Acquisition Dates Research Articles

7253 Changes In Bone Density Observed After Transitioning From Denosumab To Alendronate Or Zoledronic Acid

Abstract Disclosure: A. Khan: None. B.P. Ramchandani: None. R.A. Zielinski: None. J. Liu: None. F.S. Mirza: None. Introduction: Discontinuation of denosumab (Dmab) results in rebound increase in bone turnover and increased risk of vertebral fracture. Guidelines recommend that discontinuation of denosumab should be followed by an oral or intravenous bisphosphonate (BP), hormone therapy or selective estrogen receptor modulator (SERM). Hypothesis: We hypothesize that patients who received weekly alendronate (ALEN) after discontinuation of Dmab will have less of a decline in bone mineral density (BMD) compared to patients who received once yearly zoledronic acid (ZA). Methods: We conducted a retrospective chart review to compare the change in BMD and factors affecting BMD loss in patients who received sequential therapy with ZA versus ALEN. IRB approval was obtained. Patients were selected using Epic SlicerDicer if they had taken Dmab therapy from 2018 (year of epic implementation) onwards and had been transitioned to a BP. A total of 251 patients were initially selected. 35 patients met the inclusion criteria of being transitioned off to ALEN or ZA at least 1 year prior to date of data acquisition; 21 transitioned to ZA and 14 to ALEN. The rest of the patients were continuing Dmab and excluded from the analysis. BMD gain after Dmab therapy was evaluated. BMD change after 1 year of receiving sequential therapy was compared between the two groups at the lumbar spine (LS), femoral neck (FN), and total hip (TH). Results: A total of 35 patients were included in data analysis with mean age of 73.7 + 9 years. The mean percentage gain in BMD achieved after Dmab therapy was 9.9 + 8.5%, 4.6 + 10.6%, and 6 + 5.2% at the LS, FN and TH respectively. The mean duration of Dmab therapy was 4.4 + 2.1 years. One year following sequential therapy, mean percentage BMD loss at the LS was 4.5% with ALEN and 3.0% with ZA (p-value: 0.59, 95% CI: -0.04 to 0.08). Following sequential therapy, mean percentage loss at FN was 2.2% with ALEN and 0.2% with ZA (p-value: 0.1895% CI: (-0.01 to 0.04), mean percentage loss at TH was 2.0 % with ALEN and 2.2% with ZA (p-value: 0.78, 95% CI: -0.03 to 0.02). Mean vitamin D level of all patients at time of starting sequential therapy was 45.9 + 12.9 ng/ml. Duration of prior bisphosphonate use before starting Dmab therapy did not have any significant effect on BMD loss after sequential therapy. All patients had vitamin D level in a good range at the beginning and end of the study period. Conclusion: There is invariably some decline in BMD when transitioning from Dmab to BPs. Although our data showed somewhat greater BMD loss at LS and FN with ALEN compared to ZA, this difference was not statistically significant and failed to reject the null hypothesis. These findings need to be evaluated further in larger prospective studies. Presentation: 6/3/2024

The Impacts of Phenological Stages within the Annual Cycle on Mapping Forest Stock Volume Using Multi-Band Dual-Polarization SAR Images in Boreal Forests

SAR images with two polarizations show strong potential for mapping forest stock volume (FSV) combined with limited samples. However, accurately mapping FSV still presents challenges in selecting the optimal acquisition date to obtain the SAR images during specific phenological stages within the annual forest cycle (growth and dormant stages). To clarify the impacts of phenological stages within the annual cycle on FSV mapping, SAR images with various polarization models and bands (Sentinel-1(S), GaoFen-3(GF-3 (G)) and ALOS-2(A)) were acquired within the growth and dormant stages of an annual cycle in a boreal evergreen coniferous forest (Chinese pine) and a deciduous coniferous forest (Larch). Subsequently, single-band (G, S, and A) and multi-band combined alternative variable sets (A + G, A + S, S + G, and A + S + G) were extracted within the same stage, respectively. Finally, the forward selection approach was utilized in conjunction with four different models (MLR, KNN, RF, and SVR) to obtain the most suitable variable sets and generate FSV mapping. The results demonstrated a strong correlation between the intensity of backscattering coefficients and the phenological stages of the forest. Within the dormant stage, there was a significant decrease in the gaps of backscattering coefficients obtained from the same polarization compared to those within the growth stage. Furthermore, the results also revealed that more signals from inside the canopy could be detected during the dormant stage in both evergreen coniferous forests and deciduous coniferous forests. Subsequently, the accuracy in mapping FSV obtained from single-band SAR images within the dormant stage are slightly higher than that within the growth stage, and the accuracy was still significantly affected by both overestimation and underestimation. Moreover, the combined effects of different bands significantly improve the reliability of mapped FSV. The rRMSE values in four multi-band combinations ranged from 22.37% to 29.40% for Chinese pine forests and from 21.27% to 34.38% for Larch forests, and the optimal result was observed from combinations of A + S + G acquired within the dormant stage. It is confirmed that SAR signal and their sensitivity to FSV depends on the stages of forest annual growth cycle. In comparison to the growth period, dual-polarization SAR data acquired during the dormant stage is more suitable for estimating FSV in boreal forests.

