Search

The orbit errors, clock errors, and signal-in-space ranging errors (SISREs) of the third generation of BeiDou navigation satellite system (BDS-3) are analyzed by comparing broadcast ephemeris with precise products from January 1, 2019, to June 2, 2021. The average availability of signal-in-space (SIS) over all BDS-3 satellites is 98.41%, and those with the medium earth orbit (MEO) and inclined geosynchronous orbit (IGSO) satellites are 98.74% and 96.86%, respectively. Due to the inter-satellite link of BDS-3, the percentages of ages of data, ephemeris and ages of data, clock less than one hour increase to 97.34% and 96.85%, respectively. The root-mean-square (RMS) of the three-dimensional (3D) orbit errors of a BDS-3 satellite is 0.57 m. The eclipses and maneuvers of the IGSO satellites have a greater impact on orbit accuracy than those of the MEO satellites. The clock errors of different satellites are remarkably positively correlated. The orbit-only, global-average, and the RMS values of worst-case SISREs of BDS-3 are 0.12 m, 0.59 m and 1.10 m, respectively, and the actual URA of 2.0 m in the navigation message might be conservative for integrity since it can overbound the 95th percentile of global-average SISRE for most satellites and RMS of the daily worst-case SISRE for all satellites. A super-Gaussian distribution can describe the BDS-3 orbit errors and clock errors, although a heavier tail exists for clock errors. Since the empirical distribution of worst-case SISRE is neither symmetric nor unimodal, the two-step Gaussian overbounding method is introduced and based on the overbounding distribution parameters the integrity standards for different integrity risk levels are proposed. [ABSTRACT FROM AUTHOR]

Sustainable management of the US Army installations is critical for military training and readiness of forces. However, monitoring military training-induced vegetation cover disturbances using remote sensing data is challenging due to the lack of methodology for optimizing the selection of spectral variables or predictors and spatial modeling methods. This study aimed to propose and demonstrate a methodological solution for this purpose. The study was conducted in the Fort Riley installation in which three training areas were selected to map and monitor the training-induced vegetation cover loss. Sentinel-2 images and field observations of percentage vegetation cover (PVC) were combined at a spatial resolution of 10 m by 10 m to map PVC and its dynamics by comparison of two predictor selection methods and five spatial modeling algorithms based on a total of 304 spectral variables from the images before and after the training. Results showed that overall, the correlation-based predictor selection method reduced the relative root mean square error (RRMSE) of PVC predictions by 4.44% than the random forest (RF)-based predictor selection. Machine learning methods including RF, neural network, and support vector machine overall reduced the RRMSE of PVC predictions by 42.83% compared with multiple linear regression and k-nearest neighbors. Combining correlation-based predictor selection and RF modeling, coupled with leave one out cross validation, provided the greatest potential of increasing the accuracy of monitoring the vegetation cover loss. The findings provided useful implications to develop a near real-time system of monitoring military training-induced vegetation cover loss. [ABSTRACT FROM AUTHOR]

The BeiDou satellite-based augmentation system (BDSBAS) is expected to provide single-frequency (SF) service through B1C signals for users in Chinese mainland and the surrounding areas, in accordance with the standards of the International Civil Aviation Organization (ICAO). The SF service of the BDSBAS is designed to augment both GPS and GLONASS with the performances meeting the requirements of approach with vertical guidance (APV-I), and initial SF service has been provided for test since July 2020. To comprehensively assess the performance of the SF service before its official application to civil aviation users, experiments were conducted with data collected at more than 30 stations all over the BDSBAS service area during the year of operation to June 30, 2021. With the BDSBAS corrections being applied, the average 95th percentiles of horizontal and vertical positioning errors are improved by 45.7% and 41.9% to 1.45 m and 2.77 m, respectively. The horizontal and vertical availabilities of the BDSBAS are 99.98% and 99.89%, respectively, and no risky event of integrity occurs throughout the year. The BDSBAS has generally achieved stable operation and has the capacity to routinely provide single-frequency positioning service as designed. According to the standards of the ICAO, the SF service with the BDSBAS is able to satisfy the positioning accuracy requirement of APV-I operation and has the potential to meet the requirements on availability and integrity. [ABSTRACT FROM AUTHOR]

