Knowledge-guided mixture density network for chlorophyll-a retrieval and associated pixel-by-pixel uncertainty assessment in optically variable inland waters.

Machine learning has been increasingly used to retrieve chlorophyll-a (Chl-a) in optically variable waters. However, without the guidance of physical principles or expert knowledge, machine learning may produce biased mapping relationships, or waste considerable time searching for physically infeasible hyperparameter domains. In addition, most Chl-a retrieval models cannot evaluate retrieval uncertainty when ground observations are not available, and the retrieval uncertainty is crucial for understanding the model limitations and evaluating the reliability of retrieval results. In this study, we developed a novel knowledge-guided mixture density network to retrieve Chl-a in optically variable inland waters based on Sentinel-3 Ocean and Land Color Instrument (OLCI) imagery. The proposed method embedded prior knowledge derived from spectral shape classification into the mixture density network. Compared to another deterministic model, the knowledge-guided mixture density network outputted the conditional distribution of Chl-a given an input spectrum, enabling us to estimate the optimal retrieval and the associated uncertainty. The proposed method showed favorable correspondence with the field Chl-a, with root mean square error (RMSE) of 6.56 µg/L, and mean absolute percentage error (MAPE) of 43.64 %. Calibrated against Sentinel-3 OLCI spectrum, the proposed method also performed well when applied to field spectrum (RMSE = 4.58 µg/L, MAPE = 72.70 %), suggesting its effectiveness and good generalization. The proposed method provided the standard deviation of each estimated Chl-a, which enabled us to inspect the reliability of the estimated results and understand the model limitations. Overall, the proposed method improved the Chl-a retrieval in terms of model accuracy and uncertainty evaluation, providing a more comprehensive Chl-a observation of inland waters.

A satellite-based hybrid model for trophic state evaluation in inland waters across China.

Eutrophication is one of the major threats to the inland water ecosystem. Satellite remote sensing provides a promising way to monitor trophic state at large spatial scale in an efficient manner. Currently, most satellite-based trophic state evaluation approaches have focused on water quality parameters retrieval (e.g., transparency, chlorophyll-a), based on which trophic state was evaluated. However, the retrieval accuracies of individual parameter do not meet the demand for accurate trophic state evaluation, especially for the turbid inland waters. In this study, we proposed a novel hybrid model to estimate trophic state index (TSI) by integrating multiple spectral indices associated with different eutrophication level based on Sentinel-2 imagery. The TSI estimated by the proposed method agreed well with the in-situ TSI observations, with root mean square error (RMSE) of 6.93 and mean absolute percentage error (MAPE) of 13.77%. Compared with the independent observations from Ministry of Ecology and Environment, the estimated monthly TSI also showed good consistency (RMSE=5.91,MAPE=10.66%). Furthermore, the congruent performance of the proposed method in the 11 sample lakes (RMSE=5.91,MAPE=10.66%) and the 51 ungauged lakes (RMSE=7.16,MAPE=11.56%) indicated the favorable model generalization. The proposed method was then applied to assess the trophic state of 352 permanent lakes and reservoirs across China during the summers of 2016-2021. It showed that 10%, 60%, 28%, and 2% of the lakes/reservoirs are in oligotrophic, mesotrophic, light eutrophic, and middle eutrophic states respectively. Eutrophic waters are concentrated in the Middle-and-Lower Yangtze Plain, the Northeast Plain, and the Yunnan-Guizhou Plateau. Overall, this study improved the trophic state representativeness and revealed trophic state spatial distribution of Chinese inland waters, which has the significant meanings for aquatic environment protection and water resource management.

DiT-SLAM: Real-Time Dense Visual-Inertial SLAM with Implicit Depth Representation and Tightly-Coupled Graph Optimization.

Recently, generating dense maps in real-time has become a hot research topic in the mobile robotics community, since dense maps can provide more informative and continuous features compared with sparse maps. Implicit depth representation (e.g., the depth code) derived from deep neural networks has been employed in the visual-only or visual-inertial simultaneous localization and mapping (SLAM) systems, which achieve promising performances on both camera motion and local dense geometry estimations from monocular images. However, the existing visual-inertial SLAM systems combined with depth codes are either built on a filter-based SLAM framework, which can only update poses and maps in a relatively small local time window, or based on a loosely-coupled framework, while the prior geometric constraints from the depth estimation network have not been employed for boosting the state estimation. To well address these drawbacks, we propose DiT-SLAM, a novel real-time Dense visual-inertial SLAM with implicit depth representation and Tightly-coupled graph optimization. Most importantly, the poses, sparse maps, and low-dimensional depth codes are optimized with the tightly-coupled graph by considering the visual, inertial, and depth residuals simultaneously. Meanwhile, we propose a light-weight monocular depth estimation and completion network, which is combined with attention mechanisms and the conditional variational auto-encoder (CVAE) to predict the uncertainty-aware dense depth maps from more low-dimensional codes. Furthermore, a robust point sampling strategy introducing the spatial distribution of 2D feature points is also proposed to provide geometric constraints in the tightly-coupled optimization, especially for textureless or featureless cases in indoor environments. We evaluate our system on open benchmarks. The proposed methods achieve better performances on both the dense depth estimation and the trajectory estimation compared to the baseline and other systems.