Investigating the contribution of image time series observations to cauliflower harvest-readiness prediction.

Cauliflower cultivation is subject to high-quality control criteria during sales, which underlines the importance of accurate harvest timing. Using time series data for plant phenotyping can provide insights into the dynamic development of cauliflower and allow more accurate predictions of when the crop is ready for harvest than single-time observations. However, data acquisition on a daily or weekly basis is resource-intensive, making selection of acquisition days highly important. We investigate which data acquisition days and development stages positively affect the model accuracy to get insights into prediction-relevant observation days and aid future data acquisition planning. We analyze harvest-readiness using the cauliflower image time series of the GrowliFlower dataset. We use an adjusted ResNet18 classification model, including positional encoding of the data acquisition dates to add implicit information about development. The explainable machine learning approach GroupSHAP analyzes time points' contributions. Time points with the lowest mean absolute contribution are excluded from the time series to determine their effect on model accuracy. Using image time series rather than single time points, we achieve an increase in accuracy of 4%. GroupSHAP allows the selection of time points that positively affect the model accuracy. By using seven selected time points instead of all 11 ones, the accuracy improves by an additional 4%, resulting in an overall accuracy of 89.3%. The selection of time points may therefore lead to a reduction in data collection in the future.

Enhancing Pléiades-based crop mapping with multi-temporal and texture information

Accurate crop mapping using satellite imagery is crucial for improving the monitoring of agricultural landscapes. Very high resolution (VHR) satellite imagery offers unique capabilities in this respect, allowing for even small fields to be discerned and image texture analysis to be performed. Additionally, satellite imagery has greater efficiency than unmanned aerial vehicles due to its extensive coverage. Moreover, the operation flexibility of VHR satellites means that timely image acquisition is possible several times during the growing season. This study investigates the potential of VHR Pléiades images and the random forest classifier for accurate crop mapping. Four images acquired on April 9th, May 12th, May 31st, and June 20th were used to test 16 classification scenarios, including single-date and multi-temporal combinations of spectral bands, texture features, and vegetation Indices (VIs). The classification using the spectral bands from all four images achieved the highest overall accuracy, 93.9% and 96.3% at field and pixel levels, respectively. The bitemporal classifications had lower accuracy. Nevertheless, the combination of the May 12th and June 20th spectral bands had 90% accuracy, which indicated that two images may be sufficient for reliable mapping if the periods with phenological differences between crops are considered. Adding texture features to the spectral bands significantly enhanced the accuracy (up to 8%) of single-date classifications, making it highly recommended when only one image is available. However, the impact of texture was more pronounced on the later dates. It showed the most marked benefit for vineyards and alfalfa, with minimal or no improvement observed for other classes like winter barley. An additional increase in overall accuracy was achieved in three of the four dates by supplementing the spectral and texture bands with VIs. This study highlights the importance of considering image acquisition dates and crop types when designing satellite-based crop mapping strategies for optimal accuracy.

HIV incidence in people receiving government-subsidised pre-exposure prophylaxis in Australia: a whole-of-population retrospective cohort study

