RootPainter: deep learning segmentation of biological images with corrective annotation.

Convolutional neural networks (CNNs) are a powerful tool for plant image analysis, but challenges remain in making them more accessible to researchers without a machine-learning background. We present RootPainter, an open-source graphical user interface based software tool for the rapid training of deep neural networks for use in biological image analysis. We evaluate RootPainter by training models for root length extraction from chicory (Cichorium intybus L.) roots in soil, biopore counting, and root nodule counting. We also compare dense annotations with corrective ones that are added during the training process based on the weaknesses of the current model. Five out of six times the models trained using RootPainter with corrective annotations created within 2 h produced measurements strongly correlating with manual measurements. Model accuracy had a significant correlation with annotation duration, indicating further improvements could be obtained with extended annotation. Our results show that a deep-learning model can be trained to a high accuracy for the three respective datasets of varying target objects, background, and image quality with < 2 h of annotation time. They indicate that, when using RootPainter, for many datasets it is possible to annotate, train, and complete data processing within 1 d.

Digging roots is easier with AI.

The scale of root quantification in research is often limited by the time required for sampling, measurement, and processing samples. Recent developments in convolutional neural networks (CNNs) have made faster and more accurate plant image analysis possible, which may significantly reduce the time required for root measurement, but challenges remain in making these methods accessible to researchers without an in-depth knowledge of machine learning. We analyzed root images acquired from three destructive root samplings using the RootPainter CNN software that features an interface for corrective annotation for easier use. Root scans with and without non-root debris were used to test if training a model (i.e. learning from labeled examples) can effectively exclude the debris by comparing the end results with measurements from clean images. Root images acquired from soil profile walls and the cross-section of soil cores were also used for training, and the derived measurements were compared with manual measurements. After 200 min of training on each dataset, significant relationships between manual measurements and RootPainter-derived data were noted for monolith (R2=0.99), profile wall (R2=0.76), and core-break (R2=0.57). The rooting density derived from images with debris was not significantly different from that derived from clean images after processing with RootPainter. Rooting density was also successfully calculated from both profile wall and soil core images, and in each case the gradient of root density with depth was not significantly different from manual counts. Differences in root-length density (RLD) between crops with contrasting root systems were captured using automatic segmentation at soil profiles with high RLD (1-5 cm cm-3) as well with low RLD (0.1-0.3 cm cm-3). Our results demonstrate that the proposed approach using CNN can lead to substantial reductions in root sample processing workloads, increasing the potential scale of future root investigations.

Semantic segmentation of plant roots from RGB (mini-) rhizotron images-generalisation potential and false positives of established methods and advanced deep-learning models.

BACKGROUND: Manual analysis of (mini-)rhizotron (MR) images is tedious. Several methods have been proposed for semantic root segmentation based on homogeneous, single-source MR datasets. Recent advances in deep learning (DL) have enabled automated feature extraction, but comparisons of segmentation accuracy, false positives and transferability are virtually lacking. Here we compare six state-of-the-art methods and propose two improved DL models for semantic root segmentation using a large MR dataset with and without augmented data. We determine the performance of the methods on a homogeneous maize dataset, and a mixed dataset of > 8 species (mixtures), 6 soil types and 4 imaging systems. The generalisation potential of the derived DL models is determined on a distinct, unseen dataset. RESULTS: The best performance was achieved by the U-Net models; the more complex the encoder the better the accuracy and generalisation of the model. The heterogeneous mixed MR dataset was a particularly challenging for the non-U-Net techniques. Data augmentation enhanced model performance. We demonstrated the improved performance of deep meta-architectures and feature extractors, and a reduction in the number of false positives. CONCLUSIONS: Although correction factors are still required to match human labelled root lengths, neural network architectures greatly reduce the time required to compute the root length. The more complex architectures illustrate how future improvements in root segmentation within MR images can be achieved, particularly reaching higher segmentation accuracies and model generalisation when analysing real-world datasets with artefacts-limiting the need for model retraining.

Dynamics of Deep Water and N Uptake of Oilseed Rape (Brassica napus L.) Under Varied N and Water Supply.

