ARTICLE | doi:10.20944/preprints202002.0004.v1
Subject: Biology And Life Sciences, Biochemistry And Molecular Biology Keywords: DNA; phenotyping; intelligence; interpretation
Online: 3 February 2020 (03:27:19 CET)
The ability to predict physical characteristics from DNA presents significant opportunities for forensic science. Giving scientists an ability to make predictions about the donor of genetic material at a crime scene can then give investigators new intelligence leads for cold cases where DNA evidence has not identified any person of interest. However, the interpretation of this new form of intelligence requires careful analysis. The responses to an online survey, conducted in 2018-19, were used to examine how actors in the criminal justice system assess and interpret different types of DNA evidence and intelligence. The groups of focus for the survey were investigators, legal practitioners and the general public (as potential jurors). Several statistically significant effects were identified based on occupation and whether an individual had prior exposure to new DNA technology. Monitoring how those involved in interpreting reports from different types of DNA evidence and intelligence interpret them helps to ensure that decisions are made based on a sound understanding of their capabilities and limitations and may inform broader training and awareness strategies.
TECHNICAL NOTE | doi:10.20944/preprints202203.0341.v1
Subject: Biology And Life Sciences, Horticulture Keywords: Reflectance; Ocimum basilicum; Colour sensor; Phenotyping
Online: 25 March 2022 (08:44:33 CET)
Modern agriculture demands for comprehensive information about the plant itself. Conventional chemistry-based analytical methods - due to their low throughput and high associated cost - are no longer capable of providing these data. In recent years, remote reflectance-based characterization has developed as one of the most promising solutions for rapid assessments for plant attributes. However, in many cases, expensive equipment is required because accurate quantifications need assessment of the full reflectance spectrum. We examined the versatility of visible colour sensors as reflectance measuring devices for biological / biochemical quantifications on sweet basil (Ocimum basilicum). Our results indicate for the wide potential of spectral colour sensors for quantitative determination of leaf phenolic compounds, flavonoids in particular, and non-invasive plant phenotyping in agricultural applications by low-cost sensors.
COMMUNICATION | doi:10.20944/preprints202305.0950.v1
Subject: Biology And Life Sciences, Plant Sciences Keywords: genebank; digital phenotyping; RDA-Genebank; Nagoya Protocol
Online: 12 May 2023 (13:31:37 CEST)
The National Agrobiodiversity Center under the Rural Development Administration (RDA) in Jeonju, Republic of Korea, is the largest international genebank of Korea. The National Agrobi-odiversity Center has continuously been making efforts to introduce overseas genetic resources to its collection and has currently grown into a world-class PGR holding genebank. Currently, several steps are undertaken to, improve the accessibility of the collection to national as well as international researchers, improve data available on the resources and amend the passport in-formation of the accessions. With the implementation of the Nagoya Protocol, the origin of ge-netic resources is being highlighted as an important input in the passport information of germplasms. The RDA-Genebank actively responds to the Nagoya Protocol by supplementing passport data of resources lacking information on the origin of germplasm. In addition, a large number of conserved resources are continuously multiplied, and agronomic traits are investi-gated concurrently. With the traditional methods of characterization of the germplasm requiring a significant amount of time and effort, we have intiated high-throughput phenotyping using digital techniques to improve our germplasm data. Primarily, we have started adding seed phenotype information followed by measuring root phenotypes which are stored under agro-nomic traits. This may be the initial step for a germplasm center to use large-scale high-throughput techniques for their collection. In this shortcommunication, we aim to provide an introduction to the RDA- Genebank, adopted international standards, and establishment of high throughput phenotyping techniques for improvement of passport information.