HIV pre-exposure prophylaxis (PrEP) is highly effective and has been government subsidised in Australia since April, 2018. We examined HIV incidence over 5 years in a retrospective observational cohort of people who had received subsidised PrEP. Linked de-identified dispensing records for all government-subsidised oral PrEP, HIV antiretroviral therapy (ART), and hepatitis C treatment were used. We included all people dispensed subsidised PrEP from April 1, 2018, to March 31, 2023, and examined records up to Sept 30, 2023. Exposure was measured from date of first PrEP prescription and days covered by PrEP calculated for individuals based on quantity and date supplied. Assuming that HIV was diagnosed 30 days before ART initiation, we imputed the date of acquisition as the midpoint between the diagnosis and the later of the last PrEP prescription or 6 months before the diagnosis. We calculated HIV incidence and its predictors using Poisson regression. We included 66 206 people dispensed PrEP: 64 757 (97·8%) were men; median age was 33 years (IQR 27-43). 207 people acquired HIV, with an overall incidence of 1·07 per 1000 person-years (95% CI 0·93-1·23). Incidence was 2·61 per 1000 person-years among those dispensed PrEP once only. Using this group as a comparator, those with 60% or more days covered by PrEP had a 78·5% reduction in incidence (0·56 per 1000 person-years, p<0·0001) and those with less than 60% days covered had a 61·6% reduction (0·99 per 1000 person-years, p=0·0045). Independent predictors of HIV acquisition were a record of hepatitis C treatment (9·83 per 1000 person-years, adjusted incident rate ratio [aIRR] 8·70, 95% CI 4·86-15·56), only attending prescribers outside of areas with a high estimated prevalence of gay men (1·66 per 1000 person-years, aIRR 1·50, 1·08-2·09), age 18-29 years (1·33 per 1000 person-years, aIRR 1·56, 1·11-2·21), and earlier year of first PrEP. The low observed incidence of HIV among people receiving government-subsidised PrEP highlights the success of a national programme of oral PrEP scale-up in achieving sustained reduction in community HIV transmission. However, incidence varied greatly, indicating that more research is needed to understand why people were not taking PrEP at times of risk and emphasising the need for new interventions focused on this population to achieve elimination of HIV transmission. Individuals dispensed PrEP once only and less frequent users might benefit from more support. None.

Multitemporal sentinel-1 SAR data sensitivity analysis at phenological growth stages of green gram

ABSTRACT This research investigated the green gram crop by employing a time series of Sentinel-1A Single Look Complex (SLC) and Sentinel-2 datasets, which corresponded to different phenological growth stages of the crop. Field campaigns were conducted on the acquisition dates of Sentinel-1A datasets for the entire crop cycle to estimate crop height and soil moisture. In line with these observations, temporal variations in 10 salient Synthetic Aperture Radar (SAR) parameters, obtained from Sentinel-1A datasets, are systematically analysed, which elucidated the importance of dual-polarized SAR parameters in the diverse phenological growth stages of the crop. Moreover, the Normalized Difference Vegetation Index (NDVI) values from Sentinel-2 datasets, estimated at various green gram growth stages, are categorized into initial (NDVI ≤ 0.5), intermediate (0.5 < NDVI < 0.8), and maximum growth of the crop (NDVI ≥ 0.8), which are then aligned with corresponding SAR and ground-measured parameters. The inversion model development and sensitivity analysis were carried out to correlate ground-measured parameters with SAR parameters across various growth stages based on the categorization of NDVI values as well as for the entire growing season of the crop. Backscattering intensities are found to have good sensitivity to crop height at its intermediate growth (r ≥ 0.75). Among vegetation indices, the Polarimetric Radar Vegetation Index (PRVI) showed significant sensitivity to crop height in its intermediate growth stage (r ≈ 0.77). H-A-Alpha decomposition parameters displayed notable sensitivity to soil moisture in the initial and maximum growth stages of green gram. This study demonstrates the effectiveness of Sentinel-1 dual-polarized SAR parameters in monitoring green gram phenological growth stages. The parameters show substantial sensitivity to both green gram height and soil moisture.

Monitoring Cover Crop Biomass in Southern Brazil Using Combined PlanetScope and Sentinel-1 SAR Data

Precision agriculture integrates multiple sensors and data types to support farmers with informed decision-making tools throughout crop cycles. This study evaluated Aboveground Biomass (AGB) estimates of Rye using attributes derived from PlanetScope (PS) optical, Sentinel-1 Synthetic Aperture Radar (SAR), and hybrid (optical plus SAR) datasets. Optical attributes encompassed surface reflectance from PS’s blue, green, red, and near-infrared (NIR) bands, alongside the Normalized Difference Vegetation Index (NDVI) and Enhanced Vegetation Index (EVI). Sentinel-1 SAR attributes included the C-band Synthetic Aperture Radar Ground Range Detected, VV and HH polarizations, and both Ratio and Polarization (Pol) indices. Ground reference AGB data for Rye (Secale cereal L.) were collected from 50 samples and four dates at a farm located in southern Brazil, aligning with image acquisition dates. Multiple linear regression models were trained and validated. AGB was estimated based on individual (optical PS or Sentinel-1 SAR) and combined datasets (optical plus SAR). This process was repeated 100 times, and variable importance was extracted. Results revealed improved Rye AGB estimates with integrated optical and SAR data. Optical vegetation indices displayed higher correlation coefficients (r) for AGB estimation (r = +0.67 for both EVI and NDVI) compared to SAR attributes like VV, Ratio, and polarization (r ranging from −0.52 to −0.58). However, the hybrid regression model enhanced AGB estimation (R2 = 0.62, p &lt; 0.01), reducing RMSE to 579 kg·ha−1. Using only optical or SAR data yielded R2 values of 0.51 and 0.42, respectively (p &lt; 0.01). In the hybrid model, the most important predictors were VV, NIR, blue, and EVI. Spatial distribution analysis of predicted Rye AGB unveiled agricultural zones associated with varying biomass throughout the cover crop development. Our findings underscored the complementarity of optical with SAR data to enhance AGB estimates of cover crops, offering valuable insights for agricultural zoning to support soil and cash crop management.

