Field Validation Research Articles

Field validation as a means for continual monitoring of approved test kit’s fitness for purpose in the commercial market

Testing accuracy of a chemical contaminant requires use of a testing platform that conforms to validation criteria outlined in quality literature and standards. This study explores the application of commercial field data measured by qualified analysts using a United States Department of Agriculture – Federal Grain Inspection Service approved kit for measuring fumonisin in maize to augment method validation procedures. Analysts from seven grain testing facilities were qualified in official USDA sampling, sample preparation, and testing methodology using the Charm LF-FUMQ-WETS5. A duplicate sample was tested in the Office of the Texas State Chemist (OTSC) laboratory using UPLC-MS-MS. Data were subject to four statistical techniques using continuous and categorical methodology. This approach enabled researchers to explore if a single test or multiple comparisons were best suited to assess a field kit’s fitness for purpose across facility, toxin level, and year. The study concluded that a paired t-test and correlation analysis provided a quick and meaningful evaluation of kit performance. The correct placement of samples within the correct bin (violative versus non-violative) aligns well with market forces and regulatory compliance. The results of this study also provide a useful tool to assess all field kits’ performance at the beginning of the harvest season and subsequent years. The combination of statistical techniques presented in this research is an important tool in assessing mycotoxin field test kits fitness for purpose and represents a key step in a continuous improvement-quality systems process meant to protect the feed and food supply.

Convolutional Block Attention Module-Multimodal Feature-Fusion Action Recognition: Enabling Miner Unsafe Action Recognition.

The unsafe action of miners is one of the main causes of mine accidents. Research on underground miner unsafe action recognition based on computer vision enables relatively accurate real-time recognition of unsafe action among underground miners. A dataset called unsafe actions of underground miners (UAUM) was constructed and included ten categories of such actions. Underground images were enhanced using spatial- and frequency-domain enhancement algorithms. A combination of the YOLOX object detection algorithm and the Lite-HRNet human key-point detection algorithm was utilized to obtain skeleton modal data. The CBAM-PoseC3D model, a skeleton modal action-recognition model incorporating the CBAM attention module, was proposed and combined with the RGB modal feature-extraction model CBAM-SlowOnly. Ultimately, this formed the Convolutional Block Attention Module-Multimodal Feature-Fusion Action Recognition (CBAM-MFFAR) model for recognizing unsafe actions of underground miners. The improved CBAM-MFFAR model achieved a recognition accuracy of 95.8% on the NTU60 RGB+D public dataset under the X-Sub benchmark. Compared to the CBAM-PoseC3D, PoseC3D, 2S-AGCN, and ST-GCN models, the recognition accuracy was improved by 2%, 2.7%, 7.3%, and 14.3%, respectively. On the UAUM dataset, the CBAM-MFFAR model achieved a recognition accuracy of 94.6%, with improvements of 2.6%, 4%, 12%, and 17.3% compared to the CBAM-PoseC3D, PoseC3D, 2S-AGCN, and ST-GCN models, respectively. In field validation at mining sites, the CBAM-MFFAR model accurately recognized similar and multiple unsafe actions among underground miners.

Field validation of NDVI to identify crop phenological signatures

Purpose and MethodsCrop identification using remotely sensed imagery provides useful information to make management decisions about land use and crop health. This research used phonecams to acquire the Normalized Difference Vegetation Index (NDVI) of various crops for three crop seasons. NDVI time series from Sentinel (L121-L192) images was also acquired using Google Earth Engine (GEE) for the same period. The resolution of satellite data is low therefore gap filling and smoothening filters were applied to the time series data. The comparison of data from satellite images and phenocam provides useful insight into crop phenology. The results show that NDVI is generally underestimated when compared to phenocam data. The Savitzky-Golay (SG) and some other gap filling and smoothening methods are applied to NDVI time series based on satellite images. The smoothened NDVI curves are statistically compared with daily NDVI series based on phenocam images as a reference.ResultsThe SG method has performed better than other methods like moving average. Furthermore, polynomial order has been found to be the most sensitive parameter in applying SG filter in GEE. Sentinel (L121-L192) image was used to identify wheat during the year 2022–2023 in Sargodha district where experimental fields were located. The Random Forest Machine Leaning algorithm was used in GEE as a classifier.ConclusionThe classification accuracy has been found 97% using this algorithm which suggests its usefulness in applying to other areas with similar agro-climatic characteristics.