REVIEW | doi:10.20944/preprints202204.0228.v1
Subject: Biology And Life Sciences, Biochemistry And Molecular Biology Keywords: vegetables; high throughput phenotyping; genomic assisted breeding
Online: 26 April 2022 (06:00:45 CEST)
Conventional phenotyping breeding approaches for vegetable crops like Solanaceae, Bulb, Root crops, have made a significant contribution by developing many varieties. Despite this, conventional phenotyping approaches are not sufficient due to the longer time taken to develop a variety, low genetic gain, environmental factors and some other externalities that affect the phenotype-based selection. To address the challenges of conventional phenotype, a new recent method of high throughput phenotyping (HTP) is considered a promising tool. The development of high-throughput phenotyping technology began in the preceding decade as advancements in sensor, computer vision, automation, and advanced machine learning technologies. HTP platforms are being utilized to undertake non-destructive assessments of the complete plant system in a range of crops. HTP provides the precise measurements and suggests the collection of high-quality and accurate data which is necessary for standardizing phenotyping for the collection of genetic dissection and genomic assisted breeding such as genome-wide association studies (GWAS), linkage mapping, marker-assisted selection (MAS), genomic selection (GS). The remainder of this chapter discusses how high-throughput phenotyping technologies can be used in genomic-assisted breeding for vegetable crops
REVIEW | doi:10.20944/preprints202105.0340.v1
Subject: Biology And Life Sciences, Agricultural Science And Agronomy Keywords: Digital Biomarkers; Digital Phenotyping; Wearables; Sensors; Livestock
Online: 14 May 2021 (14:08:40 CEST)
Currently, large volumes of data are being collected on farms using multimodal sensor technol-ogies. These sensors measure the activity, housing conditions, feed intake, and health of farm animals. With traditional methods, the data from farm animals and their environment can be collected intermittently. However, with the advancement of wearable and non-invasive sensing tools, these measurements can be made in real-time for continuous quantitation relating to clinical biomarkers, resilience indicators, and behavioral predictors. The digital phenotyping of humans has drawn enormous attention recently due to its medical significance, but much research is still needed for the digital phenotyping of farm animals. Implications from human studies show great promise for the application of digital phenotyping technology in modern livestock farming, but these technologies must be directly applied to animals to understand their true capacities. Due to species-specific traits, certain technologies required to assess phenotypes need to be tailored ef-ficiently and accurately. Such devices allow for the collection of information that can better inform farmers on aspects of animal welfare and production that need improvement. By explicitly ad-dressing farm animals’ individual physiological and mental (affective states) needs, sensor-based digital phenotyping has the potential to serve as an effective intervention platform. Future re-search is warranted for the design and development of digital phenotyping technology platforms that create shared data standards, metrics, and repositories.
ARTICLE | doi:10.20944/preprints202205.0231.v1
Subject: Biology And Life Sciences, Agricultural Science And Agronomy Keywords: vegetation indices; precision farming; hybrid; phenotyping; remote sensing
Online: 17 May 2022 (12:47:44 CEST)
Abstract: Early assessment of crop development is a key aspect of precision agriculture. Shortening the time of response before a deficit of irrigation, nutrients and damage by diseases is one of the usual concerns in agriculture. Early prediction of crop yields can increase profitability in the farmer's economy. In this study we aimed to predict the yield of four maize commercial hybrids (Dekalb7508, Advanta9313, MH_INIA619 and Exp_05PMLM) using remotely sensed spectral vegetation indices (VI). A total of 10 VI (NDVI, GNDVI, GCI, RVI, NDRE, CIRE, CVI, MCARI, SAVI, and CCCI) were considered for evaluating crop yield and plant cover at 31, 39, 42, 46 and 51 days after sowing (DAS). A multivariate analysis was applied using principal component analysis (PCA), linear regression, and r-Pearson correlation. In the present study, highly significant correlations were found between plant cover with VIs at 46 (GNDVI, GCI, RVI, NDRE, CIRE and CCCI) and 51 DAS (GNDVI, GCI, NDRE, CIRE, CVI, MCARI and CCCI). The PCA indicated a clear discrimination of the dates evaluated with VIs at 31, 39 and 51 DAS. The inclusion of the CIRE and NDRE in the prediction model contributed to estimate the performance, showing greater precision at 51 DAS. The use of RPAS to monitor crops allows optimizing resources and helps in making timely decisions in agriculture in Peru.
ARTICLE | doi:10.20944/preprints201804.0209.v2
Subject: Computer Science And Mathematics, Data Structures, Algorithms And Complexity Keywords: plant phenotyping; noise filtering; binarization; accuracy evaluation; connected components
Online: 24 April 2018 (17:02:18 CEST)
Plants are such important keys of biological part of our environment, supply the human life and creatures. Understanding how the plant’s functions react with our surroundings, helps us better to make plant growth and development of food products. It means the plant phenotyping gives us bio information which needs some tools to reach the plant knowledge. Imaging tools is one of the phenotyping solutions which consists of imaging hardware such as the camera and image analysis software analyses the plant images changings such as plant growth rates. In this paper, we proposed a preprocessing algorithm to eliminate the noise and separate foreground from the background which results the plant image to help the plant image segmentation. The preprocessing is one of important levels has effect on better image segmentation and finally better plant’s image labeling and analysis. Our proposed algorithm is focused on removing noise such as converting the color space, applying the filters and local adaptive binarization step such as Niblack. Finally, we evaluate our algorithm with other algorithms by testing a variety of binarization methods.