Individual tree detection and crown delineation in the Harz National Park from 2009 to 2022 using mask R–CNN and aerial imagery

Forest diebacks pose a major threat to global ecosystems. Identifying and mapping both living and dead trees is crucial for understanding the causes and implementing effective management strategies. This study explores the efficacy of Mask R–CNN for automated forest dieback monitoring. The method detects individual trees, delineates their crowns, and classifies them as alive or dead. We evaluated the approach using aerial imagery and canopy height models in the Harz Mountains, Germany, a region severely affected by forest dieback. To assess the model's ability to track changes over time, we applied it to images from three separate flight campaigns (2009, 2016, and 2022). This evaluation considered variations in acquisition dates, cameras, post-processing techniques, and image tilting. Forest changes were analyzed based on the detected trees' number, spatial distribution, and height. A comprehensive accuracy assessment demonstrated the Mask R–CNN's robust performance, with precision scores ranging from 0.80 to 0.88 and F1-scores from 0.88 to 0.91. These results confirm the model's ability to generalize across diverse image acquisition conditions. While minor changes were observed between 2009 and 2016, the period between 2016 and 2022 witnessed substantial dieback, with a 64.57% loss of living trees. Notably, taller trees appeared to be particularly affected. This study highlights Mask R–CNN's potential as a valuable tool for automated forest dieback monitoring. It enables efficient detection, delineation, and classification of both living and dead trees, providing crucial data for informed forest management practices.

A temporal-spatial deep learning network for winter wheat mapping using time-series Sentinel-2 imagery

Accurate mapping of winter wheat provides essential information for food security and ecosystem protection. Deep learning approaches have achieved promising crop discrimination performance based on multitemporal satellite imagery. However, due to the high dimensionality of the data, sequential relations, and complex semantic information in time-series imagery, effective methods that can automatically capture temporal-spatial features with high separability and generalizability have received less attention. In this study, we proposed a U-shaped CNN-Transformer hybrid framework based on an attention mechanism, named the U-Temporal-Spatial-Transformer network (UTS-Former), for winter wheat mapping using Sentinel-2 imagery. This model includes an “encoder-decoder” structure for multiscale information mining of time series images and a temporal-spatial transformer module (TST) for learning comprehensive temporal sequence features and spatial semantic information. The results obtained from two study areas indicated that our UTS-Former achieved the best accuracy, with a mean MCC of 0.928 and an F1-score of 0.950, and the results of different band combinations also showed better performance than other popular time-series methods. We found that the MCC (MCC/All) of the UTS-Former using only RGB bands decreased by 4.53 %, while it decreased by 13.36 % and 35.02 % for UNet2d-LSTM and CNN-BiLSTM, respectively, compared with that of all the band combinations. The comparison demonstrated that the proposed UTS-Former could capture more global temporal-spatial information from winter wheat fields and achieve greater precision in terms of local details than other methods, resulting in high-quality mapping. The analysis of attention scores for the available acquisition dates revealed significant contributions of both beginning and ending growth images in winter wheat mapping, which is valuable for making appropriate selections of image dates. These findings suggest that the proposed approach has great potential for accurate, cost-effective, and high-quality winter wheat mapping.

Thinning and dynamics of the glacier terminus at the Pasterze Glacier (Austria), 2016–2021