Pengembangan Instrumen Tes Literasi Numerasi Berbasis Etnomatematika untuk Siswa SMP Negeri 2 Jambi

Numeracy literacy skills are the most important part of the abilities that every student must have. The limited number of numeracy literacy instruments at SMP Negeri 2 Jambi is one of the factors causing less than optimal mathematics learning. Students are still lacking in understanding questions, formulating, identifying and interpreting mathematical problems properly and correctly. To train these abilities, an ethnomathematics-based numeracy literacy test instrument is needed. This research aims to describe the process of developing an ethnomathematics-based numeracy literacy test instrument and determine the validity index of the ethnomathematics-based numeracy literacy test instrument. This research is development research that adapts the Plomp model which consists of initial assessment stage, design stage, realization/construction stage, and test, evaluation and revision stage. Data collection techniques in this research are field note techniques, questionnaire techniques, validation techniques and test techniques with research instruments in the form of field note sheets, questionnaire sheets, validation sheets and test questions. Data analysis uses Aiken's formula and empirical validity uses product moment correlation. The results of this research show that at the design stage, there was a difference of opinion between the two validators regarding the grid, there was one indicator question that was not in accordance with basic competencies. These differences are then analyzed and revised according to input and suggestions from the validator. The researcher improved the question grid as well as the test questions, resulting in good question items, and the results of content validation by 4 expert validators calculated using the Aiken's formula obtained a validity value of 0.785. These results indicate that the ethnomathematics-based numeracy literacy test instrument is declared valid from three aspects, namely; materials, constructs and language.

Numerical Data Based Fertilizer Recommendation System for Farmer

In modern agriculture, optimizing fertilizer use is crucial for maximizing crop yields, maintaining soil health, and reducing environmental impact. The Numerical Data-Based Fertilizer Recommendation System (NDFRS) for farmers addresses these needs by leveraging data-driven approaches to provide precise fertilizer recommendations tailored to specific crop and soil conditions. This system integrates soil nutrient analysis, crop nutrient requirements, and environmental factors to generate customized fertilizer plans. The NDFRS employs advanced machine learning algorithms to analyze extensive datasets, including soil test results, historical crop yield data, weather patterns, and geographical information. By incorporating these variables, the system can predict the optimal type and amount of fertilizers required for different crops at various growth stages. This ensures that nutrients are supplied efficiently, enhancing crop productivity while minimizing excess fertilizer use and associated environmental risks. The user- friendly interface of the NDFRS allows farmers to input relevant data easily and receive actionable recommendations. Additionally, the system offers real-time updates and adaptive recommendations based on changing environmental conditions and crop growth stages. Field trials and validation studies demonstrate the system's efficacy in improving crop yields and promoting sustainable farming practices. Keywords: Tensor Flow, Deep Learning.

Wormhole Geometry Modelling on Carbonate Matrix Acidizing: A Literature Review

Matrix acidizing is a critical well-stimulation technique used to enhance the permeability of carbonate reservoirs by creating channels or "wormholes" through the dissolution of the rock matrix. The efficiency of this process is significantly influenced by the geometry of the wormholes formed. This review paper provides a comprehensive analysis of current research on wormhole geometry modeling in carbonate matrix acidizing, synthesizing findings from experimental studies, analytical models, and numerical simulations. Key factors affecting wormhole geometry, such as acid concentration, injection rate, and rock properties, are discussed. The review highlights the contributions of various modeling approaches in predicting wormhole formation and propagation, emphasizing the importance of accurately capturing the coupled effects of fluid flow, chemical reactions, and rock dissolution. The implications for optimizing acidizing treatments and enhancing hydrocarbon recovery are explored, alongside recommendations for future research. These include the need for field validation, advanced modeling techniques, real-time monitoring, interdisciplinary collaboration, and sustainability considerations. This synthesis aims to provide a foundation for improving the design and execution of matrix acidizing operations in carbonate reservoirs.