REVIEW | doi:10.20944/preprints201712.0142.v1
Subject: Biology And Life Sciences, Plant Sciences Keywords: pre-harvest; ripeness; image analysis; machine learning; fruit phenotyping
Online: 20 December 2017 (09:35:36 CET)
Global food security for the increasing world population not only requires increased sustainable production of food but a significant reduction in pre- and post-harvest waste. The timing of when a fruit is harvested is critical for reducing waste along the supply chain and increasing fruit quality for consumers. The early in field assessment of fruit ripeness and prediction of the harvest date and yield by non-destructive technologies have the potential to revolutionize farming practices and enable the consumer to eat the tastiest and freshest fruit possible. A variety of non-destructive techniques have been applied to estimate the ripeness or maturity but not all of them are applicable for in situ (field or glasshouse) assessment. This review focuses on the non-destructive methods which are promising, or have already been, applied to the pre-harvest in field measurement including colorimetry, visible imaging, spectroscopy and spectroscopic imaging. Machine learning and regression models used in assessing ripeness are also discussed.
ARTICLE | doi:10.20944/preprints202202.0081.v1
Subject: Computer Science And Mathematics, Probability And Statistics Keywords: Online learning; Anomaly detection; Hotelling’s T-squared test; Digital phenotyping
Online: 7 February 2022 (11:54:54 CET)
To detect aberrant human behaviors from large volume of passive data collected by smartphones in real time, we propose an online anomaly detection method using Hotelling’s T-squared test. The test statistic is a weighted average, with more weight on the between-individual component when there are little data available for the individual and more weight on the within-individual component when the data are adequate. The algorithm takes only O(1) run time in each update and the required memory usage is fixed after a pre-specified number of updates. The performance of the proposed method, in terms of accuracy, sensitivity and specificity, is consistently better than or equal to the offline method that it builds upon depending on the sample size of the individual data.
ARTICLE | doi:10.20944/preprints202303.0147.v1
Subject: Medicine And Pharmacology, Oncology And Oncogenics Keywords: breast cancer; EMT; tumor microenvironment; collagen coatings; EMT-phenotyping; shape factors
Online: 8 March 2023 (06:29:48 CET)
During the progression from ductal carcinoma in situ (DCIS) to invasive breast cancer (IBC), cells have to overcome the physically restraining basement membrane (BM) which compartmentalizes the epithelium from the stroma. Since the extracellular matrix (ECM) of the epithelial and stromal compartment is biochemically and physically distinct from one another, the progression demands a certain degree of cellular plasticity being essentially required for a primary tumor to become invasive. The Epithelial-to-Mesenchymal Transition depicts such a cell program equipping cancer cells with features allowing for dissemination from the epithelial entity and stromal invasion on the single-cell level. Here, we investigated the reciprocal interference between an altering tumor microenvironment and the EMT-phenotype in vitro. BM-typical collagen IV and stroma-typical collagen I coatings were applied as provisional 2-D matrices. Pro-inflammatory growth factors were introduced to improve tissue mimicry. Whereas the growth on coated surfaces did only slightly affect the EMT-phenotype, the combinatorial action of collagen with growth factor TGF-β1 induced prominent phenotypic changes. However, the EMT-induction was independent of the collagen type and cellular accessibility for EMT-like changes was strongly cell line dependent. Summarizing the entire body of data, we computed an EMT-phenotyping model that was used to decide on cellular EMT-status and estimate EMT-like changes. We confirmed that miR200c-mediated reversion of mesenchymal MDA-MB-231 cells is reflected by our EMT-phenotype model emphasizing its potential to predict the therapeutic efficacy of EMT-targeting drugs in the future.
REVIEW | doi:10.20944/preprints202009.0308.v1
Subject: Biology And Life Sciences, Agricultural Science And Agronomy Keywords: Tomato; genetic breeding value; training population; genotyping; marker effect; phenotyping; selection schemes
Online: 14 September 2020 (00:08:23 CEST)
Genomic selection (GS) is a predictive approach that was build up to increase the rate of genetic gain per unit of time in breeding programs. It has emerged as a valuable method for improving complex traits that are controlled by many genes with small effect. GS enables the prediction of breeding value of candidate genotypes for selection. In this work we address important issues related to GS and its implementation in tomato breeding context. Genomic constrains and critical parameters affecting the accuracy of prediction in such crop such as phenotyping, genotyping training population composition and size and statistical method should be carefully evaluated. Comparison of GS approaches for facilitating the selection of tomato superior genotypes during breeding program are also discussed. GS applied to tomato breeding has already shown to be feasible. We illustrated how GS can improve the rate of gain in elite lines selection, descendent and in backcross schemes. The GS schemes begin to be delineated and computer science can provide support for future selection strategies. A new breeding framework is beginning to emerge for optimizing tomato improvement procedures.