As a result of global warming, glaciers worldwide recede and lose mass. Glacier ablation is influenced by surface characteristics, in particular the existence of debris cover. The evolution of glaciers may also be influenced by the existence of proglacial lakes, which are worldwide increasing in number and size. We examine the recent changes of the terminus of Pasterze Glacier by discussing the thinning and dynamics of the glacier terminus along with the changes of the glacial-proglacial transition zone between September 2016 and September 2021. By considering land cover in the interpretation, we find that glacier surface characteristics show different rates of elevation changes and surface movement. We mainly use orthophotos and digital elevation models (DEMs) resulting from airborne surveys of 12 different acquisition dates, but also draw on ablation stake measurements, lake water level data, water temperature data, as well as thermal infrared (TIR) imagery. While debris-covered ice lowered typically −5.0 m a−1 ± 4.6 m a−1 (mean ± standard deviation), debris-free ice was affected by ablation in the range of −7.0 m a−1 ± 2.7 m a−1, thus demonstrating that melt rates in similar elevations are related to surface characteristics. The elevation change rates were, in most cases, in a similar range throughout the observed periods of time. The horizontal displacement close to glacier-marginal crevasses amounted to up to ∼8 m a−1, but was mostly in the range of ∼1–4 m a−1. Considering both horizontal displacement and elevation changes, we find that clean and dirty ice surfaces show larger thickness losses and are moving faster compared to debris-covered glacier surfaces. Diurnal and seasonal fluctuations of lake water levels in the range of decimeters and lake water temperatures mostly clearly exceeding 0 °C are indicative of the potential influence of the proglacial lake on the glacier terminus. Given the decelerating and decaying behavior of the Pasterze Glacier in recent decades, we interpret the change rates of surface lowering and horizontal displacement 2016–2021 to be influenced by other factors such as the recently formed proglacial lake. This is also supported by direct ablation measurements that mostly show lesser ice melt differences compared to DEM-based surface elevation changes, suggesting that other factors such as thermally-induced melting contributed to an increased elevation difference.

Assessing the Potential of Multi-Temporal Conditional Generative Adversarial Networks in SAR-to-Optical Image Translation for Early-Stage Crop Monitoring

The incomplete construction of optical image time series caused by cloud contamination is one of the major limitations facing the application of optical satellite images in crop monitoring. Thus, the construction of a complete optical image time series via image reconstruction of cloud-contaminated regions is essential for thematic mapping in croplands. This study investigates the potential of multi-temporal conditional generative adversarial networks (MTcGANs) that use a single synthetic aperture radar (SAR) image acquired on a prediction date and a pair of SAR and optical images acquired on a reference date in the context of early-stage crop monitoring. MTcGAN has an advantage over conventional SAR-to-optical image translation methods as it allows input data of various compositions. As the prediction performance of MTcGAN depends on the input data composition, the variations in the prediction performance should be assessed for different input data combination cases. Such an assessment was performed through experiments using Sentinel-1 and -2 images acquired in the US Corn Belt. MTcGAN outperformed existing SAR-to-optical image translation methods, including Pix2Pix and supervised CycleGAN (S-CycleGAN), in cases representing various input compositions. In particular, MTcGAN was substantially superior when there was little change in crop vitality between the reference and prediction dates. For the SWIR1 band, the root mean square error of MTcGAN (0.021) for corn was significantly improved by 54.4% and 50.0% compared to Pix2Pix (0.046) and S-CycleGAN (0.042), respectively. Even when there were large changes in crop vitality, the prediction accuracy of MTcGAN was more than twice that of Pix2Pix and S-CycleGAN. Without considering the temporal intervals between input image acquisition dates, MTcGAN was found to be beneficial when crops were visually distinct in both SAR and optical images. These experimental results demonstrate the potential of MTcGAN in SAR-to-optical image translation for crop monitoring during the early growth stage and can serve as a guideline for selecting appropriate input images for MTcGAN.

Building Detection from SkySat Images with Transfer Learning: a Case Study over Ankara

AbstractThe detection and continuous updating of buildings in geodatabases has long been a major research area in geographic information science and is an important theme for national mapping agencies. Advancements in machine learning techniques, particularly state-of-the-art deep learning (DL) models, offer promising solutions for extracting and modeling building rooftops from images. However, tasks such as automatic labelling of learning data and the generalizability of models remain challenging. In this study, we assessed the sensor and geographic area adaptation capabilities of a pretrained DL model implemented in the ArcGIS environment using very-high-resolution (50 cm) SkySat imagery. The model was trained for digitizing building footprints via Mask R‑CNN with a ResNet50 backbone using aerial and satellite images from parts of the USA. Here, we utilized images from three different SkySat satellites with various acquisition dates and off-nadir angles and refined the pretrained model using small numbers of buildings as training data (5–53 buildings) over Ankara. We evaluated the buildings in areas with different characteristics, such as urban transformation, slums, regular, and obtained high accuracies with F‑1 scores of 0.92, 0.94, and 0.96 from SkySat 4, 7, and 17, respectively. The study findings showed that the DL model has high transfer learning capability for Ankara using only a few buildings and that the recent SkySat satellites demonstrate superior image quality.