Towards automated field ballast condition evaluation: Field validation of the ballast scanning vehicle capabilities

Ballast degradation can lead to adverse effects such as inadequate drainage, track settlement and reduced lateral stability, which could compromise track safety, daily functionality, and long-term maintenance. Field inspection of ballast for monitoring degradation and functional performance is a challenging task. Current state-of-the-practice methods for evaluating ballast primarily depend on subjective visual inspection, labor-intensive sampling, laboratory sieve analyses or Ground Penetrating Radar (GPR) technology. These methods fall short in providing an in-depth assessment of ballast, specifically in determining the degradation level and aggregate size and shape characteristics at various depths. In this regard, this research developed an innovative ballast investigation platform, the Ballast Scanning Vehicle (BSV), to automate the processes of acquiring detailed ballast inspection data. The BSV utilizes a deep learning-based pipeline for image segmentation to evaluate task-specific metrics such as coarse aggregate gradation, Fouling Index (FI), and continuous track FI depth profiles. This paper provides a detailed overview of the BSV’s functions as well as the different modules of the deep learning-based pipeline. Validation of the BSV’s capabilities was conducted at the Transportation Technology Center (TTC) and is discussed in detail. Based on the field results, the BSV is capable of providing accurate and near real-time evaluation of in-service ballast conditions, serves as a robust means for inspecting long sections of track, and can be used to investigate persistent trouble-spots related to track performance.

Preliminary field validity of ActiGraph-based energy expenditure estimation in wheelchair users with spinal cord injury.

Cross-sectional validation study. To develop a raw acceleration signal-based random forest (RF) model for predicting total energy expenditure (TEE) in manual wheelchair users (MWUs) and evaluate the preliminary field validity of this new model, along with four existing models published in prior literature, using the Doubly Labeled Water (DLW) method. General community and research institution in Pittsburgh, USA. A total of 78 participants' data from two previous studies were used to develop the new RF model. A seven-day cross-sectional study was conducted to collect participants' free-living physical activity and TEE data, resting metabolic rate, demographics, and anthropometrics. Ten MWUs with spinal cord injury (SCI) completed the study, with seven participants having valid data for evaluating the preliminary field validity of the five models. The RF model achieved a mean absolute error (MAE) of 0.59 ± 0.60 kcal/min and a mean absolute percentage error (MAPE) of 23.6% ± 24.3% on the validation set. For preliminary field validation, the five assessed models yielded MAE from 136 kcal/day to 1141 kcal/day and MAPE from 6.1% to 50.2%. The model developed by Nightingale et al. in 2015 achieved the best performance (MAE: 136 ± 96 kcal/day, MAPE: 6.1% ± 4.7%), while the RF model achieved comparable performance (MAE: 167 ± 99 kcal/day, MAPE: 7.4% ± 5.1%). Two existing models and our newly developed RF model showed good preliminary field validity for assessing TEE in MWUs with SCI and the potential to detect lifestyle change in this population. Future large-scale field validation studies and model iteration are recommended.

A MaxEnt predictive model for palaeontological sites in the Siwalik Hills: A case study from the Pinjor Formation of the Upper Siwalik Hills near Chandigarh, northern India

This study presents application of the MaxEnt predictive model to identify potential palaeontological sites in the Siwalik Hills, focusing on the Pinjor Formation near Chandigarh, northern India. The Siwalik region holds rich vertebrate palaeontological records, yet lacks comprehensive site prediction models. This research employs MaxEnt (3.4.0) software, to predict areas suitable for fossil occurrences. Georeferenced presence data was derived from literature and field surveys, for training the model. Environmental predictors including Pinjor geological deposit boundary, Vegetation Index (NDVI), and Slope and Aspect were from a digital elevation model. Furthermore, model development involved parameter tuning, for developing a potentially useful model. Field validation of the model through pedestrian surveys identified new fossil localities, demonstrating the model's practical applicability. This research emphasizes the importance of MaxEnt in developing site predictive models, offering a valuable tool for optimizing palaeontological field surveys. While highlighting the model's success, the study also recognizes its limitations, especially concerning landscape and vegetation changes over time. Overall, this work establishes a foundation for further research in predictive modelling for palaeontological exploration in the Siwalik region and emphasizes the need for multidisciplinary efforts in salvage palaeontology to mitigate anthropogenic threats.