ARTICLE | doi:10.20944/preprints201902.0126.v1
Subject: Engineering, Electrical And Electronic Engineering Keywords: microwave technique; transmittance; soil moisture; microstrip patch antennas; rhizobox; roots; plant phenotyping
Online: 13 February 2019 (16:49:06 CET)
Interactions of soil moisture with plant root systems are very important for plant growth. For non-invasive determination of volumetric soil moisture in a rhizobox, a microwave system based on transmittance of electromagnetic waves in the microwave frequency range was developed using microstrip patch antennas. Vector Network Analyzers (VNAs) were used to measure the S-parameters at frequency ranges close to 5 GHz. A transmission system with microstrip patch antennas was developed. The result of this attenuation is in the frequency domain. The antennae were designed as resonant microstrip antennae. The antennae were placed on both sides of a rhizobox, which allowed non-invasive measuring soil moisture in the box. The attenuation (S21(dB)) was used to measure the effect of temperature, and different types of soil; as well as sensitivity, reproducibility and repeatability of the system. In this work we present quantitative results of soil moisture in rhizobox. The microwave technique, using microstrip patch antennas, is a reliable and accurate system, and showed very promising potential applications for rhizobox-based investigations of root performance.
ARTICLE | doi:10.20944/preprints202208.0116.v1
Subject: Biology And Life Sciences, Agricultural Science And Agronomy Keywords: Biomass partitioning; Digital root phenotyping; Image analysis; Rhizotron; Root architecture; Root phenes; RootSnap
Online: 5 August 2022 (04:23:44 CEST)
Citron watermelon (Citrullus lanatus var. citroides) is an extremely drought-tolerant cucurbit crop widely grown in sub-Saharan Africa in arid and semi-arid environments characterized by drought. The species is a C3 xerophyte used for multiple purposes, including intercropping with maize and has a deep taproot system. The deep taproot system plays a key role in the species’ adaptation to dry conditions. Understanding root system development of this crop could be useful in identifying traits for breeding water-use efficient and drought-tolerant varieties. This study compared root system architecture of citron watermelon accessions under water-stress conditions. Nine selected and drought-tolerant citron watermelon accessions were grown under non-stress (NS) and water stress (WS) conditions using the root rhizotron procedure in a glasshouse. The following root system architecture (RSA) traits were measured, namely: root system width (RSW), root system depth (RSD), convex hull area (CHA), total root length (TRL), root branch count (RBC), total root volume (TRV), leaf area (LA), leaf number (LN), first seminal root length (FSRL), seminal root angle (SRA), root dry mass (RDM), shoot dry mass (SDM), root–shoot mass ratio (RSM), root mass ratio (RMR), shoot mass ratio (SMR) and root tissue density (RTD). The data collected on RSA traits were subjected to the analysis of variance (ANOVA), correlation and principal component analyses. ANOVA revealed a significant (p < 0.05) accession × water stress interaction effect for studied RSA traits. Under WS, RDM exhibited significant and positive correlations with RSM (r = 0.65), RMR (r = 0.66), RSD (r = 0.66), TRL (r = 0.60), RBC (r = 0.72), FSRL (r = 0.73) and LN (r = 0.70). Principal component analysis revealed high loading scores for the following RSA traits: RSW (0.89), RSD (0.97), TRL (0.99), TRV (0.90), TRL (0.99), RMR (0.96) and RDM (0.76). In conclusion, the study has shown that the identified RSA traits could be useful in crop improvement programmes for citron watermelon genotypes with enhanced drought adaptation for improved yield performance under drought-prone environments.
ARTICLE | doi:10.20944/preprints202111.0424.v1
Subject: Biology And Life Sciences, Plant Sciences Keywords: Internet of things; Raspberry Pi; LiDAR; GNSS; High-throughput plant phenotyping; Precision agriculture
Online: 23 November 2021 (14:15:26 CET)
Phenotypic characterization of crop genotypes is an essential yet challenging aspect of crop management and agriculture research. Digital sensing technologies are rapidly advancing plant phenotyping and speeding-up crop breeding outcomes. However, off-the-shelf sensors might not be fully applicable and suitable for agriculture research due to diversity in crop species and specific needs during plant breeding selections. Customized sensing systems with specialized sensor hardware and software architecture provide a powerful and low-cost solution. This study designed and developed a fully integrated Raspberry Pi-based LiDAR sensor named CropBioMass (CBM), enabled by internet of things to provide a complete end-to-end pipeline. The CBM is a low-cost sensor, provides high-throughput seamless data collection in field, small data footprint, injection of data onto the remote server, and automated data processing. Phenotypic traits of crop fresh biomass, dry biomass, and plant height estimated by CBM data had high correlation with ground truth manual measurements in wheat field trial. The CBM is readily applicable for high-throughput plant phenotyping, crop monitoring, and management for precision agricultural applications.