Background: The relationship between collagen features (CFs) in the tumor microenvironment and the treatment response to neoadjuvant chemoradiotherapy (nCRT) is still unknown. This study aimed to develop and validate a perdition model based on the CFs and clinicopathological characteristics to predict the treatment response to nCRT among locally advanced rectal cancer (LARC) patients. Methods: In this multicenter, retrospective analysis, 428 patients were included and randomly divided into a training cohort (299 patients) and validation cohort (129 patients) [7:3 ratio]. A total of 11 CFs were extracted from a multiphoton image of pretreatment biopsy, and a support vector machine (SVM) was then used to construct a CFs-SVM classifier. A prediction model was developed and presented with a nomogram using multivariable analysis. Further validation of the nomogram was performed in the validation cohort. Results: The CFs-SVM classifier, which integrated collagen area, straightness, and crosslink density, was significantly associated with treatment response. Predictors contained in the nomogram included the CFs-SVM classifier and clinicopathological characteristics by multivariable analysis. The CFs nomogram demonstrated good discrimination, with area under the receiver operating characteristic curves (AUROCs) of 0.834 in the training cohort and 0.854 in the validation cohort. Decision curve analysis indicated that the CFs nomogram was clinically useful. Moreover, compared with the traditional clinicopathological model, the CFs nomogram showed more powerful discrimination in determining the response to nCRT. Conclusions: The CFs-SVM classifier based on CFs in the tumor microenvironment is associated with treatment response, and the CFs nomogram integrating the CFs-SVM classifier and clinicopathological characteristics is useful for individualized prediction of the treatment response to nCRT among LARC patients. [ABSTRACT FROM AUTHOR]

Purpose: Determine the feasibility of a remotely delivered mobile health (mHealth)-supported intervention to improve diet and physical activity in hematologic malignancy survivors. Methods: Pilot randomized controlled trial of a 16-week intervention for improving diet and physical activity: individualized goal-setting (daily steps, sodium, saturated fat, added sugar intake) per feedback from mHealth trackers (Fitbit for activity; Healthwatch360 for diet), supplemented by a Facebook peer support group. Controls accessed the trackers without goal-setting or peer support. Everyone received standardized survivorship counseling with tailored advice from a clinician. Actigraphy and food frequency questionnaires assessed activity and diet at baseline and follow-up. Results: Forty-one participants (51.2% male; median age 45.1 years; 7.0 years from treatment) were randomized (24 intervention; 17 control). Fitbit and Healthwatch360 use were more common among intervention versus control participants (75.0% versus 70.6% and 50.0% versus 17.7% of eligible days, respectively). Most intervention participants (66.7%) engaged with Facebook; overall, 91.7% interacted with the study's mHealth applications. While no comparisons in activity or dietary outcomes between intervention versus control group met statistical significance, the intervention was associated with greater reductions in the targeted dietary factors and improvements in Healthy Eating Index-2015 score, moderate-vigorous physical activity time, and daily steps. Participant retention at 6 months was 90.2%. Conclusions: An intervention for cardiovascular risk reduction based on individualized goal-setting enhanced by mHealth and social media peer support was feasible and acceptable among cancer survivors. Implications for Cancer Survivors: Effective and easily disseminated strategies that improve diet and physical activity in this population are needed. Trial registration: Registered in ClinicalTrials.gov (NCT03574012) on June 29, 2018. [ABSTRACT FROM AUTHOR]

Accurate prediction of peritoneal metastasis for gastric cancer (GC) with serosal invasion is crucial in clinic. The presence of collagen in the tumour microenvironment affects the metastasis of cancer cells. Herein, we propose a collagen signature, which is composed of multiple collagen features in the tumour microenvironment of the serosa derived from multiphoton imaging, to describe the extent of collagen alterations. We find that a high collagen signature is significantly associated with a high risk of peritoneal metastasis (P < 0.001). A competing-risk nomogram including the collagen signature, tumour size, tumour differentiation status and lymph node metastasis is constructed. The nomogram demonstrates satisfactory discrimination and calibration. Thus, the collagen signature in the tumour microenvironment of the gastric serosa is associated with peritoneal metastasis in GC with serosal invasion, and the nomogram can be conveniently used to individually predict the risk of peritoneal metastasis in GC with serosal invasion after radical surgery. Gastric cancer can metastasise to the peritoneal cavity; predicting in which patients this will occur is important for clinical management of the disease. Here, the authors use multi-photon imaging to derive a collagen signature of the primary cancer that allows the prediction of metastasis. [ABSTRACT FROM AUTHOR]