Hierarchical-modular framework for habitat mapping through systematic and informed integration of remote sensing data with contextual information

Accurate and up-to-date habitat maps at national or regional levels are critical for informing conservation and restoration actions. Land cover maps generated from Earth Observation (EO) data can, to varying degrees, indicate the extent of some habitat categories but these are often insufficient in terms of the level of detail required. This study introduces a hierarchical-modular framework based on the globally applicable Food and Agriculture Organization (FAO) Land Cover Classification System (LCCS) whereby the land cover classes can be systematically translated to habitat categories and further differentiated by referencing contextual information and expert knowledge. Application is showcased for Wales (United Kingdom) where 10 m spatial resolution land cover maps were first constructed nationally from time-series of Sentinel-1C-band radar and Sentinel-2 optical data (2018–2020). The contextual datasets derived from non-EO sources were utilised to associate dominant vegetation types to detailed habitat classes.The accuracy assessment done using field survey data highlights that habitat classes directly derived from EO, including cultivated/managed and forest types, demonstrated higher mapping accuracies (>70% User's and Producer's Accuracies) compared to those that were more heavily dependent on non-EO-derived contextual information. These included neutral grasslands and fen/marsh/swamp that demonstrated lower accuracies (<20%) but also other detailed wetland classes (user's and producer's accuracies ranging from 25 to 60%). These reduced accuracies were largely associated with discrepancies in the contextual datasets used for their differentiation and mapping, and those integrated within field validation datasets. Dependency on contextual datasets diminished as detailed habitat maps were generalized into broader categories. The study proposes that, for enhanced habitat mapping, efforts should focus not only on EO data but also on maximising the accuracy and minimizing inconsistencies in contextual datasets as well as taxonomical systems. The flexible structure of the FAO LCCS hierarchical framework is also highlighted, emphasizing its adaptability for future improvements in habitat mapping by incorporating contextual datasets and more advanced algorithms that can improve EO-derived land cover descriptions.

Post‐Fire Sediment Yield From a Central California Watershed: Field Measurements and Validation of the WEPP Model

AbstractIn a warming climate, an intensifying fire regime and higher likelihood of extreme rain are expected to increase watershed sediment yield in many regions. Understanding regional variability in landscape response to fire and post‐fire rainfall is essential for managing water resources and infrastructure. We measured sediment yield resulting from sequential wildfire and extreme rain and flooding in the upper Carmel River watershed (116 km2), on the central California coast, USA, using changes in sediment volume mapped in a reservoir. We determined that the sediment yield after fire and post‐fire flooding was 854–1,100 t/km2/yr, a factor of 3.5–4.6 greater than the long‐term yield from this watershed and more than an order of magnitude greater than during severe drought conditions. In this first large‐scale field validation test of the WEPPcloud/wepppy framework for the Water Erosion Prediction Project (WEPP) model on a burned landscape, WEPP predicted 81%–106% of the measured sediment yield. These findings will facilitate assessing and predicting future fire effects in steep watersheds with a Mediterranean climate and indicate that the increasingly widespread use of WEPP is appropriate for evaluating post‐fire hillslope erosion even across 100‐km2 scales under conditions without debris flows.