ARTICLE | doi:10.20944/preprints202112.0325.v1
Subject: Biology And Life Sciences, Agricultural Science And Agronomy Keywords: phenotyping; proximal sensing; reflectance imaging; vegetation indices; hyperspectral reflectance; chlorophylls; carotenoids; anthocyanins; senescence; ripening
Online: 21 December 2021 (12:23:13 CET)
Hyperspectral reflectance imaging is an emerging method for rapid non-invasive quantitative screening of plant traits. This method is essential for high-throughput phenotyping and hence for accelerated breeding of crop plants as well as for precision agriculture practices. However, extraction of sensible information from reflectance images is hindered by the complexity of plant optical properties, especially when they are measured in the field. We propose using reflectance indices (Plant Senescence Reflectance Index, PSRI; Anthocyanin Reflectance Index, ARI; and spectral deconvolution) previously developed for remote sensing of vegetation and point-based reflectometers to infer the spatially resolved information on plant development and biochemical composition using ripening apple fruit as the model. Specifically, the proposed approach enables capturing data on distribution of chlorophylls and primary carotenoids as well as secondary carotenoids (both linked with fruit ripening and leaf senescence during plant development) as well as the information on spatial distribution of anthocyanins (known as stress pigments) over the plant surface. We argue that the proposed approach would enrich the phenotype assessments made on the base of reflectance image analysis with valuable information on plant physiological condition, stress acclimation state, and the progression of the plant development.
ARTICLE | doi:10.20944/preprints202009.0381.v1
Subject: Biology And Life Sciences, Biology And Biotechnology Keywords: high throughput screening; rapid phenotyping; model-based experimental design; Escherichia coli; automated bioprocess development
Online: 17 September 2020 (07:34:19 CEST)
In bioprocess development, the host and the genetic construct for a new biomanufacturing process are selected in the early developmental stages. This decision, made at the screening scale with very limited information about the performance of the selected cell factory in larger reactors, has a major influence on the performance of the final process. To overcome this, scaledown approaches are essential to run screenings that show the real cell factory performance at industrial like conditions. We present a fully automated robotic facility with 24 parallel mini-bioreactors that is operated by a model based adaptive input design framework for the characterization of clone libraries under scale-down conditions. The cultivation operation strategies are computed and continuously refined based on a macro-kinetic growth model that is continuously re-fitted to the available experimental data. The added value of the approach is demonstrated with 24 parallel fed-batch cultivations in a mini-bioreactor system with eight different Escherichia coli strains in triplicate. The 24 fed-batches ran under the desired conditions generating sufficient information to define the fastest growing strain in an environment with varying glucose concentrations similar to industrial scale bioreactors.
ARTICLE | doi:10.20944/preprints201805.0463.v1
Subject: Medicine And Pharmacology, Other Keywords: undiagnosed rare diseases; diagnostic odyssey; NGS; deep phenotyping; genomic matchmaking; secondary findings; patient involvement
Online: 31 May 2018 (09:35:32 CEST)
The time required to reach a correct diagnosis is one of the most important problems for rare disease (RD) patients. Diagnostic delay can be intolerably long, to the point that it is usually described as a “diagnostic odyssey” and, sometimes, a diagnosis might never occur. The International Rare Disease Research Consortium proposed, as ultimate goal for 2017-2027, to enable all people with a suspected RD to be diagnosed within one year if the disorder is known, and to enter a globally coordinated diagnostic and research pipeline for the unsolved cases. In-depth analysis of the genotype through next generation sequencing, together with a standardized in-depth phenotype description and sophisticated high-throughput approaches, have been applied as diagnostic tools to increase the chance of a timely and accurate diagnosis. This approach is very fruitful as, according to the Orphanet database, from 2010 to March 2017 more than 600 new RDs have been described and about 3600 genes linked to RDs have been identified. However, combination of -omics and phenotype data and international sharing of this information raise ethical concerns. Values to be assessed include not only patient autonomy but also family implications, beneficence, non-maleficence, justice, solidarity and reciprocity, which must be respected and promoted and, at the same time, balanced among each other. In this work we suggest that, to maximise patients involvement in the search for a diagnosis and identification of new causative genes, undiagnosed patients should have the possibility to: 1) actively participate in the description of their phenotype; 2) choose the level of visibility of their profile in matchmaking databases; 3) express their preferences regarding return of new findings, in particular which level of Variant of Unknown Significance (VUS) significance should be considered relevant to them. The quality of the relation between individual patients and physicians, and between the patient community and the scientific community is critically important for making the best use of available data and combining efforts in the search for matches with similar cases worldwide that will help to solve unsolved cases. The contribution of patients to collecting and coding data comprehensively is critical for efficient use of data downstream of data collection.
