Out-of-Distribution Detection Algorithms for Robust Insect Classification.

Plants encounter a variety of beneficial and harmful insects during their growth cycle. Accurate identification (i.e., detecting insects' presence) and classification (i.e., determining the type or class) of these insect species is critical for implementing prompt and suitable mitigation strategies. Such timely actions carry substantial economic and environmental implications. Deep learning-based approaches have produced models with good insect classification accuracy. Researchers aim to implement identification and classification models in agriculture, facing challenges when input images markedly deviate from the training distribution (e.g., images like vehicles, humans, or a blurred image or insect class that is not yet trained on). Out-of-distribution (OOD) detection algorithms provide an exciting avenue to overcome these challenges as they ensure that a model abstains from making incorrect classification predictions on images that belong to non-insect and/or untrained insect classes. As far as we know, no prior in-depth exploration has been conducted on the role of the OOD detection algorithms in addressing agricultural issues. Here, we generate and evaluate the performance of state-of-the-art OOD algorithms on insect detection classifiers. These algorithms represent a diversity of methods for addressing an OOD problem. Specifically, we focus on extrusive algorithms, i.e., algorithms that wrap around a well-trained classifier without the need for additional co-training. We compared three OOD detection algorithms: (a) maximum softmax probability, which uses the softmax value as a confidence score; (b) Mahalanobis distance (MAH)-based algorithm, which uses a generative classification approach; and (c) energy-based algorithm, which maps the input data to a scalar value, called energy. We performed an extensive series of evaluations of these OOD algorithms across three performance axes: (a) Base model accuracy: How does the accuracy of the classifier impact OOD performance? (b) How does the level of dissimilarity to the domain impact OOD performance? (c) Data imbalance: How sensitive is OOD performance to the imbalance in per-class sample size? Evaluating OOD algorithms across these performance axes provides practical guidelines to ensure the robust performance of well-trained models in the wild, which is a key consideration for agricultural applications. Based on this analysis, we proposed the most effective OOD algorithm as wrapper for the insect classifier with highest accuracy. We presented the results of its OOD detection performance in the paper. Our results indicate that OOD detection algorithms can significantly enhance user trust in insect pest classification by abstaining classification under uncertain conditions.

Using Machine Learning to Develop a Fully Automated Soybean Nodule Acquisition Pipeline (SNAP).

Nodules form on plant roots through the symbiotic relationship between soybean (Glycine max L. Merr.) roots and bacteria (Bradyrhizobium japonicum) and are an important structure where atmospheric nitrogen (N2) is fixed into bioavailable ammonia (NH3) for plant growth and development. Nodule quantification on soybean roots is a laborious and tedious task; therefore, assessment is frequently done on a numerical scale that allows for rapid phenotyping, but is less informative and suffers from subjectivity. We report the Soybean Nodule Acquisition Pipeline (SNAP) for nodule quantification that combines RetinaNet and UNet deep learning architectures for object (i.e., nodule) detection and segmentation. SNAP was built using data from 691 unique roots from diverse soybean genotypes, vegetative growth stages, and field locations and has a good model fit (R 2 = 0.99). SNAP reduces the human labor and inconsistencies of counting nodules, while acquiring quantifiable traits related to nodule growth, location, and distribution on roots. The ability of SNAP to phenotype nodules on soybean roots at a higher throughput enables researchers to assess the genetic and environmental factors, and their interactions on nodulation from an early development stage. The application of SNAP in research and breeding pipelines may lead to more nitrogen use efficiency for soybean and other legume species cultivars, as well as enhanced insight into the plant-Bradyrhizobium relationship.

Dissecting the Root Phenotypic and Genotypic Variability of the Iowa Mung Bean Diversity Panel.

Mung bean [Vigna radiata (L.) Wilczek] is a drought-tolerant, short-duration crop, and a rich source of protein and other valuable minerals, vitamins, and antioxidants. The main objectives of this research were (1) to study the root traits related with the phenotypic and genetic diversity of 375 mung bean genotypes of the Iowa (IA) diversity panel and (2) to conduct genome-wide association studies of root-related traits using the Automated Root Image Analysis (ARIA) software. We collected over 9,000 digital images at three-time points (days 12, 15, and 18 after germination). A broad sense heritability for days 15 (0.22-0.73) and 18 (0.23-0.87) was higher than that for day 12 (0.24-0.51). We also reported root ideotype classification, i.e., PI425425 (India), PI425045 (Philippines), PI425551 (Korea), PI264686 (Philippines), and PI425085 (Sri Lanka) that emerged as the top five in the topsoil foraging category, while PI425594 (unknown origin), PI425599 (Thailand), PI425610 (Afghanistan), PI425485 (India), and AVMU0201 (Taiwan) were top five in the drought-tolerant and nutrient uptake "steep, cheap, and deep" ideotype. We identified promising genotypes that can help diversify the gene pool of mung bean breeding stocks and will be useful for further field testing. Using association studies, we identified markers showing significant associations with the lateral root angle (LRA) on chromosomes 2, 6, 7, and 11, length distribution (LED) on chromosome 8, and total root length-growth rate (TRL_GR), volume (VOL), and total dry weight (TDW) on chromosomes 3 and 5. We discussed genes that are potential candidates from these regions. We reported beta-galactosidase 3 associated with the LRA, which has previously been implicated in the adventitious root development via transcriptomic studies in mung bean. Results from this work on the phenotypic characterization, root-based ideotype categories, and significant molecular markers associated with important traits will be useful for the marker-assisted selection and mung bean improvement through breeding.

Soybean Root System Architecture Trait Study through Genotypic, Phenotypic, and Shape-Based Clusters.