Proposed Methods Accelerate Permanent CO2-Storage Process

_ This article, written by JPT Technology Editor Chris Carpenter, contains highlights of paper SPE 215352,“ Accelerating the Permanent CO2 Storage Process: A Safer and Faster Route to Net Zero,” by Shubham Mishra, SPE, Boston Consulting Group. The paper has not been peer reviewed. _ The complete paper proposes two methods of accelerating the solidification (or mineralization) of CO2 in subsurface conditions, thus reducing the time required in the CO2 storage process. It also reviews industry and academic works devoted to the subject. Introduction Two main concerns with CO2 storage in subsurface reservoirs and CO2 sequestration are the large time cycle involved in its permanent storage and the associated environmental risks. Because other CO2 trapping mechanisms can be reversible, these concerns hinge heavily on the time required for solidification or mineralization of CO2 into a component such as calcite (CaCO3). Using current technologies and practices, it is practically impossible to complete the soaking period—or, in simpler words, a CO2 storage project—in one lifetime. Thus, technological development of CO2 storage, which depends on field validations, becomes an extremely long, multigenerational process. On the other hand, environmental safety is always a concern with underground CO2 storage. Three of four CO2 trapping mechanisms (structural, solubility-based, and residual) pose greater risk to the environment than does the fourth, which is mineralization or solidification. Therefore, from the point of view of environmental safety, mineralization is the most permanent form of CO2 storage that minimizes the risk of CO2 existing in gaseous form in the reservoir and thus flowing upward to shallower zones or the surface. Related Efforts Several research projects devoted to finding a commercial solution for accelerating the mineralization process are ongoing, including the following: - The CarbFix project in Iceland is based upon injection of CO2 dissolved in water streams into basalt rocks for accelerating mineral trapping. Through tracer surveys and mass calculations, the project has proposed that mineralization can be achieved in as little as 2 years for the size of the reservoir under consideration and the amount of CO2 injected. - A Canadian startup (Carbon Engineering) is turning carbon emissions into pellets that could be used as a synthetic fuel source, while a Swiss company called Climeworks is pumping extracted carbon to farms for agricultural use. - Another start-up is using calcium oxide, a waste product from the steel industry called steel slag, to react with CO2 and form CaCO3. This is used in controlled surface conditions in a cement-industry setup. This process, however, highlights how the reactions converting CO2 into solid can be sped up. - University research projects are ongoing that focus on use of various catalysts for increasing the rate of the mineralization chemical reaction, resulting in the formation of CaCO3. Proposed Methods While industry and academia are focusing on niche setups to solve this problem, not much focus is being placed on general large-scale applications such as increasing the mineralization chemical reaction rate in sandstone saline aquifers, where approximately 80% of potential carbon capture, utilization, and storage (CCUS) projects will be developed. Thus, technologies or workflows that can be applied in CO2 storage projects in saline aquifers to increase the mineralization reaction rate can be essential to bringing a shift to the larger vision of CCUS. In this effort, an integrated approach incorporating experience from oil and gas, cement, and other adjacent industries can help develop potential solutions.

Frequency Identification using Resonance Curve-Based Drive-by Monitoring: Field Validation

Railway bridges have a long service life - in Germany, for example, the average is 122 years. During this time, vehicles are constantly evolving, leading to innovative axle configurations and higher speeds. This leads to loads for which the bridges were not originally designed, which can lead to resonant structural responses. A key parameter for calculating and evaluating the dynamic behaviour of bridge structures, is the resonance frequency. Due to the large number of bridges in the network, conventional methods using sensors on the structure would be very laborious. A cost-effective alternative would be drive-by monitoring, using sensors on a passing train. In railway bridge contexts, drive-by monitoring to identify frequencies faces challenges from short spans and high speeds, leading to brief contact periods. The brief periods cause a low resolution of the signals in the frequency domain. An approach to overcome this issue is the use of resonance curve-based drive-by monitoring. The term resonance curve describes the maximum structural responses as a function of speed. In conjunction with the known axle configuration of the train, the resonance frequency can be determined from the resonance curve. In this paper, data from two field tests are analysed, in which synchronised acceleration measurements were conducted on the main girders of the bridges and the axle boxes of the trains. An ICE 4 was used on a bridge with a 19.5-metre span, and an ICE TD on a bridge spanning 16.4 metres. The analyses show that frequency identification using indirect resonance curves is feasible under real operating conditions. Thus, this approach enables the determination of a key parameter essential for calibrating structural models. Furthermore, this method enables the avoidance of resonance crossings by speed adaptation, either by acceleration or deceleration. This strategy has the potential to significantly increase the service life of bridges.