Enhanced nitrogen (N) and water uptake from deep soil layers may increase resource use efficiency while maintaining yield under stressed conditions. Winter oilseed rape (Brassica napus L.) can develop deep roots and access deep-stored resources such as N and water to sustain its growth and productivity. Less is known of the performance of deep roots under varying water and N availability. In this study, we aimed to evaluate the effects of reduced N and water supply on deep N and water uptake for oilseed rape. Oilseed rape plants grown in outdoor rhizotrons were supplied with 240 and 80 kg N ha-1, respectively, in 2019 whereas a well-watered and a water-deficit treatment were established in 2020. To track deep water and N uptake, a mixture of 2H2O and Ca(15NO3)2 was injected into the soil column at 0.5- and 1.7-m depths. δ2H in transpiration water and δ15N in leaves were measured after injection. δ15N values in biomass samples were also measured. Differences in N or water supply had less effect on root growth. The low N treatment reduced water uptake throughout the soil profile and altered water uptake distribution. The low N supply doubled the 15N uptake efficiency at both 0.5 and 1.7 m. Similarly, water deficit in the upper soil layers led to compensatory deep water uptake. Our findings highlight the increasing importance of deep roots for water uptake, which is essential for maintaining an adequate water supply in the late growing stage. Our results further indicate the benefit of reducing N supply for mitigating N leaching and altering water uptake from deep soil layers, yet at a potential cost of biomass reduction.

RootPainter3D: Interactive-machine-learning enables rapid and accurate contouring for radiotherapy.

PURPOSE: Organ-at-risk contouring is still a bottleneck in radiotherapy, with many deep learning methods falling short of promised results when evaluated on clinical data. We investigate the accuracy and time-savings resulting from the use of an interactive-machine-learning method for an organ-at-risk contouring task. METHODS: We implement an open-source interactive-machine-learning software application that facilitates corrective-annotation for deep-learning generated contours on X-ray CT images. A trained-physician contoured 933 hearts using our software by delineating the first image, starting model training, and then correcting the model predictions for all subsequent images. These corrections were added into the training data, which was used for continuously training the assisting model. From the 933 hearts, the same physician also contoured the first 10 and last 10 in Eclipse (Varian) to enable comparison in terms of accuracy and duration. RESULTS: We find strong agreement with manual delineations, with a dice score of 0.95. The annotations created using corrective-annotation also take less time to create as more images are annotated, resulting in substantial time savings compared to manual methods. After 923 images had been delineated, hearts took 2 min and 2 s to delineate on average, which includes time to evaluate the initial model prediction and assign the needed corrections, compared to 7 min and 1 s when delineating manually. CONCLUSIONS: Our experiment demonstrates that interactive-machine-learning with corrective-annotation provides a fast and accessible way for non computer-scientists to train deep-learning models to segment their own structures of interest as part of routine clinical workflows.

Gating has a negligible impact on dose delivered in MRI-guided online adaptive radiotherapy of prostate cancer.

BACKGROUND AND PURPOSE: MR-guided radiotherapy (MRgRT) allows real-time beam-gating to compensate for intra-fractional target position variations. This study investigates the dosimetric impact of beam-gating and the impact of PTV margin on prostate coverage for prostate cancer patients treated with online-adaptive MRgRT. MATERIALS AND METHODS: 20 consecutive prostate cancer patients were treated with online-adaptive MRgRT SBRT with 36.25 Gy in 5 fractions (PTV D95% ≥ 95% (N = 5) and PTV D95% ≥ 100% (N = 15)). Sagittal 2D cine MRIs were used for gating on the prostate with a 3 mm expansion as the gating window. We computed motion-compensated dose distributions for (i) all prostate positions during treatment (simulating non-gated treatments) and (ii) for prostate positions within the gating window (gated treatments). To evaluate the impact of PTV margin on prostate coverage, we simulated coverage with smaller margins than clinically applied both for gated and non-gated treatments. Motion-compensated fraction doses were accumulated and dose metrics were compared. RESULTS: We found a negligible dosimetric impact of beam-gating on prostate coverage (median of 0.00 Gy for both D95% and Dmean). For 18/20 patients, prostate coverage (D95% ≥ 100%) would have been ensured with a prostate-to-PTV margin of 3 mm, even without gating. The same was true for all but one fraction. CONCLUSION: Beam-gating has negligible dosimetric impact in online-adaptive MRgRT of prostate cancer. Accounting for motion, the clinically used prostate-to-PTV margin could potentially be reduced from 5 mm to 3 mm for 18/20 patients.