We aimed to investigate whether geniposide, a main component extracted from Gardenia jasminoides Ellis fruit, could exert neuroprotective functions against traumatic brain injury (TBI). Enzyme-linked immunosorbent assay (ELISA) was used for detection of plasma cytokines. Real-time polymerase chain reaction (RT-PCR) was employed for measurements of mRNA levels of cytokines. Neurological outcomes were evaluated by modified neurological severity score (mNSS) and Rota-Rod. Blood–brain barrier (BBB) integrity and brain edema were assessed. Protein expression was tested by Western blot. The plasma levels of interleukin (IL)-1β, IL-6, IL-8 and IL-10 were all elevated in patients with TBI compared to those of healthy controls. TBI rats treated with geniposide showed lower mNSS and longer fall latency time than untreated TBI rats. BBB integrity was maintained and brain edema was reduced by geniposide treatment in TBI rats. Plasma levels of IL-1β, IL-6 and IL-8 were significantly repressed by geniposide treatment in TBI rats, whereas IL-10 level was upregulated. mRNA expression levels of these cytokines in the brain tissues of TBI rats exhibited the same trends of changes. By testing p38 mitogen-activated protein kinase and NF-κB p65 activities, it was observed that phosphorylated (p)-p38 and p-p65 were dramatically inhibited by geniposide. In conclusion, geniposide exerts neuroprotective functions against TBI by inhibiting p-p38 and p-p65. [ABSTRACT FROM AUTHOR]

Since millimeter accuracy is required in many GNSS applications such as real-time zenith wet delay (ZWD) estimation, the higher-order ionospheric delays on GNSS signals are no longer negligible. We calculated the second-order ionospheric delays ( I 2 ) and analyzed the impact on the ZWD estimation with GPS-only and combined GPS/BDS observations. The undifferenced PPP model with fixed coordinates was used to estimate the ZWD and horizontal gradients. The method of blockwise sequential least squares was utilized to eliminate the receiver clock biases and compute the I 2 impact on the ZWDs. The I 2 delays on each GNSS satellite observations were calculated with the CODE final TEC map and the 12th generation of the international geomagnetic reference field (IGRF-12) model. The statistical results with the actual observation geometry show that the I 2 delays can reach over 10 mm during the daytime, and the corresponding impact on the estimated ZWD can reach up to several millimeters. At station HKWS, the maximum I 2 impact with GPS only reaches up to 3.1 mm and is still 2.4 mm when both GPS and BDS observations are used. The simulated I 2 impact on the ZWD could reach several millimeters, even though the TEC and geomagnetic values were calculated from relatively moderate background models. Compared with the 5–10 mm precision of real-time ZWD estimation, the I 2 delays must not be ignored, especially during high VTEC periods. [ABSTRACT FROM AUTHOR]

Southwest China is one of three major forest regions in China and plays an important role in carbon sequestration. Accurate estimations of changes in aboveground biomass are critical for understanding forest carbon cycling and promoting climate change mitigation. Southwest China is characterized by complex topographic features and forest canopy structures, complicating methods for mapping aboveground biomass and its dynamics. The integration of continuous Landsat images and national forest inventory data provides an alternative approach to develop a long-term monitoring program of forest aboveground biomass dynamics. This study explores the development of a methodological framework using historical national forest inventory plot data and Landsat TM time-series images. This method was formulated by comparing two parametric methods: Linear Regression for Multiple Independent Variables (MLR), and Partial Least Square Regression (PLSR); and two nonparametric methods: Random Forest (RF) and Gradient Boost Regression Tree (GBRT) based on the state of forest aboveground biomass and change models. The methodological framework mapped Pinus densata aboveground biomass and its changes over time in Shangri-la, Yunnan, China. Landsat images and national forest inventory data were acquired for 1987, 1992, 1997, 2002 and 2007. The results show that: (1) correlation and homogeneity texture measures were able to characterize forest canopy structures, aboveground biomass and its dynamics; (2) GBRT and RF predicted Pinus densata aboveground biomass and its changes better than PLSR and MLR; (3) GBRT was the most reliable approach in the estimation of aboveground biomass and its changes; and, (4) the aboveground biomass change models showed a promising improvement of prediction accuracy. This study indicates that the combination of GBRT state and change models developed using temporal Landsat and national forest inventory data provides the potential for developing a methodological framework for the long-term mapping and monitoring program of forest aboveground biomass and its changes in Southwest China. [ABSTRACT FROM AUTHOR]

