Boosting precision crop protection towards agriculture 5.0 via machine learning and emerging technologies: A contextual review.

Crop protection is a key activity for the sustainability and feasibility of agriculture in a current context of climate change, which is causing the destabilization of agricultural practices and an increase in the incidence of current or invasive pests, and a growing world population that requires guaranteeing the food supply chain and ensuring food security. In view of these events, this article provides a contextual review in six sections on the role of artificial intelligence (AI), machine learning (ML) and other emerging technologies to solve current and future challenges of crop protection. Over time, crop protection has progressed from a primitive agriculture 1.0 (Ag1.0) through various technological developments to reach a level of maturity closelyin line with Ag5.0 (section 1), which is characterized by successfully leveraging ML capacity and modern agricultural devices and machines that perceive, analyze and actuate following the main stages of precision crop protection (section 2). Section 3 presents a taxonomy of ML algorithms that support the development and implementation of precision crop protection, while section 4 analyses the scientific impact of ML on the basis of an extensive bibliometric study of >120 algorithms, outlining the most widely used ML and deep learning (DL) techniques currently applied in relevant case studies on the detection and control of crop diseases, weeds and plagues. Section 5 describes 39 emerging technologies in the fields of smart sensors and other advanced hardware devices, telecommunications, proximal and remote sensing, and AI-based robotics that will foreseeably lead the next generation of perception-based, decision-making and actuation systems for digitized, smart and real-time crop protection in a realistic Ag5.0. Finally, section 6 highlights the main conclusions and final remarks.

Network topological determinants of pathogen spread.

How do we best constrain social interactions to decrease transmission of communicable diseases? Indiscriminate suppression is unsustainable long term and presupposes that all interactions carry equal importance. Instead, transmission within a social network has been shown to be determined by its topology. In this paper, we deploy simulations to understand and quantify the impact on disease transmission of a set of topological network features, building a dataset of 9000 interaction graphs using generators of different types of synthetic social networks. Independently of the topology of the network, we maintain constant the total volume of social interactions in our simulations, to show how even with the same social contact some network structures are more or less resilient to the spread. We find a suitable intervention to be specific suppression of unfamiliar and casual interactions that contribute to the network's global efficiency. This is, pathogen spread is significantly reduced by limiting specific kinds of contact rather than their global number. Our numerical studies might inspire further investigation in connection to public health, as an integrative framework to craft and evaluate social interventions in communicable diseases with different social graphs or as a highlight of network metrics that should be captured in social studies.

Seasonal Arctic sea ice forecasting with probabilistic deep learning.

Anthropogenic warming has led to an unprecedented year-round reduction in Arctic sea ice extent. This has far-reaching consequences for indigenous and local communities, polar ecosystems, and global climate, motivating the need for accurate seasonal sea ice forecasts. While physics-based dynamical models can successfully forecast sea ice concentration several weeks ahead, they struggle to outperform simple statistical benchmarks at longer lead times. We present a probabilistic, deep learning sea ice forecasting system, IceNet. The system has been trained on climate simulations and observational data to forecast the next 6 months of monthly-averaged sea ice concentration maps. We show that IceNet advances the range of accurate sea ice forecasts, outperforming a state-of-the-art dynamical model in seasonal forecasts of summer sea ice, particularly for extreme sea ice events. This step-change in sea ice forecasting ability brings us closer to conservation tools that mitigate risks associated with rapid sea ice loss.

Spatio-chromatic contrast sensitivity under mesopic and photopic light levels.

Contrast sensitivity functions (CSFs) characterize the sensitivity of the human visual system at different spatial scales, but little is known as to how contrast sensitivity for achromatic and chromatic stimuli changes from a mesopic to a highly photopic range reflecting outdoor illumination levels. The purpose of our study was to further characterize the CSF by measuring both achromatic and chromatic sensitivities for background luminance levels from 0.02 cd/m2 to 7,000 cd/m2. Stimuli consisted of Gabor patches of different spatial frequencies and angular sizes, varying from 0.125 to 6 cpd, which were displayed on a custom high dynamic range (HDR) display with luminance levels up to 15,000 cd/m2. Contrast sensitivity was measured in three directions in color space, an achromatic direction, an isoluminant "red-green" direction, and an S-cone isolating "yellow-violet" direction, selected to isolate the luminance, L/M-cone opponent, and S-cone opponent pathways, respectively, of the early postreceptoral processing stages. Within each session, observers were fully adapted to the fixed background luminance (0.02, 2, 20, 200, 2,000, or 7,000 cd/m2). Our main finding is that the background luminance has a differential effect on achromatic contrast sensitivity compared to chromatic contrast sensitivity. The achromatic contrast sensitivity increases with higher background luminance up to 200 cd/m2 and then shows a sharp decline when background luminance is increased further. In contrast, the chromatic sensitivity curves do not show a significant sensitivity drop at higher luminance levels. We present a computational luminance-dependent model that predicts the CSF for achromatic and chromatic stimuli of arbitrary size.

From pairwise comparisons and rating to a unified quality scale.

The goal of psychometric scaling is the quantification of perceptual experiences, understanding the relationship between an external stimulus, the internal representation and the response. In this paper, we propose a probabilistic framework to fuse the outcome of different psychophysical experimental protocols, namely rating and pairwise comparisons experiments. Such a method can be used for merging existing datasets of subjective nature and for experiments in which both measurements are collected. We analyze and compare the outcomes of both types of experimental protocols in terms of time and accuracy in a set of simulations and experiments with benchmark and real-world image quality assessment datasets, showing the necessity of scaling and the advantages of each protocol and mixing. Although most of our examples focus on image quality assessment, our findings generalize to any other subjective quality-of-experience task.