High-resolution satellite imagery to assess orchard characteristics impacting water use

Most Orchards throughout the Mediterranean basin rely heavily on irrigation, a dependency increasing due to climate changes. Assessing the water requirement (WR) is crucial and depends on different factors, including orchard age, tree density per field, inter-row management. This study proposes new methods to evaluate these characteristics with remote sensing (RS). Various remote sensors providing high and very high spatial resolution images are investigated and their accuracy is assessed. The final objective is to assess WR using variables derived from remote sensing compared to data provided by water managers and from the FAO method. A typical Mediterranean watershed was selected in South-Eastern France, with orchards having various agricultural practices. Original methods were developed with Sentinel 2 (S2) data (2016–2023), 1 Pleiades image (2022) and the extraction of Google-satellite-hybrid images (GSH, 2017), and assessed using a large ground observation dataset (information on water use collected on 366 fields). Five orchards were monitored by capacitive sensors to assess the water balance. Irrigation durations ranged from 3–300 hours/year, with decision influenced by tree density and plot age. To identify young orchards, a thresholding approach on S2 derived NDVI effectively identified young orchards achieving a 98% accuracy rate. Grassed and non-grassed orchards were mapped using two methods, with a random forest classification using three spectral bands with 72% accuracy and a supervised approach yielding 81% accuracy for GSH and 57% for Pleiades. The performance depends on the acquisition date of images. A pattern detection algorithm applied to GSH and Pleaides determined tree density, showing a high correlation (r²=0.9) with observed data. These RS derived variables allowed to compute orchard water requirements at the watershed scale, ranging from 70 to 550 mm annually depending on management practices. The proposed methods can be extrapolated to other territories and are implemented using open access softwares.

High-Resolution Flood Monitoring Based on Advanced Statistical Modeling of Sentinel-1 Multi-Temporal Stacks

High-resolution flood monitoring can be achieved relying on multi-temporal analysis of remote sensing SAR data, through the implementation of semi-automated systems. Exploiting a Bayesian inference framework, conditioned probabilities can be estimated for the presence of floodwater at each image location and each acquisition date. We developed a procedure for efficient monitoring of floodwaters from SAR data cubes, which adopts a statistical modelling framework for SAR backscatter time series over normally unflooded areas based on Gaussian processes (GPs), in order to highlight flood events as outliers, causing abrupt variations in the trends. We found that non-parametric time series modelling improves the performances of Bayesian probabilistic inference with respect to state-of-the-art methodologies using, e.g., parametric fits based on periodic functions, by both reducing false detections and increasing true positives. Our approach also exploits ancillary data derived from a digital elevation model, including slopes, normalized heights above nearest drainage (HAND), and SAR imaging parameters such as shadow and layover conditions. It is here tested over an area that includes the so-called Metaponto Coastal Plain (MCP), in the Basilicata region (southern Italy), which is recurrently subject to floods. We illustrate the ability of our system to detect known (although not ground-truthed) and smaller, undocumented inundation events over large areas, and propose some consideration about its prospective use for contexts affected by similar events, over various land cover scenarios and climatic settings.

Spotting Culex pipiens from satellite: modeling habitat suitability in central Italy using Sentinel-2 and deep learning techniques.

Culex pipiens, an important vector of many vector borne diseases, is a species capable to feeding on a wide variety of hosts and adapting to different environments. To predict the potential distribution of Cx. pipiens in central Italy, this study integrated presence/absence data from a four-year entomological survey (2019-2022) carried out in the Abruzzo and Molise regions, with a datacube of spectral bands acquired by Sentinel-2 satellites, as patches of 224 × 224 pixels of 20 meters spatial resolution around each site and for each satellite revisit time. We investigated three scenarios: the baseline model, which considers the environmental conditions at the time of collection; the multitemporal model, focusing on conditions in the 2 months preceding the collection; and the MultiAdjacency Graph Attention Network (MAGAT) model, which accounts for similarities in temperature and nearby sites using a graph architecture. For the baseline scenario, a deep convolutional neural network (DCNN) analyzed a single multi-band Sentinel-2 image. The DCNN in the multitemporal model extracted temporal patterns from a sequence of 10 multispectral images; the MAGAT model incorporated spatial and climatic relationships among sites through a graph neural network aggregation method. For all models, we also evaluated temporal lags between the multi-band Earth Observation datacube date of acquisition and the mosquito collection, from 0 to 50 days. The study encompassed a total of 2,555 entomological collections, and 108,064 images (patches) at 20 meters spatial resolution. The baseline model achieved an F1 score higher than 75.8% for any temporal lag, which increased up to 81.4% with the multitemporal model. The MAGAT model recorded the highest F1 score of 80.9%. The study confirms the widespread presence of Cx. pipiens throughout the majority of the surveyed area. Utilizing only Sentinel-2 spectral bands, the models effectively capture early in advance the temporal patterns of the mosquito population, offering valuable insights for directing surveillance activities during the vector season. The methodology developed in this study can be scaled up to the national territory and extended to other vectors, in order to support the Ministry of Health in the surveillance and control strategies for the vectors and the diseases they transmit.

