Are unmanned aerial vehicle-based hyperspectral imaging and machine learning advancing crop science?

The past few years have seen increased interest in unmanned aerial vehicle (UAV)-based hyperspectral imaging (HSI) and machine learning (ML) in agricultural research, concomitant with an increase in published research on these topics. We provide an updated review, written for agriculturalists, highlighting the benefits in the retrieval of biophysical parameters of crops via UAVs relative to less sophisticated options. We reviewed >70 recent papers and found few consistent results between similar studies. Owing to their high complexity and cost, especially when applied to crops of low value, the benefits of most of the research reviewed are difficult to explain. Future effort will be necessary to distill research findings into lower-cost options for end-users.

Phenological stage and vegetation index for predicting corn yield under rainfed environments.

Uncrewed aerial systems (UASs) provide high temporal and spatial resolution information for crop health monitoring and informed management decisions to improve yields. However, traditional in-season yield prediction methodologies are often inconsistent and inaccurate due to variations in soil types and environmental factors. This study aimed to identify the best phenological stage and vegetation index (VI) for estimating corn yield under rainfed conditions. Multispectral images were collected over three years (2020-2022) during the corn growing season and over fifty VIs were analyzed. In the three-year period, thirty-one VIs exhibited significant correlations (r ≥ 0.7) with yield. Sixteen VIs were significantly correlated with the yield at least for two years, and five VIs had a significant correlation with the yield for all three years. A strong correlation with yield was achieved by combining red, red edge, and near infrared-based indices. Further, combined correlation and random forest an alyses between yield and VIs led to the identification of consistent and highest predictive power VIs for corn yield prediction. Among them, leaf chlorophyll index, Medium Resolution Imaging Spectrometer (MERIS) terrestrial chlorophyll index and modified normalized difference at 705 were the most consistent predictors of corn yield when recorded around the reproductive stage (R1). This study demonstrated the dynamic nature of canopy reflectance and the importance of considering growth stages, and environmental conditions for accurate corn yield prediction.

Seasonal changes of trace elements, nutrients, dissolved organic matter, and coastal acidification over the largest oyster reef in the Western Mississippi Sound, USA.

Seasonal changes of trace elements, nutrients, dissolved organic matter (DOM), and carbonate system parameters were evaluated over the largest deteriorating oyster reef in the Western Mississippi Sound using data collected during spring, summer, and winter of 2018, and summer of 2019. Higher concentrations of Pb (224%), Cu (211%), Zn (2400%), and Ca (240%) were observed during winter of 2018 compared to summer 2019. Phosphate and ammonia concentrations were higher (> 800%) during both summers of 2018 and 2019 than winter of 2018. Among the three distinct DOM components identified, two terrestrial humic-like components were more abundant during both spring (12% and 36%) and summer (11% and 33%) of 2018 than winter of 2018, implying a relatively lesser supply of humic-like components from terrestrial sources during winter. On the other hand, the protein-like component was more abundant during summer of 2019 compared to rest of the study period, suggesting a higher rate of autochthonous production during summer 2019. In addition, to their significant depth-wise variation, ocean acidification parameters including pH, pCO2, CO32-, and carbonate saturation states were all higher during both summers of 2018 and 2019. The measured variables such as trace elements, organic carbon, suspended particulates, and acidification parameters exhibited conservative mixing behavior against salinity. These observations have strong implications for the health of the oyster reefs, which provides ecologically important habitats and supports the economy of the Gulf Coast.

Class-Aware Fish Species Recognition Using Deep Learning for an Imbalanced Dataset.

Fish species recognition is crucial to identifying the abundance of fish species in a specific area, controlling production management, and monitoring the ecosystem, especially identifying the endangered species, which makes accurate fish species recognition essential. In this work, the fish species recognition problem is formulated as an object detection model to handle multiple fish in a single image, which is challenging to classify using a simple classification network. The proposed model consists of MobileNetv3-large and VGG16 backbone networks and an SSD detection head. Moreover, a class-aware loss function is proposed to solve the class imbalance problem of our dataset. The class-aware loss takes the number of instances in each species into account and gives more weight to those species with a smaller number of instances. This loss function can be applied to any classification or object detection task with an imbalanced dataset. The experimental result on the large-scale reef fish dataset, SEAMAPD21, shows that the class-aware loss improves the model over the original loss by up to 79.7%. The experimental result on the Pascal VOC dataset also shows the model outperforms the original SSD object detection model.