Investigate the optimum design of atomizing nozzles for coal dust suppression by using multifactor level response surface methodology

Mechanized coal extraction generates substantial dust, adversely affecting the physical and mental wellbeing of underground workers and degrading the operational environment. To address the limitations of traditional spray systems, which often suffer from poor atomization effects and limited ability to withstand wind interference, the innovative vortex atomizing nozzle has been developed specifically for dust suppression. This advanced solution was rigorously tested through detailed numerical simulations and comprehensive experimental investigations. These studies focused on examining the impact of critical parameters, including the velocity at the airflow director's exit, the angle of the exit hole, and its radius, on the efficacy of dust suppression. Findings highlighted the paramount importance of the airflow director's exit velocity in enhancing dust removal, with the exit hole's radius having the least impact. Optimal conditions for dust suppression were identified as an airflow director exit velocity of 20 m/s, an angle of 60°, and a radius of 0.6 mm. Field validations under these optimum parameters demonstrated unparalleled dust control efficiency, achieving a reduction in coal dust concentration exceeding 90 %. This apparatus offers a highly effective, user-friendly solution for comprehensive coal dust management across various mine locations.

Exploring ICT-based learning adoption in higher education: An extended perspective of the technology acceptance model

This article analyses the self-perception of digital technology learning competence among engineering and science students. The use of information and communication technologies (ICT) is now considered highly essential in higher education. Today's higher education institutions (HEIs) seldom employ traditional methods of instruction; instead, sophisticated ICT is starting to emerge as a feasible paradigm for fundamental change. To this end, Pakistan's Higher Education Commission (HEC) has heavily invested in technology in the education sector and initiated several programs. The ramifications of ICT utilization on the environment, student academic performance, and capabilities remain ambiguous. In delving into how students embrace and utilize ICT systems for learning in HEIs, researchers felt the need to grasp the factors that influence students' acceptance and utilization of ICT skills through the lens of the Extended Technology Acceptance Model (TAM). This study develops an exclusive conceptual model by incorporating TAM with a set of latent variables identified in the available literature: interest, ICT self-efficacy, economic cost, and performance expectancy. The adapted TAM model developed in this study considers the influence of these variables on students’ ICT acceptance and its impact on academic performance in emerging countries. As part of this study, reports on the development of instruments and validation in the research field were managed using a cross-sectional survey method and SPSS-22 and Smart PLS-4 software. In addition, association rule mining applied in demographic data. A sample of 69 students was randomly selected from three universities representing engineering, medical, and general HEIs in the Sindh province. Though the sample size was small, it showcases that the reliability of the scales is within an acceptable range and can be used to test the main study hypotheses. A conceptual framework model is introduced to offer a comprehensive framework derived from the amalgamation of various acceptance and usage models of technology. The results demonstrate that the survey items are appropriate and suitable for further research.

AIoT-enabled decentralized sensor fault diagnosis for structural health monitoring

Structural health monitoring (SHM) systems collect and analyze data recorded from civil infrastructure to detect changes in the structural condition and to predict potential damage. However, sensor faults in SHM systems may lead to erroneous structural assessments, thus to inefficient maintenance operations, which could jeopardize the integrity of the structures. Artificial intelligence (AI) algorithms based on analytical redundancy have proven being effective foundations to implement sensor fault diagnosis but are computationally expensive and require large amounts of raw sensor data to be transmitted to centralized servers. Moreover, operating with centralized servers increases the likelihood of single points of failure, which compromises the robustness of SHM systems. To overcome the limitations and to enhance the robustness of SHM systems, this paper proposes a decentralized approach towards fault-tolerant SHM systems based on the concept of Artificial Intelligence of Things (AIoT). Fault diagnosis algorithms based on AI are embedded into sensor nodes deployed for SHM, using Internet of Things technologies to be computed at the edge of civil infrastructure, avoiding large amounts of data to be sent to central servers. In addition, the sensor nodes embed algorithms that collect, process, and analyze the sensor data to assess the condition of the structures being monitored. The result is a smart (i.e., decentralized, fault-tolerant, and AIoT-based) SHM system that is calibrated in a laboratory experiment and then deployed on a bridge for field validation. It is expected that the approach presented in this paper will improve the robustness of SHM systems in assessing the structural condition of civil infrastructure.