Grave from Nezabylice, Chomutov District. On the Phenomenon of Inhumation in Stage B1 of the Early Roman Period in Bohemia

The article describes the discovery of a skeletal grave in 2015 in Nezabylice (NW Bohemia, Chomutov district) at a cremation burial ground, which was dated to stage B1 of the early Roman period according to characteristic metal objects. Today, only four dozen similarly dated skeletal graves are registered in Bohemia, which, together with the early date of acquisition, makes them one of the rarest and most difficult to recognize archaeological monuments in this area. In the given situation, every recently researched skeleton grave from the early Roman period brings a wealth of new and important information about this distinctive phenomenon of burial rite among the Elbe-Germanic tribes.

Assessment of a deep learning model for COVID-19 classification on chest radiographs: a comparison across image acquisition techniques and clinical factors.

The purpose is to assess the performance of a pre-trained deep learning model in the task of classifying between coronavirus disease (COVID)-positive and COVID-negative patients from chest radiographs (CXRs) while considering various image acquisition parameters, clinical factors, and patient demographics. Standard and soft-tissue CXRs of 9860 patients comprised the "original dataset," consisting of training and test sets and were used to train a DenseNet-121 architecture model to classify COVID-19 using three classification algorithms: standard, soft tissue, and a combination of both types of images via feature fusion. A larger more-current test set of 5893 patients (the "current test set") was used to assess the performance of the pretrained model. The current test set contained a larger span of dates, incorporated different variants of the virus and included different immunization statuses. Model performance between the original and current test sets was evaluated using area under the receiver operating characteristic curve (ROC AUC) [95% CI]. The model achieved AUC values of 0.67 [0.65, 0.70] for cropped standard images, 0.65 [0.63, 0.67] for cropped soft-tissue images, and 0.67 [0.65, 0.69] for both types of cropped images. These were all significantly lower than the performance of the model on the original test set. Investigations regarding matching the acquisition dates between the test sets (i.e., controlling for virus variants), immunization status, disease severity, and age and sex distributions did not fully explain the discrepancy in performance. Several relevant factors were considered to determine whether differences existed in the test sets, including time period of image acquisition, vaccination status, and disease severity. The lower performance on the current test set may have occurred due to model overfitting and a lack of generalizability.

Six Decades of Changes in Pool Characteristics on a Concentric-Patterned Raised Bog

AbstractRaised bogs are wetland ecosystems which, under the right climatic conditions, feature patterns of pool hollows and hummock ridges. The relative cover and the spatial arrangement of pool and ridge microforms are thought to be influential on peatland atmosphere carbon gas fluxes and plant biodiversity. The mechanisms responsible for the formation and maintenance of pools, and the stability of these features in response to warming climates, remain topics of ongoing research. We employed historical aerial imagery, combined with a contemporary uncrewed aerial vehicle survey, to study 61 years of changes in pools at a patterned raised bog in central Sweden. We used a pool inheritance method to track individual pools between image acquisition dates throughout the time series. These data show a rapid loss of open-water pool area during the study period, primarily due to overgrowth of open-water pools by Sphagnum. We postulate that these changes are driven by ongoing climate warming that is accelerating Sphagnum colonisation. Open-water pool area declined by 26.8% during the study period, equivalent to a loss of 1001 m2 y−1 across the 150-hectare site. This is contradictory to an existing theory that states pools are highly stable, once formed, and can only convert to a terrestrial state through catastrophic drainage. The pool inheritance analysis shows that smaller pools are liable to become completely terrestrialised and expire. Our findings form part of a growing body of evidence for the loss of open-water habitats in peatlands across the boreal and elsewhere.

Estimating population density using open-access satellite images and geographic information system: Case of Al Ain city, UAE

Efficient and real-time estimates of population density are crucial for monitoring urban systems and disaster response. To develop this capability, this study estimates population density using covariates derived from remote-sensing images [e.g., building density, vegetation temperature light population index (VTLPI), and day–night band (DNB)]. Data were obtained from open-access satellite images, specifically Sentinel 2A, Landsat 8, and NPP-VIIRS images. Building density, VTLPI, and DNB were extracted at the district level and compared with population density. Furthermore, vector data from geographic information systems, such as human points of interest and road networks, were used to develop a geographically weighted regression model for estimating population density. Building density and DNB correlated positively with population density (r = 0.89), and population density correlated positively with the VTLPI (r = 0.75). Although all covariates correlated positively with estimates of population density, building density and the DNB showed the best correlation. Furthermore, both parameters were considerably related to human points of interest and road network density. Finally, using geographically weighted regression (GWR) and multiple linear regression (MLR) with the building density, DNB, and human points of interest as input, we predicted the population density. The GWR model built from independent variables correlates well with the population density (AICc = 526; R2 adj. = 0.89). Overall, the GWR and MLR produce similar results. To assess the transferability of the GWR model used for Al Ain, it was tested on Abu Dhabi, and the results produce an accuracy of 36%. This low accuracy may be attributed to the acquisition dates of the census data being different from those of the human points of interest. Further research is required to address such limitations in the proposed method and to provide additional support for using open-access remote-sensing images to estimate population density.