Survey and Insights into Unmanned Aerial-Vehicle-Based Detection and Documentation of Clandestine Graves and Human Remains.

Numerous biological and archaeological studies have demonstrated the legitimacy of remote sensing in anthropology. This article focuses on detecting and documenting terrestrial clandestine graves and surface remains (CGSR) of humans using unmanned aerial vehicles (UAVs), sensors, and automatic processing algorithms. CGSR is a problem of complex decision making under uncertainty that requires the identification and intelligent reasoning about direct evidence of human remains and their environmental fingerprints. As such, it is as much an engineering and geospatial problem as it is an anthropology problem. This article is an effort to survey existing work across disciplines and to provide insights and recommendations to assist future research. To support our claims, preliminary experiments were performed at the Forensic Anthropological Research Facility at Texas State University using UAVs, hyperspectral imaging, thermal imaging, and structure from motion. Prior work, our experience, and preliminary results indicate that both great potential and extreme challenges face remote sensing of CGSR.

Influence of land use and land cover on the spatial variability of dissolved organic matter in multiple aquatic environments.

Water quality of lakes, estuaries, and coastal areas serves as an indicator of the overall health of aquatic ecosystems as well as the health of the terrestrial ecosystem that drains to the water body. Land use and land cover plays not only a significant role in controlling the quantity of the exported dissolved organic matter (DOM) but also influences the quality of DOM via various biogeochemical and biodegradation processes. We examined the characteristics and spatial distribution of DOM in five major lakes, in an estuary, and in the coastal waters of the Mississippi, USA, and investigated the influence of the land use and land cover of their watersheds on the DOM composition. We employed absorption and fluorescence spectroscopy including excitation-emission matrix (EEM) combined with parallel factor (PARAFAC) analysis modeling techniques to determine optical properties of DOM and its characteristics in this study. We developed a site-specific PARAFAC model to evaluate DOM characteristics resulting in five diverse DOM compositions that included two terrestrial humic-like (C1 and C3), two microbial humic-like (C2 and C5), and one protein-like (C4) DOM. Our results showed elevated fluorescence levels of microbial humic-like or protein-like DOM in the lakes and coastal waters, while the estuarine waters showed relatively high fluorescence levels of terrestrial humic-like DOM. The results also showed that percent forest and wetland coverage explained 68 and 82% variability, respectively, in terrestrial humic-like DOM exports, while 87% variability in microbially derived humiclike DOM was explained by percent agricultural lands. Strong correlations between microbial humic-like DOM and fluorescence-derived DOM indices such as biological index (BIX) and fluorescence index (FI) indicated autochthonous characteristics in the lakes, while the estuary showed largely allochthonous DOM of terrestrial origin. We also observed higher concentrations of total dissolved phosphorous (TDP) and ammonium nitrogen (NH4-N) in coastal waters potentially due to photodegradation of refractory DOM derived from the sediment-bound organic matter in the coastal wetlands. This study highlights the relationships between the DOM compositions in the water and the land use and land cover in the watershed. The spatial variability of DOM in three different types of aquatic environments enhances the understanding of the role of land use and land cover in carbon cycling through export of organic matter to the aquatic ecosystems..

Multimorbidity in a marginalised, street-health Australian population: a retrospective cohort study.