Rapid urban sprawl and re-construction of old towns have been leading to great changes of land use in cities of China. To witness short-term urban land use changes, rapid or real time remote sensing images and effective detection methods are required. With the availability of short repeat cycle, relatively high spatial resolution, and weather-independent Synthetic Aperture Radar (SAR) remotely sensed data, detection of short-term urban land use changes becomes possible. This paper adopts newly released Sentinel-1 SAR data for urban change detection in Tianhe District of Guangzhou City in Southern China, where dramatic urban redevelopment practices have been taking place in past years. An integrative method that combines the SAR time series data and a spectral angle mapping (SAM) was developed and applied to detect the short-term land use changes. Linear trend transformations of the SAR time series data were first conducted to reveal patterns of substantial changes. Spectral mixture analysis was then conducted to extract temporal endmembers to reflect the land development patterns based on the SAR backscattering intensities over time. Moreover, SAM was applied to extract the information of significant increase and decrease patterns. The results of validation and method comparison showed a significant capability of both the proposed method and the SAR time series images for detecting the short-term urban land use changes. The method received an overall accuracy of 78%, being more accurate than that using a bi-temporal image change detection method. The results revealed land use conversions due to the removal of old buildings and their replacement by new construction. This implies that SAR time series data reflects the spatiotemporal evolution of urban constructed areas within a short time period and this study provided the potential for detecting changes that requires continuously short-term capability, and could be potential in other landscapes. [ABSTRACT FROM AUTHOR]

Key message : Crown width models developed using nonlinear simultaneous equations with a two-step procedure provided the best performance and are recommended to predict the crown components and crown width of Prince Rupprecht larch. Abstract: Crown width (CW) is defined as an average of two crown diameters at two cardinal directions: east-west and south-north, obtained from measurements of four crown radii (crown components) at four directions: east, west, south, and north. CW is one of the important tree variables in forest growth and yield modeling, and forest management. Reliable estimates of CW are central elements of forest management. However, the additivity of CW and crown components and their inherent correlations have not been addressed in existing CW models. In this study, two alternative procedures for nonlinear simultaneous equations (NSE) were used to develop CW models. The procedures included a disaggregation model structure with one- and two-step, proportional weighting systems, and two commonly used additivity methods, adjustment in proportion (AP) and ordinary least squares with separating regression (OLSSR). These methods were compared using data from a total of 3369 Prince Rupprecht larch ( Larix principis- rupprechtii Mayr.) trees in 116 permanent sample plots in northern China. It was found that these methods effectively ensured that the sum of the crown components was equal to twice the total CW. The NSE accounted for the inherent correlations among the crown components and CW. The CW models developed using the NSE with the two-step procedure provided the best performance, followed by the models developed with AP and OLSSR. This methodology can be adapted to develop a system of CW models for other tree species. [ABSTRACT FROM AUTHOR]

Understanding the spatial distribution of PM concentration and its contributing environmental variables is critical to develop strategies of addressing adverse effects of the particulate pollution. In this study, a range of meteorological and land use factors were incorporated into a linear regression (LR) model and a logistic model-based regression (LMR) model to simulate the annual and winter PM concentrations. The vegetation cover, derived from a linear spectral unmixing analysis (LSUA), and the normalized difference built-up index (NDBI), were found to improve the goodness of fit of the models. The study shows that (1) both the LR and the LMR agree on the predicted spatial patterns of PM concentration and (2) the goodness of fit is higher for the models established based on the annual PM concentration than that based on the winter PM. The modeling results show that higher PM concentration coincided with the major urban area for the annual average but focused on the suburban and rural areas for the winter. The methods introduced in this study can potentially be applied to similar regions in other developing countries. [ABSTRACT FROM AUTHOR]