Interferometric Synthetic Aperture Radar (InSAR)-Based Absence Sampling for Machine-Learning-Based Landslide Susceptibility Mapping: The Three Gorges Reservoir Area, China

The accurate prediction of landslide susceptibility relies on effectively handling landslide absence samples in machine learning (ML) models. However, existing research tends to generate these samples in feature space, posing challenges in field validation, or using physics-informed models, thereby limiting their applicability. The rapid progress of interferometric synthetic aperture radar (InSAR) technology may bridge this gap by offering satellite images with extensive area coverage and precise surface deformation measurements at millimeter scales. Here, we propose an InSAR-based sampling strategy to generate absence samples for landslide susceptibility mapping in the Badong–Zigui area near the Three Gorges Reservoir, China. We achieve this by employing a Small Baseline Subset (SBAS) InSAR to generate the annual average ground deformation. Subsequently, we select absence samples from slopes with very slow deformation. Logistic regression, support vector machine, and random forest models demonstrate improvement when using InSAR-based absence samples, indicating enhanced accuracy in reflecting non-landslide conditions. Furthermore, we compare different integration methods to integrate InSAR into ML models, including absence sampling, joint training, overlay weights, and their combination, finding that utilizing all three methods simultaneously optimally improves landslide susceptibility models.

Retrieval of snow liquid water content from radiative transfer model, field data and PRISMA satellite data

The amount of liquid water (LWC) present in the snowpack is critical for predicting wet snow avalanches, forecasting meltwater release, and assessing water availability in river basins. However, measuring this variable using traditional in situ methods is challenging. Space imaging spectroscopy is emerging as a promising approach to map the spatial and temporal variations of snow parameters. While some studies suggest the potential of hyperspectral remote sensing to infer liquid water content, field validation is still lacking. In this context, we propose a new spectral index, namely Snow Surficial Water Index (SSWI), which is designed to be sensitive to the percentage of surficial liquid water content in snow. We developed the index using the BioSNICAR radiative transfer model and then we tested it on both field spectral data and satellite PRISMA imagery. Validation was performed using field data collected with a Snow Sensor during four campaigns in alpine environments, one of which simultaneously with PRISMA. Through a k-fold cross-validation analysis, we achieved a coefficient of determination of 0.7 and a Root Mean Square Error equal to 3%, demonstrating the effectiveness of the proposed index in retrieving LWC from field data and mapping LWC from PRISMA data. A spatial analysis at the catchment level reinforced the results, showing an LWC distribution consistent with orography. The proposed method can be easily applied to other space imaging spectroscopy missions.

Screening of malaria infections in human blood samples with varying parasite densities and anaemic conditions using AI-Powered mid-infrared spectroscopy

