in epi-F2:4 lines that showed increased yield when compared to epi-F2:6 . These data support the potential of MSH1-derived epigenetic variation in plant breeding for enhanced yield and yield stability.
| S-EPMC6181216 | biostudies-literature
Project description:Early Leaf Spot (ELS) caused by the fungus Passalora arachidicola and Late Leaf Spot (LLS) also caused by the fungus Nothopassalora personata, are the two major groundnut (Arachis hypogaea L.) destructive diseases in Ghana. Accurate phenotyping and genotyping to develop groundnut genotypes resistant to Leaf Spot Diseases (LSD) and to increase groundnut production is critically important in Western Africa. Two experiments were conducted at the Council for Scientific and Industrial Research-Savanna Agricultural Research Institute located in Nyankpala, Ghana to explore the effectiveness of using RGB-image method as a high-throughput phenotyping tool to assess groundnut LSD and to estimate yield components. Replicated plots arranged in a rectangular alpha lattice design were conducted during the 2020 growing season using a set of 60 genotypes as the training population and 192 genotypes for validation. Indirect selection models were developed using Red-Green-Blue (RGB) color space indices. Data was collected on conventional LSD ratings, RGB imaging, pod weight per plant and number of pods per plant. Data was analyzed using a mixed linear model with R statistical software version 4.0.2. The results showed differences among the genotypes for the traits evaluated. The RGB-image method traits exhibited comparable or better broad sense heritability to the conventionally measured traits. Significant correlation existed between the RGB-image method traits and the conventionally measured traits. Genotypes 73-33, Gha-GAF 1723, Zam-ICGV-SM 07599, and Oug-ICGV 90099 were among the most resistant genotypes to ELS and LLS, and they represent suitable sources of resistance to LSD for the groundnut breeding programs in Western Africa.
| S-EPMC9387199 | biostudies-literature
Project description:Yield prediction is of great significance for yield mapping, crop market planning, crop insurance, and harvest management. Remote sensing is becoming increasingly important in crop yield prediction. Based on remote sensing data, great progress has been made in this field by using machine learning, especially the Deep Learning (DL) method, including Convolutional Neural Network (CNN) or Long Short-Term Memory (LSTM). Recent experiments in this area suggested that CNN can explore more spatial features and LSTM has the ability to reveal phenological characteristics, which both play an important role in crop yield prediction. However, very few experiments combining these two models for crop yield prediction have been reported. In this paper, we propose a deep CNN-LSTM model for both end-of-season and in-season soybean yield prediction in CONUS at the county-level. The model was trained by crop growth variables and environment variables, which include weather data, MODIS Land Surface Temperature (LST) data, and MODIS Surface Reflectance (SR) data; historical soybean yield data were employed as labels. Based on the Google Earth Engine (GEE), all these training data were combined and transformed into histogram-based tensors for deep learning. The results of the experiment indicate that the prediction performance of the proposed CNN-LSTM model can outperform the pure CNN or LSTM model in both end-of-season and in-season. The proposed method shows great potential in improving the accuracy of yield prediction for other crops like corn, wheat, and potatoes at fine scales in the future.
| S-EPMC6832950 | biostudies-literature
Project description:Genomic selection (GS) has become viable for selection of quantitative traits for which marker-assisted selection has often proven less effective. The potential of GS for soybean was characterized using 483 elite breeding lines, genotyped with BARCSoySNP6K iSelect BeadChips. Cross validation was performed using RR-BLUP and predictive abilities (r MP) of 0.81, 0.71, and 0.26 for protein, oil, and yield, were achieved at the largest tested training set size. Minimal differences were observed when comparing different marker densities and there appeared to be inflation in r MP due to population structure. For comparison purposes, two additional methods to predict breeding values for lines of four bi-parental populations within the GS dataset were tested. The first method predicted within each bi-parental population (WP method) and utilized a training set of full-sibs of the validation set. The second method utilized a training set of all remaining breeding lines except for full-sibs of the validation set to predict across populations (AP method). The AP method is more practical as the WP method would likely delay the breeding cycle and leverage smaller training sets. Averaging across populations for protein and oil content, r MP for the AP method (0.55, 0.30) approached r MP for the WP method (0.60, 0.52). Though comparable, r MP for yield was low for both AP and WP methods (0.12, 0.13). Based on increases in r MP as training sets increased and the effectiveness of WP vs. AP method, the AP method could potentially improve with larger training sets and increased relatedness between training and validation sets.
| S-EPMC6643879 | biostudies-literature
Project description:Content-based medical image retrieval continues to gain attention for its potential to assist radiological image interpretation and decision making. Many approaches have been proposed to improve the performance of medical image retrieval system, among which visual features such as SIFT, LBP, and intensity histogram play a critical role. Typically, these features are concatenated into a long vector to represent medical images, and thus traditional dimension reduction techniques such as locally linear embedding (LLE), principal component analysis (PCA), or laplacian eigenmaps (LE) can be employed to reduce the "curse of dimensionality". Though these approaches show promising performance for medical image retrieval, the feature-concatenating method ignores the fact that different features have distinct physical meanings. In this paper, we propose a new method called multiview locally linear embedding (MLLE) for medical image retrieval. Following the patch alignment framework, MLLE preserves the geometric structure of the local patch in each feature space according to the LLE criterion. To explore complementary properties among a range of features, MLLE assigns different weights to local patches from different feature spaces. Finally, MLLE employs global coordinate alignment and alternating optimization techniques to learn a smooth low-dimensional embedding from different features. To justify the effectiveness of MLLE for medical image retrieval, we compare it with conventional spectral embedding methods. We conduct experiments on a subset of the IRMA medical image data set. Evaluation results show that MLLE outperforms state-of-the-art dimension reduction methods.