ARTICLE | doi:10.20944/preprints202210.0477.v1
Subject: Computer Science And Mathematics, Analysis Keywords: High Throughput Plant Phenotyping; Deep Neural Network; Flower Detection; Temporal Phenotypes; Benchmark Dataset; Flower Status Report
Online: 31 October 2022 (10:00:24 CET)
A phenotype is the composite of an observable expression of a genome for traits in a given environment. The trajectories of phenotypes computed from an image sequence and timing of important events in a plant’s life cycle can be viewed as temporal phenotypes and indicative of the plant’s growth pattern and vigor. In this paper, we introduce a novel method called FlowerPhenoNet which uses deep neural networks for detecting flowers from multiview image sequences for high throughput temporal plant phenotyping analysis. Following flower detection, a set of novel flower-based phenotypes are computed, e.g., the day of emergence of the first flower in a plant’s life cycle, the total number of flowers present in the plant at a given time, the highest number of flowers bloomed in the plant, growth trajectory of a flower and the blooming trajectory of a plant. To develop a new algorithm and facilitate performance evaluation based on experimental analysis, a benchmark dataset is indispensable. Thus, we introduce a benchmark dataset called FlowerPheno which comprises image sequences of three flowering plant species, e.g., sunflower, coleus, and canna, captured by a visible light camera in a high throughput plant phenotyping platform from multiple view angles. The experimental analyses on the FlowerPheno dataset demonstrate the efficacy of the FlowerPhenoNet.
Subject: Biology And Life Sciences, Agricultural Science And Agronomy Keywords: automated machine learning; Neural Architecture Search; high-throughput plant phenotyping; wheat lodging assessment; unmanned aerial vehicle.
Online: 1 February 2021 (14:11:08 CET)
Automated machine learning (AutoML) has been heralded as the next wave in artificial intelligence with its promise to deliver high performance end-to-end machine learning pipelines with minimal effort from the user. However, despite AutoML showing great promise for computer vision tasks, to the best of our knowledge, no study has used AutoML for image-based plant phenotyping. To address this gap in knowledge, we examined the application of AutoML for image-based plant phenotyping using wheat lodging assessment with UAV imagery as an example. We compared the performance of an open-source AutoML framework, AutoKeras in image classification and regression tasks to transfer learning using modern convolutional neural network (CNN) architectures. For image classification which classified plot images as lodged or non-lodged, transfer learning with Xception and DenseNet-201 achieved best classification accuracy of 93.2%, whereas Autokeras had 92.4% accuracy. For image regression which predicted lodging scores from plot images, transfer learning with DenseNet-201 had the best performance (R2=0.8303, RMSE=9.55, MAE=7.03, MAPE=12.54%), followed closely by AutoKeras (R2=0.8273, RMSE=10.65, MAE=8.24, MAPE=13.87%). Interestingly, in both tasks, AutoKeras models had up to 40-fold faster inference times compared to the pretrained CNNs. The merits and drawbacks of AutoML compared to transfer learning for image-based plant phenotyping are discussed.
ARTICLE | doi:10.20944/preprints201712.0108.v1
Subject: Computer Science And Mathematics, Computer Science Keywords: Plant phenotyping, Plant pixel classification, Colour space, , Gaussian mixture model, Earth mover distance, Variance ratio, Plant segmentation.
Online: 15 December 2017 (16:52:23 CET)
Segmentation of a region of interest is an important pre-processing step for many colour image analysis techniques. Similarly segmentation of plant in digital images is an important preprocessing step in phenotying plants by image analysis. In this paper we present an analytical study to statistically determine the suitability of colour space representation of an image to best detect plant pixels and separate them from background pixels. Our hypothesis is that the colour space representation in which the separation of the distributions representing plant pixels and background pixels is maximized would be the best for detection of plant pixels. The two classes of pixels are modelled as a Gaussian mixture model (GMM). In our GM modelling we don't make any prior assumption about the number of Gaussians in the model. Rather a constant bandwidth mean-shift filter is used to cluster the data and the number of clusters and hence the number of Gaussians is automatically determined. Here we have analysed following representative colour spaces like $RGB$, $rgb$, $HSV$, $Ycbcr$ and $CIE-Lab$. This is because these colour spaces represent several other similar colour spaces and also an exhaustive study of all the colour space will be too voluminous. We also analyse the colour space feature from the two-class variance ratio perspective and compare the results of our hypothesis with this metric. The dataset for this empirical study consist of 378 digital images of plants and their manual segmentation. Dataset consist of various species of plants (arabidopsi, tobacco, wheat, rye grass etc.) imaged under different lighting conditions, indoor and outdoor, controlled and uncontrolled background. In results we obtain better segmentation of the plants in $HSV$ colour space, which is supported by its Earth mover distance (EMD) on the GMM distribution of plant and background pixels.