Evaluation of trophic state for inland waters through combining Forel-Ule Index and inherent optical properties.

Eutrophication is a severe environmental pollution problem for inland waters and poses significant threats to the water safety. Monitoring trophic state of inland waters using optical remote sensing generally requires the inversion of water quality parameters, such as chlorophyll-a, secchi depth, etc. However, the accurate inversion of these individual indicators remains challenging, while the associated retrieval errors can propagate and degrade the evaluation of trophic state. Hence, we proposed a novel monitoring method by developing a Trophic State Index (TSI) based on optical remote-sensing parameters, i.e., Forel-Ule index (FUI) and non-water absorption coefficient at 674 nm (referred to as at-w(674)) retrieved from Sentinel-3 Ocean and Land Color Instrument (OLCI) imagery. The estimated TSI showed favorable correspondence with observed water quality data, including coefficient of determination (r2 = 0.91), root mean squared error (RMSE = 5.54), and mean absolute percentage error (MAPE = 10.69%). Using the Sentinel-3 OLCI data, the proposed method also had very good performance in the field spectrum (MAPE = 5.25 % , RMSE = 3.36). The monthly trophic state evaluation also showed congruence (MAPE = 12.51 % , RMSE = 6.41) with surface water quality monthly report (SWQMR) from the Ministry of Environment and Ecology of the People's Republic of China. The monthly TSI showed favorable agreement for 23 ungauged lakes (RMSE = 7.26, MAPE = 12.78%), indicating potential utility for regional lake water quality monitoring. The proposed method was then applied to 47 other large (>50 km2) water bodies in the Middle-and-Lower watershed of Yangtze River and the Huaihe watershed to evaluate the spatial and temporal variation of trophic state from 2016 to 2020. The TSI results revealed several lakes, such as Lake Honghu and Lake Luoma, with rapidly deteriorating water quality during the study period, while other lakes show relative improvement (e.g., Xiashan Reservoir), indicating unbalanced environmental pressure over the region. Overall, this study showed promising performance and potential for satellite-based monitoring of regional aquatic environments.

Comparison of comfort and complications in breast cancer patients of implantable venous access port (IVAP) with ultrasound guided internal jugular vein (IJV) and axillary vein/subclavian vein (AxV/SCV) puncture: a randomized controlled study protocol.

BACKGROUND: Internal jugular vein (IJV) and axillary vein/subclavian vein (AxV/SCV) are commonly used for implantable venous access port (IVAP) implantation in breast cancer (BC) patients with chemotherapy. Previous studies focused on complications between these different approaches and ignored patient comfort. In this study, we aim to compare patient comfort between IJV and AxV/SCV approaches, as well as surgery duration and complications. METHODS: This is a single-center prospective randomized controlled clinical trial. A total of 200 patients diagnosed with invasive BC will be enrolled in this study. After signing written informed consent, patients schedule to undergo IVAP implantation will be randomized at a 1:1 ratio to receive central venous catheters (CVC) with either IJV or AxV/SCV approaches. Baseline as well as demographic data and procedure time of port implantation will be recorded. All patients will receive assessment of comfort with a comfort scale table at days 1, 2 and 7 after implantation surgery. Patients will be followed up and complications will be recorded until devices are removed at the end of the treatment period, or in case of complications. Patient comfort, procedure time of implantation and complications will be compared and analyzed between these two arms. DISCUSSION: To the best of our knowledge, this is the first study to compare patient comfort as primary outcome measure between IJV and AxV/SCV puncture. This study will further confirm the benefits of ultrasound guidance and may provide a better choice of IVAP implantation for BC patients. TRIAL REGISTRATION: This study has been registered at Chinese Clinical Trial Registry (ChiCTR, www. chictr.org.cn) and Chinese Ethics Committee of Registering Clinical Trials (No. ChiCTR2000034986).

Mobile Breast Cancer e-Support Program for Chinese Women With Breast Cancer Undergoing Chemotherapy (Part 3): Secondary Data Analysis.