Key message: A climate-sensitive aboveground biomass model developed by combining a nonlinear mixed-effects model and dummy variable approach led to higher prediction accuracy of biomass than those without climatic variables for three larch species. Abstract: As native species and being widely distributed in northeastern and northern China, larch forests play a pivotal role in maintaining forest ecosystem functions and mitigation of carbon concentration at the atmosphere. However, the spatial sensitivity of growing and aboveground biomass (AGB) of larch species to climate change is not known. In this study, a climate-sensitive AGB model was developed by combining nonlinear mixed-effects modeling and a dummy variable approach to account for the spatial sensitivity of AGB of three larch species including Dahurian larch ( Larix gmelinii (Rupr.) Kuzen.), Korean larch ( Larix olgensis Henry.) and Prince Rupprecht larch ( Larix principis rupprechtii Mayr.) to climate change in northeastern and northern China. We examined the AGB values of 256 larch trees growing in five secondary climate zones: mid-temperature humid climatic zone, mid-temperature sub-humid climatic zone, mid-temperature semi-arid climatic zone, cool temperature humid climatic zone and warm temperature sub-humid climatic zone in northeastern and northern China. The results showed that the long-term averages of growing season temperature and total growing season precipitation, and the mean temperature and precipitation of wettest quarter, had significant ( α = 0.05) effects on AGB. The prediction accuracy of the developed model was much higher than that of the model at the population average level and its base model. Excessive rain and high mean temperature during the growing season increased the AGB values of larch trees, while abundant precipitation and high mean temperature in the wettest quarter led to the decrease of AGB. The AGB of the larch species from the mid-temperature semi-arid climatic zone had the greatest values, which indicated that climate conditions were more favorable in this zone than in the other four secondary climate zones. It is expected that the findings from this study can be combined with the knowledge of adaptive management to reduce the risks and uncertainties associated with forest management decisions. [ABSTRACT FROM AUTHOR]

Key message : Nonlinear error-in-variable models can advance the development of the systems of additive biomass equations and lead to much higher prediction accuracy of tree biomass than nonlinear seemingly unrelated regression. Abstract: In this study, the approach of nonlinear error-in-variable models (NEIVM) was compared with nonlinear seemingly unrelated regressions (NSUR) for developing a system of nonlinear additive biomass equations using the data collected in Southern China for Pinus massoniana Lamb. Various tree variables were assessed to explore their contributions to improvement of biomass prediction using the systems of equations. It was found that diameter at breast height ( D), total tree height ( H) and crown width (CW) significantly contributed to the increase of prediction accuracy. The combinations of D, H, and CW led to three sets of independent variables: (1) D alone; (2) both D and H; and (3) D, H and CW together, which were used for the development of one-predictor, two-predictor and three-predictor systems of biomass equations, respectively. The results showed that both NEIVM and NSUR had high prediction accuracy of biomass for all the systems of biomass equations. For the one-predictor systems of biomass equations, both NEIVM and NSUR led to very similar predictions. However, for the two-predictor and three-predictor systems of biomass equations, the prediction accuracy of NEIVM was much higher than that of NSUR. When the two-predictor system of equations was used, in particular, NEIVM with one-step procedure, that is, by directly partitioning total tree biomass into four basic components, showed a higher accuracy of biomass prediction than NSUR for all the one-predictor, two-predictor and three-predictor systems of equations. This study implies that the NEIVM approach could provide a greater potential to develop a system of biomass equations that are dependent on the predictors with significant measurement errors. [ABSTRACT FROM AUTHOR]

Fly ash is a hazardous byproduct of municipal solid wastes incineration (MSWI). An alkali activated blast furnace slag-based cementitious material was used to stabilize/solidify the fly ash at experimental level. The characteristics of the stabilized/solidified fly ash, including metal leachability, mineralogical characteristics and the distributions of metals in matrices, were tested by toxic characteristic leaching procedure (TCLP), X-ray diffraction (XRD) and scanning electron microscopy-energy dispersive spectrometer (SEM-EDS) respectively. Continuous acid extraction was utilized to extract metal ions and characterize their leaching behavior. The stabilization/ solidification procedure for MSWI fly ash demonstrates a strong fixing capacity for the metals by the formation of CS-H phase, hydrated calcium aluminosilicate and ettringite. The stabilized/solidified fly ash shows a dense and homogeneous microstructure. Cr is mainly solidified in hydrated calcium aluminosilicate, C-S-H and ettringite phase through physical encapsulation, precipitation, adsorption or substitution mechanisms, and Pb is mainly solidified in C-S-H phase and absorbed in the Si-O structure. [ABSTRACT FROM AUTHOR]