| S-EPMC3862625 | biostudies-literature
Project description:Image memorability is a very hard problem in image processing due to its subjective nature. But due to the introduction of Deep Learning and the large availability of data and GPUs, great strides have been made in predicting the memorability of an image. In this paper, we propose a novel deep learning architecture called ResMem-Net that is a hybrid of LSTM and CNN that uses information from the hidden layers of the CNN to compute the memorability score of an image. The intermediate layers are important for predicting the output because they contain information about the intrinsic properties of the image. The proposed architecture automatically learns visual emotions and saliency, shown by the heatmaps generated using the GradRAM technique. We have also used the heatmaps and results to analyze and answer one of the most important questions in image memorability: "What makes an image memorable?". The model is trained and evaluated using the publicly available Large-scale Image Memorability dataset (LaMem) from MIT. The results show that the model achieves a rank correlation of 0.679 and a mean squared error of 0.011, which is better than the current state-of-the-art models and is close to human consistency (p = 0.68). The proposed architecture also has a significantly low number of parameters compared to the state-of-the-art architecture, making it memory efficient and suitable for production.
| S-EPMC8594589 | biostudies-literature
Project description:Recent substantial advances in high-throughput field phenotyping have provided plant breeders with affordable and efficient tools for evaluating a large number of genotypes for important agronomic traits at early growth stages. Nevertheless, the implementation of large datasets generated by high-throughput phenotyping tools such as hyperspectral reflectance in cultivar development programs is still challenging due to the essential need for intensive knowledge in computational and statistical analyses. In this study, the robustness of three common machine learning (ML) algorithms, multilayer perceptron (MLP), support vector machine (SVM), and random forest (RF), were evaluated for predicting soybean (Glycine max) seed yield using hyperspectral reflectance. For this aim, the hyperspectral reflectance data for the whole spectra ranged from 395 to 1005 nm, which were collected at the R4 and R5 growth stages on 250 soybean genotypes grown in four environments. The recursive feature elimination (RFE) approach was performed to reduce the dimensionality of the hyperspectral reflectance data and select variables with the largest importance values. The results indicated that R5 is more informative stage for measuring hyperspectral reflectance to predict seed yields. The 395 nm reflectance band was also identified as the high ranked band in predicting the soybean seed yield. By considering either full or selected variables as the input variables, the ML algorithms were evaluated individually and combined-version using the ensemble-stacking (E-S) method to predict the soybean yield. The RF algorithm had the highest performance with a value of 84% yield classification accuracy among all the individual tested algorithms. Therefore, by selecting RF as the metaClassifier for E-S method, the prediction accuracy increased to 0.93, using all variables, and 0.87, using selected variables showing the success of using E-S as one of the ensemble techniques. This study demonstrated that soybean breeders could implement E-S algorithm using either the full or selected spectra reflectance to select the high-yielding soybean genotypes, among a large number of genotypes, at early growth stages.
| S-EPMC7835636 | biostudies-literature
Project description:Fully convolutional neural networks (FCNs), and in particular U-Nets, have achieved state-of-the-art results in semantic segmentation for numerous medical imaging applications. Moreover, batch normalization and Dice loss have been used successfully to stabilize and accelerate training. However, these networks are poorly calibrated i.e. they tend to produce overconfident predictions for both correct and erroneous classifications, making them unreliable and hard to interpret. In this paper, we study predictive uncertainty estimation in FCNs for medical image segmentation. We make the following contributions: 1) We systematically compare cross-entropy loss with Dice loss in terms of segmentation quality and uncertainty estimation of FCNs; 2) We propose model ensembling for confidence calibration of the FCNs trained with batch normalization and Dice loss; 3) We assess the ability of calibrated FCNs to predict segmentation quality of structures and detect out-of-distribution test examples. We conduct extensive experiments across three medical image segmentation applications of the brain, the heart, and the prostate to evaluate our contributions. The results of this study offer considerable insight into the predictive uncertainty estimation and out-of-distribution detection in medical image segmentation and provide practical recipes for confidence calibration. Moreover, we consistently demonstrate that model ensembling improves confidence calibration.
| S-EPMC7704933 | biostudies-literature
Project description:Yield monitoring is an important parameter to evaluate cotton productivity during cotton harvest. Nondestructive and accurate yield monitoring is of great significance to cotton production. Unmanned aerial vehicle (UAV) remote sensing has fast and repetitive acquisition ability. The visible vegetation indices has the advantages of low cost, small amount of calculation and high resolution. The combination of the UAV and visible vegetation indices has been more and more applied to crop yield monitoring. However, there are some shortcomings in estimating cotton yield based on visible vegetation indices only as the similarity between cotton and mulch film makes it difficult to differentiate them and yields may be saturated based on vegetation index estimates near harvest. Texture feature is another important remote sensing information that can provide geometric information of ground objects and enlarge the spatial information identification based on original image brightness. In this study, RGB images of cotton canopy were acquired by UAV carrying RGB sensors before cotton harvest. The visible vegetation indices and texture features were extracted from RGB images for cotton yield monitoring. Feature parameters were selected in different methods after extracting the information. Linear and nonlinear methods were used to build cotton yield monitoring models based on visible vegetation indices, texture features and their combinations. The results show that (1) vegetation indices and texture features extracted from the ultra-high-resolution RGB images obtained by UAVs were significantly correlated with the cotton yield; (2) The best model was that combined with vegetation indices and texture characteristics RF_ELM model, verification set R 2 was 0.9109, and RMSE was 0.91277 t.ha-1. rRMSE was 29.34%. In conclusion, the research results prove that UAV carrying RGB sensor has a certain potential in cotton yield monitoring, which can provide theoretical basis and technical support for field cotton production evaluation.
| S-EPMC9240637 | biostudies-literature