ARTICLE | doi:10.20944/preprints202104.0755.v1
Subject: Biology And Life Sciences, Biochemistry And Molecular Biology Keywords: wheat; leaf rust; powdery mildew; septoria; stem rust; yellow rust; image recognition; deep learning; convolutional neural network; phenotyping
Online: 28 April 2021 (15:35:37 CEST)
Diseases of cereals caused by pathogenic fungi can significantly reduce crop yields. Many cultures are exposed to them. The disease is difficult to control on a large scale, thus one of the relevant approaches is the crop field monitoring, which helps to identify the disease at an early stage and take measures to prevent its spread. One of the effective control methods is disease identification based on the analysis of digital images with the possibility of obtaining them in field conditions using mobile devices. In this work, we propose a method for the recognition of five fungal diseases of wheat shoots (leaf rust, stem rust, yellow rust, powdery mildew, and septoria), both separately and in combination, with the possibility of identifying the stage of plant development. A set of 2414 images of wheat fungi diseases (WFD2020) was generated, for which expert labeling was performed by the type of disease. WFD2020 data are available freely at http://wfd.sysbio.ru/. In the process of creating this set, a method was applied to reduce the degeneracy of the training data based on the image hashing algorithm. The disease recognition algorithm is based on the convolutional neural network with the EfficientNet architecture. The best accuracy (0.942) was shown by a network with a training strategy based on augmentation and transfer of image styles. The recognition method was implemented by the authors as a bot on the Telegram platform, which allows assessing plants by lesions in the field conditions.
ARTICLE | doi:10.20944/preprints202206.0120.v1
Subject: Environmental And Earth Sciences, Environmental Science Keywords: Miscanthus; remote sensing; UAV; multispectral images; high-throughput phenotyping; machine learning; yield prediction; trait estimation; PROSAIL; multi-sensor interoperability
Online: 8 June 2022 (09:44:59 CEST)
Miscanthus holds a great potential in the frame of the bioeconomy and yield prediction can help improving Miscanthus logistic supply chain. Breeding programs in several countries are attempting to produce high-yielding Miscanthus hybrids better adapted to different climates and end-uses. Multispectral images acquired from unmanned aerial vehicles (UAVs) in Italy and in the UK in 2021 and 2022 were used to investigate the feasibility of high-throughput phenotyping (HTP) of novel Miscanthus hybrids for yield prediction and crop traits estimation. An intercalibration procedure was performed using simulated data from the PROSAIL model to link vegetation indices (VIs) derived from two different multispectral sensors. Random forest algorithm estimated with good accuracy yield traits (light interception, plant height, green leaf biomass and standing biomass) using VIs time series and predicted yield using peak descriptor derived from VIs time series with 2.3 Mg DM ha-1 of RMSE. The study demonstrates the potential of UAVs multispectral in HTP applications and in yield prediction for providing important information needed to increase sustainable biomass production.
REVIEW | doi:10.20944/preprints202202.0048.v1
Subject: Biology And Life Sciences, Biochemistry And Molecular Biology Keywords: Plant Breeding; Speed Breeding; Training Population; Field Design; Multi-Environment; Multi-Trait; Deep Learning; High-Throughput Phenotyping; Genetic Gain
Online: 3 February 2022 (10:41:44 CET)
Plant geneticists and breeders have used marker technology since the 1980s in quantitative trait locus (QTL) identification. Marker-assisted selection is effective for large-effect QTL but has been challenging to use with quantitative traits controlled by multiple minor effect alleles. Therefore, genomic selection (GS) was proposed to estimate all markers simultaneously, thereby capturing all their effects. However, breeding programs are still struggling to identify the best strategy to implement it into their programs. Traditional breeding programs need to be optimized to implement GS effectively. This review explores the optimization of breeding programs for variety release based on aspects of the breeder’s equation. Optimizations include reorganizing field designs, training populations, increasing the number of lines evaluated, and leveraging the large amount of genomic and phenotypic data collected across different growing seasons and environments to increase heritability estimates, selection intensity, and selection accuracy. Breeding programs can leverage their phenotypic and genotypic data to maximize genetic gain and selection accuracy through GS methods utilizing multi-trait and, multi-environment models, high-throughput phenotyping, and deep learning approaches. Overall, this review describes various methods that plant breeders can utilize to increase genetic gains and effectively implement GS in breeding .