OBJECTIVES: Demographic and presentation profile of patients using an innovative mobile outreach clinic compared with mainstream practice. DESIGN: Retrospective cohort study. SETTING: Primary care mobile street health clinic and mainstream practice in Western Australia. PARTICIPANTS: 2587 street health and 4583 mainstream patients. MAIN OUTCOME MEASURES: Prevalence and patterns of chronic diseases in anatomical domains across the entire age spectrum of patients and disease severity burden using Cumulative Illness Rating Scale (CIRS). RESULTS: Multimorbidity (2+ CIRS domains) prevalence was significantly higher in the street health cohort (46.3%, 1199/2587) than age-sex-adjusted mainstream estimate (43.1%, 2000/4583), p=0.011. Multimorbidity prevalence was significantly higher in street health patients <45 years (37.7%, 615/1649) compared with age-sex-adjusted mainstream patients (33%, 977/2961), p=0.003 but significantly lower if 65+ years (62%, 114/184 vs 90.7%, 322/355, p<0.001). Controlling for age and gender, the mean CIRS Severity Index score for street health (M=1.4, SD=0.91) was significantly higher than for mainstream patients (M=1.1, SD=0.80), p<0.001. Furthermore, 44.2% (530/1199) of street health patients had at least one level 3 or 4 score across domains compared with 18.3% (420/2294) for mainstream patients, p<0.001. Street health population comprised 29.6% (766/2587) Aboriginal patients with 50.4% (386/766) having multimorbidity compared with 44.6% (813/1821) for non-Aboriginals, p=0.007. There were no comprehensive data on Indigenous status in the mainstream cohort available for comparison. Musculoskeletal, respiratory and psychiatric domains were most commonly affected with multimorbidity significantly associated with male gender, increasing age and Indigenous status. CONCLUSIONS: Age-sex-adjusted multimorbidity prevalence and disease severity is higher in the street health cohort. Earlier onset (23-34 years) multimorbidity is found in the street health cohort but prevalence is lower in 65+ years than in mainstream patients. Multimorbidity prevalence is higher for Aboriginal patients of all ages.

Multimorbidity in patients attending 2 Australian primary care practices.

PURPOSE: Multiple chronic conditions in a single patient can be a challenging health burden. We aimed to examine patterns and prevalence of multimorbidity among patients attending 2 large Australian primary care practices and to estimate disease severity burden using the Cumulative Illness Rating Scale (CIRS). METHODS: Using published CIRS guidelines and a disease severity index calculated for each individual, we extracted data from the medical records of all 7,247 patients (58.5% female) seen over 6 months in 2008 who were rated for chronic conditions across 14 anatomical domains. RESULTS: Fifty-two percent of patients had multimorbidity in 2 or more CIRS domains, ranging from 20.6% if younger than 25 years, 43.7% if aged 25 to 44 years, 75.5% if aged 45 to 64 years, 87.5% if aged 65 to 74 years, and 97.1% if aged 75 years and older. Using a cutoff of 3 or more CIRS domains, 34.5% had multimorbidity ranging from 4.8% if younger than 25 years, 22.3% if aged 25 to 44 years, 56.1% if aged 45 to 64 years, 74.6% if aged 65 to 74 years, and 92.0% if aged 75 years and older. Musculoskeletal, singularly or in combination with others, was the commonest morbidity domain. The moderate severity index category increased with increasing age. CONCLUSIONS: Multimorbidity is a significant problem in men and women across all age-groups, and the moderate severity index increases with age. The musculoskeletal domain was most commonly affected. Mild and moderate severity index categories may underrepresent disease burden. Severity burden assessment in the primary care setting needs to take into account the severity index, as well as levels of domain severity within the index categories.

A 2D flow visualization user study using explicit flow synthesis and implicit task design.

This paper presents a 2D flow visualization user study that we conducted using new methodologies to increase the objectiveness. We evaluated grid-based variable-size arrows, evenly spaced streamlines, and line integral convolution (LIC) variants (basic, oriented, and enhanced versions) coupled with a colorwheel and/or rainbow color map, which are representative of many geometry-based and texture-based techniques. To reduce data-related bias, template-based explicit flow synthesis was used to create a wide variety of symmetric flows with similar topological complexity. To suppress task-related bias, pattern-based implicit task design was employed, addressing critical point recognition, critical point classification, and symmetric pattern categorization. In addition, variable-duration and fixed-duration measurement schemes were utilized for lightweight precision-critical and heavyweight judgment intensive flow analysis tasks, respectively, to record visualization effectiveness. We eliminated outliers and used the Ryan REGWQ post-hoc homogeneous subset tests in statistical analysis to obtain reliable findings. Our study shows that a texture-based dense representation with accentuated flow streaks, such as enhanced LIC, enables intuitive perception of the flow, while a geometry-based integral representation with uniform density control, such as evenly spaced streamlines, may exploit visual interpolation to facilitate mental reconstruction of the flow. It is also shown that inappropriate color mapping (e.g., colorwheel) may add distractions to a flow representation.

Noodles: a tool for visualization of numerical weather model ensemble uncertainty.