BACKGROUND: Many app-based interventions targeting women with breast cancer have been developed and tested for effectiveness. However, information regarding the evaluation of the usage of these interventions is scarce. A better understanding of usage data is important to determine how women use apps and how these interventions affect health outcomes. OBJECTIVE: This study aimed to examine the usage duration and login frequency of an app-based intervention, the Breast Cancer e-Support (BCS) program, and to investigate the association between usage data and participants' demographic and medical characteristics. METHODS: This study is a secondary data analysis of a randomized controlled trial assessing the effectiveness of the BCS program. The BCS program contains four modules: Learning Forum, Discussion Forum, Ask-the-Expert Forum, and Your Story Forum. A total of 57 women in the intervention group accessed the BCS program during their 12-week chemotherapy. The app's background system tracked the usage duration and login frequency for each forum and the entire BCS program. RESULTS: The total usage duration per participant ranged from 0 to 9371 minutes, and the login frequency per participant ranged from 0 to 774 times. The Discussion Forum and the Learning Forum were the most frequently used modules. The general linear model showed that age, education, family monthly income, and employment were associated with BCS usage duration and/or login frequency. Age (F1,45=10.09, P=.003, B=115.34, 95% CI 42.22-188.47) and education level (F1,45=7.22, P=.01, B=1949.63, 95% CI 487.76-3411.50) were positively associated with the usage duration of the entire BCS program. Family monthly income was positively associated with the usage duration of the Learning Forum (F1,45=11.85, P=.001, B=1488.55, 95% CI 617.58-2359.51) and the login frequency of the entire BCS program (F1,45=4.47, P=.04, B=113.68, 95% CI 5.33-222.03). Employment was negatively associated with the usage duration of the Ask-the-expert Forum (F1,45=4.50, P=.04, B=-971.87, 95% CI -1894.66 to -49.07) and the Your Story Forum (F1,45=5.36, P=.03, B=-640.71, 95% CI -1198.30 to -83.11) and positively associated with the login frequency of the entire BCS program (F1,45=10.86, P=.002, B=192.88, 95% CI 75.01-310.74). No statistical differences were found between BCS usage data and cancer stage, BMI, comorbidity, types of surgery, or cycles of chemotherapy. CONCLUSIONS: Overall, this study found considerable variability in the usage of app-based interventions. When health care professionals incorporate app-based interventions into their routine care for women with breast cancer, the learning and discussion functions of apps should be strengthened to promote engagement. Additionally, characteristics of women with breast cancer, such as age, level of education, income, and employment status, should be taken in consideration to develop tailored apps that address their particular needs and therefore improve their engagement with the app. TRIAL REGISTRATION: Australian New Zealand Clinical Trials Registry ACTRN12616000639426; http://www.ANZCTR.org.au/ACTRN12616000639426.aspx.

Akaike's Bayesian Information Criterion for the Joint Inversion of Terrestrial Water Storage Using GPS Vertical Displacements, GRACE and GLDAS in Southwest China.

Global navigation satellite systems (GNSS) techniques, such as GPS, can be used to accurately record vertical crustal movements induced by seasonal terrestrial water storage (TWS) variations. Conversely, the TWS data could be inverted from GPS-observed vertical displacement based on the well-known elastic loading theory through the Tikhonov regularization (TR) or the Helmert variance component estimation (HVCE). To complement a potential non-uniform spatial distribution of GPS sites and to improve the quality of inversion procedure, herein we proposed in this study a novel approach for the TWS inversion by jointly supplementing GPS vertical crustal displacements with minimum usage of external TWS-derived displacements serving as pseudo GPS sites, such as from satellite gravimetry (e.g., Gravity Recovery and Climate Experiment, GRACE) or from hydrological models (e.g., Global Land Data Assimilation System, GLDAS), to constrain the inversion. In addition, Akaike's Bayesian Information Criterion (ABIC) was employed during the inversion, while comparing with TR and HVCE to demonstrate the feasibility of our approach. Despite the deterioration of the model fitness, our results revealed that the introduction of GRACE or GLDAS data as constraints during the joint inversion effectively reduced the uncertainty and bias by 42% and 41% on average, respectively, with significant improvements in the spatial boundary of our study area. In general, the ABIC with GRACE or GLDAS data constraints displayed an optimal performance in terms of model fitness and inversion performance, compared to those of other GPS-inferred TWS methodologies reported in published studies.

Analysis of Key Issues on GNSS-R Soil Moisture Retrieval Based on Different Antenna Patterns.