ARTICLE | doi:10.20944/preprints201810.0664.v1
Subject: Computer Science And Mathematics, Hardware And Architecture Keywords: RGB-D sensors; empirical analysis; sensors in agriculture; phenotyping; microsoft kinect; Intel D-435; Intel SR300; ORBBEC ASTRA S
Online: 29 October 2018 (09:15:45 CET)
Phenotyping is the task of measuring plant attributes for analyzing the current state of the plant. In agriculture, phenotyping can be used to make decisions concerning the management of crops, such as the watering policy, or whether to spray for a certain pest. Currently, large scale phenotyping in fields is typically done using manual labor, which is a costly, low throughput process. Researchers often advocate the use of automated systems for phenotyping, relying on the use of sensors for making measurements. The recent rise of low cost, yet reasonably accurate, RGB-D sensors has opened the way for using these sensors in field phenotyping applications. In this paper, we investigate the applicability of 4 different RGB-D sensors for this task. We conduct an outdoor experiment, measuring plant attribute in various distances and light conditions. Our results show that modern RGB-D sensors, in particular, the Intel D435 sensor, provides a viable tool for close range phenotyping tasks in fields.
REVIEW | doi:10.20944/preprints202206.0040.v1
Subject: Chemistry And Materials Science, Analytical Chemistry Keywords: Microsampling; sample miniaturisation; dried blood spot (DBS); dried plasma spot (DPS); dried serum spot (DSS); metabolic phenotyping; gas chromatography-mass spectrometry (GC-MS); liquid chromatog-raphy-mass spectrometry (LC-MS)
Online: 3 June 2022 (09:53:13 CEST)
Microsamples (collections usually less than 50 µL) have been introduced in pre-clinical, clinical, and research settings to overcome obstacles in sampling via traditional venipuncture. However, venipuncture remains the sampling gold standard for metabolic phenotyping of blood. This pre-sents several challenges in metabolic phenotyping workflows: accessibility for individuals in ru-ral and remote underserved areas (due to the need for trained personnel), the unamenable nature to frequent sampling protocols in longitudinal research (for its invasive nature), and sample col-lection difficulty in the young and elderly. Furthermore, venous sample stability may be compro-mised when temperate conditions necessary for cold-chain transport are beyond control. Alter-natively, research utilising microsamples extends phenotyping possibilities to inborn errors of metabolism, therapeutic drug monitoring, nutrition, as well as sport and anti-doping. Although the application of microsamples in metabolic phenotyping exists, it is still in its infancy, with whole blood being overwhelmingly the primary biofluid collected through the collection method of dried blood spots. Research into metabolic phenotyping of microsamples is limited; however, with advances in commercially available microsampling devices, common barriers such as volumetric inaccuracies and the ‘haematocrit effect’ in dried blood spot microsampling can be overcome. In this review, we provide an overview of the common uses and workflows for mi-crosampling in metabolic phenotyping research. We discuss the advancements in technologies, highlighting key considerations and remaining knowledge gaps for employment of microsamples in metabolic phenotyping research. Supporting the translation of research from the ‘bench to the community’.
ARTICLE | doi:10.20944/preprints202304.0344.v1
Subject: Medicine And Pharmacology, Epidemiology And Infectious Diseases Keywords: Deep phenotyping; AI; 2D and 3D facial scans; Genetic diseases; Early treatment; Big data; 2D and 3D facial scans; Facial traits; Healthcare; Citizen privacy; Ethical concerns; EU legal framework; Forensic medicine; Orwellian ramifications
Online: 14 April 2023 (03:56:59 CEST)
One in 12 babies is born with a rare genetic disease. Sadly, most cases are undetected until later age, missing time for early treatment and opportunity to prevent complications. Humanity has entered a new era where Big Data collected by governments, including 2D and 3D facial scans, are available. Many rare genetic diseases can be identified by artificial intelligence (AI) analysis of the facial photo. Phenotyping AI utilizations facilitate comprehensive and accurate genetic evaluations. AI processing of this Big Data to identify rare genetic diseases could bring unimaginable benefits to healthcare, although this would be a questionable step in terms of citizen privacy and could lead to future "Orwellian" ramifications with government abuse. Going forward, a balance must be found between protecting the privacy of citizens and the enticing use of AI for their health risks and cost savings through prevention. The unimaginable potential of AI early diagnostics from facial photos also raises various ethical and legal concerns. This paper presents concept, protentional methods and legal and other limitations within EU legal framework in contrast with potential benefits. This paper is focused on AI utilization to early diagnostic of rare genetic diseases. Shift of paradigm in the screening for rare genetic diseases in population with AI face analysis is expected to have a significant impact. The potential of AI algorithms utilizations similar to face2gene app in general population or systematically on Big governmental datasets recording facial traits changes in time can have significant impact on public health but at the same time give raise to profound concern as violation of one’s privacy.