Numerical weather prediction ensembles are routinely used for operational weather forecasting. The members of these ensembles are individual simulations with either slightly perturbed initial conditions or different model parameterizations, or occasionally both. Multi-member ensemble output is usually large, multivariate, and challenging to interpret interactively. Forecast meteorologists are interested in understanding the uncertainties associated with numerical weather prediction; specifically variability between the ensemble members. Currently, visualization of ensemble members is mostly accomplished through spaghetti plots of a single mid-troposphere pressure surface height contour. In order to explore new uncertainty visualization methods, the Weather Research and Forecasting (WRF) model was used to create a 48-hour, 18 member parameterization ensemble of the 13 March 1993 "Superstorm". A tool was designed to interactively explore the ensemble uncertainty of three important weather variables: water-vapor mixing ratio, perturbation potential temperature, and perturbation pressure. Uncertainty was quantified using individual ensemble member standard deviation, inter-quartile range, and the width of the 95% confidence interval. Bootstrapping was employed to overcome the dependence on normality in the uncertainty metrics. A coordinated view of ribbon and glyph-based uncertainty visualization, spaghetti plots, iso-pressure colormaps, and data transect plots was provided to two meteorologists for expert evaluation. They found it useful in assessing uncertainty in the data, especially in finding outliers in the ensemble run and therefore avoiding the WRF parameterizations that lead to these outliers. Additionally, the meteorologists could identify spatial regions where the uncertainty was significantly high, allowing for identification of poorly simulated storm environments and physical interpretation of these model issues.

Opportunistic screening for chlamydia in young men.

BACKGROUND: Information on prevalence of genital Chlamydia trachomatis in young men attending Australian general practices is scarce. OBJECTIVE: We aimed to estimate the prevalence of chlamydia in this group; establish behaviours associated with infection, and evaluate general practitioner follow up of positive cases. METHODS: Sexually active men (aged 15-29 years) attending 10 general practices in Perth, Western Australia, were tested for chlamydia and completed a self report questionnaire concerning sexual practices and symptoms. RESULTS: Prevalence of chlamydia was 3.7% (95% CI: 2-6%, n=14/383). High rates of risky sexual practices were observed in both chlamydia positive and negative participants. The association between chlamydia status and risky sexual practices however, was not statistically significant. Treatment and notification of positive cases were usually undertaken, but GPs did not always check that patients had contacted sexual partners. CONCLUSION: We found little relationship between reported sexual behaviour and chlamydia infection in young men. It may be appropriate to offer screening to all at risk individuals.

A user study to compare four uncertainty visualization methods for 1D and 2D datasets.

Many techniques have been proposed to show uncertainty in data visualizations. However, very little is known about their effectiveness in conveying meaningful information. In this paper, we present a user study that evaluates the perception of uncertainty amongst four of the most commonly used techniques for visualizing uncertainty in one-dimensional and two-dimensional data. The techniques evaluated are traditional errorbars, scaled size of glyphs, color-mapping on glyphs, and color-mapping of uncertainty on the data surface. The study uses generated data that was designed to represent the systematic and random uncertainty components. Twenty-seven users performed two types of search tasks and two types of counting tasks on 1D and 2D datasets. The search tasks involved finding data points that were least or most uncertain. The counting tasks involved counting data features or uncertainty features. A 4x4 full-factorial ANOVA indicated a significant interaction between the techniques used and the type of tasks assigned for both datasets indicating that differences in performance between the four techniques depended on the type of task performed. Several one-way ANOVAs were computed to explore the simple main effects. Bonferronni's correction was used to control for the family-wise error rate for alpha-inflation. Although we did not find a consistent order among the four techniques for all the tasks, there are several findings from the study that we think are useful for uncertainty visualization design. We found a significant difference in user performance between searching for locations of high and searching for locations of low uncertainty. Errorbars consistently underperformed throughout the experiment. Scaling the size of glyphs and color-mapping of the surface performed reasonably well. The efficiency of most of these techniques were highly dependent on the tasks performed. We believe that these findings can be used in future uncertainty visualization design. In addition, the framework developed in this user study presents a structured approach to evaluate uncertainty visualization techniques, as well as provides a basis for future research in uncertainty visualization.

Retirement intentions of general practitioners aged 45-65 years.