BackgroundEffective testing for malaria, including the detection of infections at very low densities, is vital for the successful elimination of the disease. Unfortunately, existing methods are either inexpensive but poorly sensitive or sensitive but costly. Recent studies have shown that mid-infrared spectroscopy coupled with machine learning (MIRs-ML) has potential for rapidly detecting malaria infections but requires further evaluation on diverse samples representative of natural infections in endemic areas. The aim of this study was, therefore, to demonstrate a simple AI-powered, reagent-free, and user-friendly approach that uses mid-infrared spectra from dried blood spots to accurately detect malaria infections across varying parasite densities and anaemic conditions.MethodsPlasmodium falciparum strains NF54 and FCR3 were cultured and mixed with blood from 70 malaria-free individuals to create various malaria parasitaemia and anaemic conditions. Blood dilutions produced three haematocrit ratios (50%, 25%, 12.5%) and five parasitaemia levels (6%, 0.1%, 0.002%, 0.00003%, 0%). Dried blood spots were prepared on Whatman™ filter papers and scanned using attenuated total reflection-Fourier Transform Infrared (ATR-FTIR) for machine-learning analysis. Three classifiers were trained on an 80%/20% split of 4655 spectra: (I) high contrast (6% parasitaemia vs. negative), (II) low contrast (0.00003% vs. negative) and (III) all concentrations (all positive levels vs. negative). The classifiers were validated with unseen datasets to detect malaria at various parasitaemia levels and anaemic conditions. Additionally, these classifiers were tested on samples from a population survey in malaria-endemic villages of southeastern Tanzania.ResultsThe AI classifiers attained over 90% accuracy in detecting malaria infections as low as one parasite per microlitre of blood, a sensitivity unattainable by conventional RDTs and microscopy. These laboratory-developed classifiers seamlessly transitioned to field applicability, achieving over 80% accuracy in predicting natural P. falciparum infections in blood samples collected during the field survey. Crucially, the performance remained unaffected by various levels of anaemia, a common complication in malaria patients.ConclusionThese findings suggest that the AI-driven mid-infrared spectroscopy approach holds promise as a simplified, sensitive and cost-effective method for malaria screening, consistently performing well despite variations in parasite densities and anaemic conditions. The technique simply involves scanning dried blood spots with a desktop mid-infrared scanner and analysing the spectra using pre-trained AI classifiers, making it readily adaptable to field conditions in low-resource settings. In this study, the approach was successfully adapted to field use, effectively predicting natural malaria infections in blood samples from a population-level survey in Tanzania. With additional field trials and validation, this technique could significantly enhance malaria surveillance and contribute to accelerating malaria elimination efforts.

Predicating risk habitats of Oncomelania hupensis, the intermediate host of schistosoma japonicum under multiple environmental drivers

BackgroundOncomelania hupensis (O. hupensis), the unique intermediate host for Schistosoma japonicum, exerts a substantial influence on the risk of schistosomiasis. Being amphibious freshwater snails, the growth, development, and reproductive distribution of O. hupensis are intricately tied to climatic environmental variables. This study aims to predict O. hupensis habitat risks along the Yangtze River in China, considering multiple environmental factors. MethodsData pertaining to the distribution of O. hupensis, including both presence and absence records, with the Jiangsu section of the Yangtze River basin for the period 2017–2021, were retrieved from the Jiangsu Schistosomiasis Control Information Platform. Ten machine learning algorithms and an ensemble model were used to explore environmental drivers. Three datasets (Snail_CLIM, Snail_TOPO, and Snail_ALL) incorporating climatic and topographic variables were examined for their impact on model accuracy. We conducted validation using the AUC and TSS metrics. Moreover, we utilized the data from the 2022 snail field survey for model external validation. ResultsThe findings demonstrate that snail_ALL, which incorporates both climatic and topographic variables, exhibits superior performance (ensemble model: sensitivity = 98.000, specificity = 95.960, AUC = 0.994). Among the ten model algorithms, Random Forest (RF) exhibited the highest degree of accuracy and stability (Snail_ALL: AUC = 1.000 ± 0.000, TSS = 0.985 ± 0.005). The key environmental factors affecting snail distribution included the distance to the nearest river, elevation, annual precipitation, and annual average pressure. High-risk areas manifested as two distinct concentrations: downstream of the Luhe District in Nanjing and at the confluence of Zhenjiang and Yangzhou. The results of 2022 field validation showed that over 90 % of the data points for snail breeding sites are concentrated in medium to high-risk areas. ConclusionBy selecting pertinent environmental variables and employing ensemble modeling techniques, we can accurately predict O. hupensis habitats. The resulting risk distribution map for snail habitats not only provides valuable insights but also serves as a guiding tool for targeted monitoring and control measures. The holds particular significance within the contest of the Yangtze River protection and restoration projects.