We report a root system architecture (RSA) traits examination of a larger scale soybean accession set to study trait genetic diversity. Suffering from the limitation of scale, scope, and susceptibility to measurement variation, RSA traits are tedious to phenotype. Combining 35,448 SNPs with an imaging phenotyping platform, 292 accessions (replications = 14) were studied for RSA traits to decipher the genetic diversity. Based on literature search for root shape and morphology parameters, we used an ideotype-based approach to develop informative root (iRoot) categories using root traits. The RSA traits displayed genetic variability for root shape, length, number, mass, and angle. Soybean accessions clustered into eight genotype- and phenotype-based clusters and displayed similarity. Genotype-based clusters correlated with geographical origins. SNP profiles indicated that much of US origin genotypes lack genetic diversity for RSA traits, while diverse accession could infuse useful genetic variation for these traits. Shape-based clusters were created by integrating convolution neural net and Fourier transformation methods, enabling trait cataloging for breeding and research applications. The combination of genetic and phenotypic analyses in conjunction with machine learning and mathematical models provides opportunities for targeted root trait breeding efforts to maximize the beneficial genetic diversity for future genetic gains.

A farm-level precision land management framework based on integer programming.

Farmland management involves several planning and decision making tasks including seed selection and irrigation management. A farm-level precision farmland management model based on mixed integer linear programming is proposed in this study. Optimal decisions are designed for pre-season planning of crops and irrigation water allocation. The model captures the effect of size and shape of decision scale as well as special irrigation patterns. The authors illustrate the model with a case study on a farm in the state of California in the U.S. and show the model can capture the impact of precision farm management on profitability. The results show that threefold increase of annual net profit for farmers could be achieved by carefully choosing irrigation and seed selection. Although farmers could increase profits by applying precision management to seed or irrigation alone, profit increase is more significant if farmers apply precision management on seed and irrigation simultaneously. The proposed model can also serve as a risk analysis tool for farmers facing seasonal irrigation water limits as well as a quantitative tool to explore the impact of precision agriculture.

10,000-fold concentration increase of the biomarker cardiac troponin I in a reducing union microfluidic chip using cationic isotachophoresis.

This paper describes the preconcentration of the biomarker cardiac troponin I (cTnI) and a fluorescent protein (R-phycoerythrin) using cationic isotachophoresis (ITP) in a 3.9 cm long poly(methyl methacrylate) (PMMA) microfluidic chip. The microfluidic chip includes a channel with a 5× reduction in depth and a 10× reduction in width. Thus, the overall cross-sectional area decreases by 50× from inlet (anode) to outlet (cathode). The concentration is inversely proportional to the cross-sectional area so that as proteins migrate through the reductions, the concentrations increase proportionally. In addition, the proteins gain additional concentration by ITP. We observe that by performing ITP in a cross-sectional area reducing microfluidic chip we can attain concentration factors greater than 10,000. The starting concentration of cTnI was 2.3 µg mL⁻¹ and the final concentration after ITP concentration in the microfluidic chip was 25.52 ± 1.25 mg mL⁻¹. To the author's knowledge this is the first attempt at concentrating the cardiac biomarker cTnI by ITP. This experimental approach could be coupled to an immunoassay based technique and has the potential to lower limits of detection, increase sensitivity, and quantify different isolated cTnI phosphorylation states.

10,000-fold concentration increase in proteins in a cascade microchip using anionic ITP by a 3-D numerical simulation with experimental results.

This paper describes both the experimental application and 3-D numerical simulation of isotachophoresis (ITP) in a 3.2 cm long "cascade" poly(methyl methacrylate) (PMMA) microfluidic chip. The microchip includes 10 × reductions in both the width and depth of the microchannel, which decreases the overall cross-sectional area by a factor of 100 between the inlet (cathode) and outlet (anode). A 3-D numerical simulation of ITP is outlined and is a first example of an ITP simulation in three dimensions. The 3-D numerical simulation uses COMSOL Multiphysics v4.0a to concentrate two generic proteins and monitor protein migration through the microchannel. In performing an ITP simulation on this microchip platform, we observe an increase in concentration by over a factor of more than 10,000 due to the combination of ITP stacking and the reduction in cross-sectional area. Two fluorescent proteins, green fluorescent protein and R-phycoerythrin, were used to experimentally visualize ITP through the fabricated microfluidic chip. The initial concentration of each protein in the sample was 1.995 µg/mL and, after preconcentration by ITP, the final concentrations of the two fluorescent proteins were 32.57 ± 3.63 and 22.81 ± 4.61 mg/mL, respectively. Thus, experimentally the two fluorescent proteins were concentrated by over a factor of 10,000 and show good qualitative agreement with our simulation results.

Chemically modified solid state nanopores for high throughput nanoparticle separation.

The separation of biomolecules and other nanoparticles is a vital step in several analytical and diagnostic techniques. Towards this end we present a solid state nanopore-based set-up as an efficient separation platform. The translocation of charged particles through a nanopore was first modeled mathematically using the multi-ion model and the surface charge density of the nanopore membrane was identified as a critical parameter that determines the selectivity of the membrane and the throughput of the separation process. Drawing from these simulations a single 150 nm pore was fabricated in a 50 nm thick free-standing silicon nitride membrane by focused-ion-beam milling and was chemically modified with (3-aminopropyl)triethoxysilane to change its surface charge density. This chemically modified membrane was then used to separate 22 and 58 nm polystyrene nanoparticles in solution. Once optimized, this approach can readily be scaled up to nanopore arrays which would function as a key component of next-generation nanosieving systems.