OBJECTIVE: To ascertain the retirement intentions of a cohort of Australian general practitioners. DESIGN AND SETTING: Postal questionnaire survey of members of four Divisions of General Practice in Western Australia, sent out November 2007 - January 2008. PARTICIPANTS: A sample of 178 GPs aged 45-65 years. MAIN OUTCOME MEASURES: Intention to work in general practice until retirement; reasons for retiring before age 65 years; factors that might encourage working beyond chosen retirement age; and perceived obstacles to working in general practice. RESULTS: 63% of GPs intended to work to at least age 65 years, with men more likely to retire early. Of 63 GPs intending to retire early, 46% gave pressure of work, exhaustion and burnout as reasons for early retirement. Better remuneration, better staffing levels and more general support were incentives to continue working for 46% of the 64 GPs who responded to the question about incentives, and more flexible working hours, part-time work and reduced workload for 41%. Of 169 participants, 65% gave increasing bureaucracy, poor job satisfaction and disillusionment with the medical system or Medicare as obstacles to working in general practice in Australia, whereas workforce shortage, increasing patient demands and diminishing lifestyle through overwork were obstacles named by 48%. CONCLUSION: Many GPs are planning to retire early, reflecting an emerging trend among professionals and society generally. Declining job satisfaction, falling workforce numbers, excessive workload and increasing bureaucracy were recurrent concerns of older WA GPs considering premature retirement.

Breaking new ground: the story of Dagmar Berne.

In 1885 a young country girl enrolled in medicine at Sydney University. She was the first woman in Australia to enroll in a medical school. The medical school at this time was heavily influenced by academics who had graduated from the University of Edinburgh. At one point in her medical education she experienced difficulties with her examinations, the reasons for which are a point for debate. Under the mentorship of a famous British female medical practitioner who was visiting Australia at the time, she continued her studies in the United Kingdom. Here she was successful and returned to practice in New South Wales in 1895. Her experience of medical education was not without suffering. This young woman became part of an international group of women who at the time were trying to break into a field dominated by men.

How Western Australian registrars choose training practices.

BACKGROUND: While general practice registrars receive advice on how to choose a training practice, little is known about what underlies their decision. METHOD: A questionnaire was sent to all registrars in Western Australian General Practice Education and Training practices seeking information about the choice of their practice. RESULTS: Most registrars reported that they had chosen their practice. The supervisor was most commonly the person spoken to before the practice was chosen. DISCUSSION: The supervisor's personality was an important influence on choice and the practice manager was an important contact for registrars who were parents, presumably to arrange a part time attachment. Those who reported speaking to the supervisor also reported that they subsequently received greater remuneration. Registrars from a rural background chose more often to have all their training in a rural area.

An advanced evenly-spaced streamline placement algorithm.

This paper presents an advanced evenly-spaced streamline placement algorithm for fast, high-quality, and robust layout of flow lines. A fourth-order Runge-Kutta integrator with adaptive step size and error control is employed for rapid accurate streamline advection. Cubic Hermite polynomial interpolation with large sample-spacing is adopted to create fewer evenly-spaced samples along each streamline to reduce the amount of distance checking. We propose two methods to enhance placement quality. Double queues are used to prioritize topological seeding and to favor long streamlines to minimize discontinuities. Adaptive distance control based on the local flow variance is explored to reduce cavities. Furthermore, we propose a universal, effective, fast, and robust loop detection strategy to address closed and spiraling streamlines. Our algorithm is an order-of-magnitude faster than Jobard and Lefer's algorithm with better placement quality and over 5 times faster than Mebarki et al.'s algorithm with comparable placement quality, but with a more robust solution to loop detection.

Qualities of a good training practice--perceptions of advanced general practice registrars.

BACKGROUND: General practice registrars report satisfaction with their placements in Australian training practices. However, more needs to be known about which criteria comprise this global satisfaction. METHOD: A questionnaire was sent to 61 general practice registrars in Western Australia who had recently completed a 6 month advanced term in a training practice, seeking feedback on the practice. They were asked to mark the training practice on a scale of one (lowest) to 10 (highest) according to how well the training practice met a range of criteria. RESULTS: General practice registrars who rated their training practice highly believed the required amount of effective teaching and supervision had been provided to them during the placement, and adequate time made available for consultations, case reviews, discussion, and tuition. DISCUSSION: General practice registrars rating their training practice highly were more likely to have been allowed time to undertake outside education activities. Registrars not rating their training practices highly may be compromised by clinical service commitments. Registrars who believed members of the practice staff (medical and nonmedical) other than the general practice trainer had contributed to their learning were more likely to rate their training practice highly.