Prediction of Time-to-First Treatment (TTFT) in Favorable-Risk Chronic Lymphocytic Leukemia (CLL) Patients Using a Novel Stratification through Transcriptome Risk Scoring

Background: CLL is the most prevalent adult leukemia with a highly heterogenous genetic landscape leading to extremely varied clinical outcomes. Previously, CLL patients with immunoglobulin heavy chain variable region ( IGHV) mutated status and presence of deletion 13q14 [del(13q)] were identified as favorable-risk with most patients remaining on lifetime observation. However, other intrinsic factors contributing to CLL disease progression and leading to treatment initiation remain unknown. Our goal was to identify patients at high-risk of progression with added predictive value (APV) beyond IGHV-status and presence of del(13q). Methods: Flow cytometry was used to sort CD19+CD5+ malignant cells in peripheral blood from CLL patients at Huntsman Cancer Institute at University of Utah and triple sequencing generated (tumor whole transcriptome and paired tumor-normal whole exome data). We used SPECTRA, a novel dimension reduction transcriptome approach, to derive a set of independent, intrinsic, quantitative transcriptome tumor variables from 249 CLL patients with quality-controlled transcriptome sequence data. From this set, 182 treatment-naïve (TN) patients with quality-controlled clinical variables were used as a discovery cohort set to derive a SPECTRA-based risk score for TTFT using multivariable Cox proportional hazards modeling. Event time was date of patient specimen acquisition to initiation of frontline therapy with patients censored at last follow-up or death. Gaussian mixture modeling (GMM) was performed to identify evidence for distinct risk groups within the quantitative risk score. GMM-risk groups, and further stratifications by IGHV mutational and del(13q) status, were used in hazard ratios (HR) and to illustrate differences in Kaplan-Meier time-to-event curves. An independent publicly available dataset of 548 TN CLL patients with clinical and transcriptome data for CD19+ enriched cells was used for replication. Results: SPECTRA produced 19 quantitative spectra variables to describe most of the expression variation across the 249 CLL patients. In multivariable Cox modeling for TTFT, 7 spectra were significant and retained in a final model, which GMM separated into a bi-modal distribution and defined high (N=83) and low-risk (N=99) groups (HR=5.7, 95% confidence interval [CI] 3.4-9.7). In a post-hoc model including IGHV mutational and del(13q) status, the spectra-risk score showed significant APV beyond both factors (p&amp;lt;0.001 for both). The discovery study timeframe was 6.3 years (y) and included 109 IGHV-mutated and 75 del(13q) patients. In the IGHV-mutated patients, median TTFT (mTTFT) was significantly shorter for spectra-high risk patients (3.1y) vs. spectra-low risk patients (median not reached [NR]; p&amp;lt;0.001). Similarly, in the del(13q) patients, mTTFT was significantly shorter in high-risk patients (3.3y) compared to median NR for spectra-low (p&amp;lt;0.001) ( Figure). Importantly, stratification of patients into low- and high-risk SPECTRA scores provided significant additional information for prediction of TTFT than IGHV mutational and del(13q) status ( Figure). For validation, we applied the same spectra-risk score to the independent data set containing 301 IGHV-mutated and 278 del(13q) CLL patients. In the replication set, the mTTFT in the IGHV-mutated patients who were spectra-high was 3.3y vs. 11.2y in spectra-low (p&amp;lt;0.001). Similarly, in the replication set of del(13q) patients, spectra-high-risk patients had a mTTFT of 2.7y vs. 10.7y for spectra-low (p&amp;lt;0.001). Conclusion: A transcriptome risk score utilizing novel SPECTRA variables identified a substantial subset of CLL patients with IGHV-mutated status and presence of del(13q) who have uncharacteristically short TTFT, but who would be classified as favorable risk under current best practice stratification. Future directions include validating our results with external patient cohorts from Mayo Clinic and investigating the genetic and transcriptomic profile of SPECTRA low- and high-risk cohorts to determine targetable pathways for treatment.