Relative importance of genotype, gene expression, and DNA methylation on complex traits in perennial ryegrass.

The growing demand for food and feed crops in the world because of growing population and more extreme weather events requires high-yielding and resilient crops. Many agriculturally important traits are polygenic, controlled by multiple regulatory layers, and with a strong interaction with the environment. In this study, 120 F2 families of perennial ryegrass (Lolium perenne L.) were grown across a water gradient in a semifield facility with subsoil irrigation. Genomic (single-nucleotide polymorphism [SNP]), transcriptomic (gene expression [GE]), and DNA methylomic (MET) data were integrated with feed quality trait data collected from control and drought sections in the semifield facility, providing a treatment effect. Deep root length (DRL) below 110 cm was assessed with convolutional neural network image analysis. Bayesian prediction models were used to partition phenotypic variance into its components and evaluated the proportion of phenotypic variance in all traits captured by different regulatory layers (SNP, GE, and MET). The spatial effects and effects of SNP, GE, MET, the interaction between GE and MET (GE × MET) and GE × treatment (GEControl and GEDrought ) interaction were investigated. Gene expression explained a substantial part of the genetic and spatial variance for all the investigated phenotypes, whereas MET explained residual variance not accounted for by SNPs or GE. For DRL, MET also contributed to explaining spatial variance. The study provides a statistically elegant analytical paradigm that integrates genomic, transcriptomic, and MET information to understand the regulatory mechanisms of polygenic effects for complex traits.

Developing and validating COVID-19 adverse outcome risk prediction models from a bi-national European cohort of 5594 patients.

Patients with severe COVID-19 have overwhelmed healthcare systems worldwide. We hypothesized that machine learning (ML) models could be used to predict risks at different stages of management and thereby provide insights into drivers and prognostic markers of disease progression and death. From a cohort of approx. 2.6 million citizens in Denmark, SARS-CoV-2 PCR tests were performed on subjects suspected for COVID-19 disease; 3944 cases had at least one positive test and were subjected to further analysis. SARS-CoV-2 positive cases from the United Kingdom Biobank was used for external validation. The ML models predicted the risk of death (Receiver Operation Characteristics-Area Under the Curve, ROC-AUC) of 0.906 at diagnosis, 0.818, at hospital admission and 0.721 at Intensive Care Unit (ICU) admission. Similar metrics were achieved for predicted risks of hospital and ICU admission and use of mechanical ventilation. Common risk factors, included age, body mass index and hypertension, although the top risk features shifted towards markers of shock and organ dysfunction in ICU patients. The external validation indicated fair predictive performance for mortality prediction, but suboptimal performance for predicting ICU admission. ML may be used to identify drivers of progression to more severe disease and for prognostication patients in patients with COVID-19. We provide access to an online risk calculator based on these findings.

Segmentation of roots in soil with U-Net.

BACKGROUND: Plant root research can provide a way to attain stress-tolerant crops that produce greater yield in a diverse array of conditions. Phenotyping roots in soil is often challenging due to the roots being difficult to access and the use of time consuming manual methods. Rhizotrons allow visual inspection of root growth through transparent surfaces. Agronomists currently manually label photographs of roots obtained from rhizotrons using a line-intersect method to obtain root length density and rooting depth measurements which are essential for their experiments. We investigate the effectiveness of an automated image segmentation method based on the U-Net Convolutional Neural Network (CNN) architecture to enable such measurements. We design a data-set of 50 annotated chicory (Cichorium intybus L.) root images which we use to train, validate and test the system and compare against a baseline built using the Frangi vesselness filter. We obtain metrics using manual annotations and line-intersect counts. RESULTS: Our results on the held out data show our proposed automated segmentation system to be a viable solution for detecting and quantifying roots. We evaluate our system using 867 images for which we have obtained line-intersect counts, attaining a Spearman rank correlation of 0.9748 and an r 2 of 0.9217. We also achieve an F 1 of 0.7 when comparing the automated segmentation to the manual annotations, with our automated segmentation system producing segmentations with higher quality than the manual annotations for large portions of the image. CONCLUSION: We have demonstrated the feasibility of a U-Net based CNN system for segmenting images of roots in soil and for replacing the manual line-intersect method. The success of our approach is also a demonstration of the feasibility of deep learning in practice for small research groups needing to create their own custom labelled dataset from scratch.