Accurate spatial estimation of forest carbon stocks and their spatial uncertainties at local, regional, national, and global scales is a critical step in global carbon cycle modeling and management. This study aimed at enhancing the methods that are currently used in this area by combining plot data from the forest inventory and analysis program of the U.S. Forest Service and free landsat thematic mapper image data. Three mapping methods including linear regression, sequential Gaussian co-simulation, and block co-simulation algorithm were compared with respect to the accuracy of forest carbon stock estimates obtained for a study area in Southern Illinois, USA. The results indicated that although the linear regression resulted in smaller prediction errors than the sequential Gaussian co-simulation and the block co-simulation approaches, it also produced both negative and unreasonably large estimates, which is a serious drawback. Moreover, the sequential Gaussian co-simulation and the block co-simulation produced not only accurate carbon predictions, but also uncertainties for the local estimates. In addition, the block co-simulation approach scaled up both forest carbon stocks and the input uncertainties from finer to coarser spatial resolutions as is required for mapping forest carbon at national and global scales. Thus, the co-simulation and block co-simulation algorithms resolved an important current methodological challenge. [ABSTRACT FROM AUTHOR]

Having non-negligible impact on the code range observables, multipath delay is one of the error sources that limit GNSS positioning accuracy. Due to the relatively stationary geometry, multipath effects for signals from geostationary earth orbit (GEO) satellites are even more difficult to mitigate by merely increasing the observing periods or averaging over multiple epochs. To investigate the characteristics of code, multipath effects for the BeiDou Navigation Satellite System (BDS) GEO satellites, a linear combination of observations reflecting code multipath was employed and BDS multipath time series over long periods were analyzed with the Fourier transform, correlation and wavelet transform. The amplitudes of GEO multipath series vary from <1.0 m to around 2.0 m, and the periods of the dominant daily repeating components fall between 86,130 and 86,280 s. The low-frequency components were extracted, and most cross-correlation coefficients between the low-frequency components of two consecutive days are larger than 0.7. When the low-frequency components of the first day are subtracted from the multipath time series of the second day, a decrease of more than 25 % is found in terms of the code standard deviations. By correcting the observables with low-frequency multipath of the previous day, the precisions of code-only single-point positioning using ionosphere-free linear combination of BDS first and second or first and third frequencies can be improved. Precision improvements in north, east and up components for two stations in Perth, Australia were shown to be 0.2, 0.5 and 0.4 m, and 0.3, 0 and 0.5 m, respectively. [ABSTRACT FROM AUTHOR]