Validation of artificial neural networks as a methodology for donor-recipient matching for liver transplantation.

In 2014, we reported a model for donor-recipient (D-R) matching in liver transplantation (LT) based on artificial neural networks (ANNs) from a Spanish multicenter study (Model for Allocation of Donor and Recipient in España [MADR-E]). The aim is to test the ANN-based methodology in a different European health care system in order to validate it. An ANN model was designed using a cohort of patients from King's College Hospital (KCH; n = 822). The ANN was trained and tested using KCH pairs for both 3- and 12-month survival models. End points were probability of graft survival (correct classification rate [CCR]) and nonsurvival (minimum sensitivity [MS]). The final model is a rule-based system for facilitating the decision about the most appropriate D-R matching. Models designed for KCH had excellent prediction capabilities for both 3 months (CCR-area under the curve [AUC] = 0.94; MS-AUC = 0.94) and 12 months (CCR-AUC = 0.78; MS-AUC = 0.82), almost 15% higher than the best obtained by other known scores such as Model for End-Stage Liver Disease and balance of risk. Moreover, these results improve the previously reported ones in the multicentric MADR-E database. In conclusion, the use of ANN for D-R matching in LT in other health care systems achieved excellent prediction capabilities supporting the validation of these tools. It should be considered as the most advanced, objective, and useful tool to date for the management of waiting lists. Liver Transplantation 24 192-203 2018 AASLD.

Dynamically weighted evolutionary ordinal neural network for solving an imbalanced liver transplantation problem.

OBJECTIVE: Create an efficient decision-support model to assist medical experts in the process of organ allocation in liver transplantation. The mathematical model proposed here uses different sources of information to predict the probability of organ survival at different thresholds for each donor-recipient pair considered. Currently, this decision is mainly based on the Model for End-stage Liver Disease, which depends only on the severity of the recipient and obviates donor-recipient compatibility. We therefore propose to use information concerning the donor, the recipient and the surgery, with the objective of allocating the organ correctly. METHODS AND MATERIALS: The database consists of information concerning transplants conducted in 7 different Spanish hospitals and the King's College Hospital (United Kingdom). The state of the patients is followed up for 12 months. We propose to treat the problem as an ordinal classification one, where we predict the organ survival at different thresholds: less than 15 days, between 15 and 90 days, between 90 and 365 days and more than 365 days. This discretization is intended to produce finer-grain survival information (compared with the common binary approach). However, it results in a highly imbalanced dataset in which more than 85% of cases belong to the last class. To solve this, we combine two approaches, a cost-sensitive evolutionary ordinal artificial neural network (ANN) (in which we propose to incorporate dynamic weights to make more emphasis on the worst classified classes) and an ordinal over-sampling technique (which adds virtual patterns to the minority classes and thus alleviates the imbalanced nature of the dataset). RESULTS: The results obtained by our proposal are promising and satisfactory, considering the overall accuracy, the ordering of the classes and the sensitivity of minority classes. In this sense, both the dynamic costs and the over-sampling technique improve the base results of the considered ANN-based method. Comparing our model with other state-of-the-art techniques in ordinal classification, competitive results can also be appreciated. The results achieved with this proposal improve the ones obtained by other state-of-the-art models: we were able to correctly predict more than 73% of the transplantation results, with a geometric mean of the sensitivities of 31.46%, which is much higher than the one obtained by other models. CONCLUSIONS: The combination of the proposed cost-sensitive evolutionary algorithm together with the application of an over-sampling technique improves the predictive capability of our model in a significant way (especially for minority classes), which can help the surgeons make more informed decisions about the most appropriate recipient for an specific donor organ, in order to maximize the probability of survival after the transplantation and therefore the fairness principle.

Oversampling the Minority Class in the Feature Space.

The imbalanced nature of some real-world data is one of the current challenges for machine learning researchers. One common approach oversamples the minority class through convex combination of its patterns. We explore the general idea of synthetic oversampling in the feature space induced by a kernel function (as opposed to input space). If the kernel function matches the underlying problem, the classes will be linearly separable and synthetically generated patterns will lie on the minority class region. Since the feature space is not directly accessible, we use the empirical feature space (EFS) (a Euclidean space isomorphic to the feature space) for oversampling purposes. The proposed method is framed in the context of support vector machines, where the imbalanced data sets can pose a serious hindrance. The idea is investigated in three scenarios: 1) oversampling in the full and reduced-rank EFSs; 2) a kernel learning technique maximizing the data class separation to study the influence of the feature space structure (implicitly defined by the kernel function); and 3) a unified framework for preferential oversampling that spans some of the previous approaches in the literature. We support our investigation with extensive experiments over 50 imbalanced data sets.

Projection-based ensemble learning for ordinal regression.

The classification of patterns into naturally ordered labels is referred to as ordinal regression. This paper proposes an ensemble methodology specifically adapted to this type of problem, which is based on computing different classification tasks through the formulation of different order hypotheses. Every single model is trained in order to distinguish between one given class (k) and all the remaining ones, while grouping them in those classes with a rank lower than k , and those with a rank higher than k. Therefore, it can be considered as a reformulation of the well-known one-versus-all scheme. The base algorithm for the ensemble could be any threshold (or even probabilistic) method, such as the ones selected in this paper: kernel discriminant analysis, support vector machines and logistic regression (LR) (all reformulated to deal with ordinal regression problems). The method is seen to be competitive when compared with other state-of-the-art methodologies (both ordinal and nominal), by using six measures and a total of 15 ordinal datasets. Furthermore, an additional set of experiments is used to study the potential scalability and interpretability of the proposed method when using LR as base methodology for the ensemble.