GNSS-R (Global Navigation Satellite System-Reflectometry) has been demonstrated to be a new and powerful tool to sense soil moisture in recent years. Multi-antenna pattern and single-antenna pattern have been proposed regarding how to receive and process reflected signals. Great efforts have been made concerning ground-based and air-borne observations. Meanwhile, a number of satellite-based missions have also been implemented. For the in-depth study of soil moisture remote sensing by the technique of GNSS-R, regardless of the extraction methods of the reflected signals or the types of the observation platform, three key issues have to be determined: The specular reflection point, the spatial resolution and the detection depth in the soil. However, in current literatures, there are no comprehensive explanations of the above three key issues. This paper conducts theoretical analysis and formula derivation, aiming to systematically and quantitatively determine the extent of soil moisture being detected in three dimensions from the above-mentioned aspects. To further explain how the three factors behave in the specific application, the results of two application scenarios are shown: (1) a ground-based GPS measurement in Marshall, Colorado, US from the Plate Boundary Observatory, corresponding to single-antenna pattern. The relative location of the specular reflection points, the average area of the First Fresnel Ellipse Clusters and the sensing depth of the time-series soil moisture are analyzed, and (2) an aviation experiment conducted in Zhengzhou to retrieve soil moisture content, corresponding to the multi-antenna pattern. The spatial distribution of soil moisture estimation with a certain resolution based on the flight tracks and the relevant sensing depth are manifested. For remote sensing using GNSS reflected signals, BeiDou is different from GPS mainly in the carrier frequency. Therefore, the results of this study can provide references for China's future development of the BeiDou-R technique.

Double-Windows-Based Motion Recognition in Multi-Floor Buildings Assisted by a Built-In Barometer.

Accelerometers, gyroscopes and magnetometers in smartphones are often used to recognize human motions. Since it is difficult to distinguish between vertical motions and horizontal motions in the data provided by these built-in sensors, the vertical motion recognition accuracy is relatively low. The emergence of a built-in barometer in smartphones improves the accuracy of motion recognition in the vertical direction. However, there is a lack of quantitative analysis and modelling of the barometer signals, which is the basis of barometer's application to motion recognition, and a problem of imbalanced data also exists. This work focuses on using the barometers inside smartphones for vertical motion recognition in multi-floor buildings through modelling and feature extraction of pressure signals. A novel double-windows pressure feature extraction method, which adopts two sliding time windows of different length, is proposed to balance recognition accuracy and response time. Then, a random forest classifier correlation rule is further designed to weaken the impact of imbalanced data on recognition accuracy. The results demonstrate that the recognition accuracy can reach 95.05% when pressure features and the improved random forest classifier are adopted. Specifically, the recognition accuracy of the stair and elevator motions is significantly improved with enhanced response time. The proposed approach proves effective and accurate, providing a robust strategy for increasing accuracy of vertical motions.

An Indoor Continuous Positioning Algorithm on the Move by Fusing Sensors and Wi-Fi on Smartphones.

Wi-Fi indoor positioning algorithms experience large positioning error and low stability when continuously positioning terminals that are on the move. This paper proposes a novel indoor continuous positioning algorithm that is on the move, fusing sensors and Wi-Fi on smartphones. The main innovative points include an improved Wi-Fi positioning algorithm and a novel positioning fusion algorithm named the Trust Chain Positioning Fusion (TCPF) algorithm. The improved Wi-Fi positioning algorithm was designed based on the properties of Wi-Fi signals on the move, which are found in a novel "quasi-dynamic" Wi-Fi signal experiment. The TCPF algorithm is proposed to realize the "process-level" fusion of Wi-Fi and Pedestrians Dead Reckoning (PDR) positioning, including three parts: trusted point determination, trust state and positioning fusion algorithm. An experiment is carried out for verification in a typical indoor environment, and the average positioning error on the move is 1.36 m, a decrease of 28.8% compared to an existing algorithm. The results show that the proposed algorithm can effectively reduce the influence caused by the unstable Wi-Fi signals, and improve the accuracy and stability of indoor continuous positioning on the move.

[Vegetation index estimation by chlorophyll content of grassland based on spectral analysis].

Comparing the methods of existing remote sensing research on the estimation of chlorophyll content, the present paper confirms that the vegetation index is one of the most practical and popular research methods. In recent years, the increasingly serious problem of grassland degradation. This paper, firstly, analyzes the measured reflectance spectral curve and its first derivative curve in the grasslands of Songpan, Sichuan and Gongger, Inner Mongolia, conducts correlation analysis between these two spectral curves and chlorophyll content, and finds out the regulation between REP (red edge position) and grassland chlorophyll content, that is, the higher the chlorophyll content is, the higher the REIP (red-edge inflection point) value would be. Then, this paper constructs GCI (grassland chlorophyll index) and selects the most suitable band for retrieval. Finally, this paper calculates the GCI by the use of satellite hyperspectral image, conducts the verification and accuracy analysis of the calculation results compared with chlorophyll content data collected from field of twice experiments. The result shows that for grassland chlorophyll content, GCI has stronger sensitivity than other indices of chlorophyll, and has higher estimation accuracy. GCI is the first proposed to estimate the grassland chlorophyll content, and has wide application potential for the remote sensing retrieval of grassland chlorophyll content. In addition, the grassland chlorophyll content estimation method based on remote sensing retrieval in this paper provides new research ideas for other vegetation biochemical parameters' estimation, vegetation growth status' evaluation and grassland ecological environment change's monitoring.