Background: Accurate prediction of peritoneal recurrence for gastric cancer (GC) is crucial in clinic. The collagen alterations in tumor microenvironment affect the migration and treatment response of cancer cells. Herein, we proposed multitask machine learning-based tumor-associated collagen signatures (TACS), which are composed of quantitative collagen features derived from multiphoton imaging, to simultaneously predict peritoneal recurrence (TACSPR) and disease-free survival (TACSDFS).Among 713 consecutive patients, with 275 in training cohort, 222 patients in internal validation cohort, and 216 patients in external validation cohort, we developed and validated a multitask machine learning model for simultaneously predicting peritoneal recurrence (TACSPR) and disease-free survival (TACSDFS). The accuracy of the model for prediction of peritoneal recurrence and prognosis as well as its association with adjuvant chemotherapy were evaluated.The TACSPR and TACSDFS were independently associated with peritoneal recurrence and disease-free survival in three cohorts, respectively (all P < 0.001). The TACSPR demonstrated a favorable performance for peritoneal recurrence in all three cohorts. In addition, the TACSDFS also showed a satisfactory accuracy for disease-free survival among included patients. For stage II and III diseases, adjuvant chemotherapy improved the survival of patients with low TACSPR and low TACSDFS, or high TACSPR and low TACSDFS, or low TACSPR and high TACSDFS, but had no impact on patients with high TACSPR and high TACSDFS.The multitask machine learning model allows accurate prediction of peritoneal recurrence and survival for GC and could distinguish patients who might benefit from adjuvant chemotherapy.Methods: Accurate prediction of peritoneal recurrence for gastric cancer (GC) is crucial in clinic. The collagen alterations in tumor microenvironment affect the migration and treatment response of cancer cells. Herein, we proposed multitask machine learning-based tumor-associated collagen signatures (TACS), which are composed of quantitative collagen features derived from multiphoton imaging, to simultaneously predict peritoneal recurrence (TACSPR) and disease-free survival (TACSDFS).Among 713 consecutive patients, with 275 in training cohort, 222 patients in internal validation cohort, and 216 patients in external validation cohort, we developed and validated a multitask machine learning model for simultaneously predicting peritoneal recurrence (TACSPR) and disease-free survival (TACSDFS). The accuracy of the model for prediction of peritoneal recurrence and prognosis as well as its association with adjuvant chemotherapy were evaluated.The TACSPR and TACSDFS were independently associated with peritoneal recurrence and disease-free survival in three cohorts, respectively (all P < 0.001). The TACSPR demonstrated a favorable performance for peritoneal recurrence in all three cohorts. In addition, the TACSDFS also showed a satisfactory accuracy for disease-free survival among included patients. For stage II and III diseases, adjuvant chemotherapy improved the survival of patients with low TACSPR and low TACSDFS, or high TACSPR and low TACSDFS, or low TACSPR and high TACSDFS, but had no impact on patients with high TACSPR and high TACSDFS.The multitask machine learning model allows accurate prediction of peritoneal recurrence and survival for GC and could distinguish patients who might benefit from adjuvant chemotherapy.Results: Accurate prediction of peritoneal recurrence for gastric cancer (GC) is crucial in clinic. The collagen alterations in tumor microenvironment affect the migration and treatment response of cancer cells. Herein, we proposed multitask machine learning-based tumor-associated collagen signatures (TACS), which are composed of quantitative collagen features derived from multiphoton imaging, to simultaneously predict peritoneal recurrence (TACSPR) and disease-free survival (TACSDFS).Among 713 consecutive patients, with 275 in training cohort, 222 patients in internal validation cohort, and 216 patients in external validation cohort, we developed and validated a multitask machine learning model for simultaneously predicting peritoneal recurrence (TACSPR) and disease-free survival (TACSDFS). The accuracy of the model for prediction of peritoneal recurrence and prognosis as well as its association with adjuvant chemotherapy were evaluated.The TACSPR and TACSDFS were independently associated with peritoneal recurrence and disease-free survival in three cohorts, respectively (all P < 0.001). The TACSPR demonstrated a favorable performance for peritoneal recurrence in all three cohorts. In addition, the TACSDFS also showed a satisfactory accuracy for disease-free survival among included patients. For stage II and III diseases, adjuvant chemotherapy improved the survival of patients with low TACSPR and low TACSDFS, or high TACSPR and low TACSDFS, or low TACSPR and high TACSDFS, but had no impact on patients with high TACSPR and high TACSDFS.The multitask machine learning model allows accurate prediction of peritoneal recurrence and survival for GC and could distinguish patients who might benefit from adjuvant chemotherapy.Conclusions: Accurate prediction of peritoneal recurrence for gastric cancer (GC) is crucial in clinic. The collagen alterations in tumor microenvironment affect the migration and treatment response of cancer cells. Herein, we proposed multitask machine learning-based tumor-associated collagen signatures (TACS), which are composed of quantitative collagen features derived from multiphoton imaging, to simultaneously predict peritoneal recurrence (TACSPR) and disease-free survival (TACSDFS).Among 713 consecutive patients, with 275 in training cohort, 222 patients in internal validation cohort, and 216 patients in external validation cohort, we developed and validated a multitask machine learning model for simultaneously predicting peritoneal recurrence (TACSPR) and disease-free survival (TACSDFS). The accuracy of the model for prediction of peritoneal recurrence and prognosis as well as its association with adjuvant chemotherapy were evaluated.The TACSPR and TACSDFS were independently associated with peritoneal recurrence and disease-free survival in three cohorts, respectively (all P < 0.001). The TACSPR demonstrated a favorable performance for peritoneal recurrence in all three cohorts. In addition, the TACSDFS also showed a satisfactory accuracy for disease-free survival among included patients. For stage II and III diseases, adjuvant chemotherapy improved the survival of patients with low TACSPR and low TACSDFS, or high TACSPR and low TACSDFS, or low TACSPR and high TACSDFS, but had no impact on patients with high TACSPR and high TACSDFS.The multitask machine learning model allows accurate prediction of peritoneal recurrence and survival for GC and could distinguish patients who might benefit from adjuvant chemotherapy. [ABSTRACT FROM AUTHOR]