Estimating The Distribution of Livestock In Space And Time: A Time-Series Contribution To Livestock Density Mapping

More than one billion people rely on livestock for income, nutrition, and social cohesion, however livestock keeping can facilitate disease transmission and contribute to climate change. While data on the distribution of livestock thus have broad utility across a range of applications, efforts to map the distribution of livestock on a large scale are limited to the Gridded Livestock of the World (GLW) project. We present a complimentary effort to map the distribution of cattle and pigs in Malawi, Uganda, Democratic Republic of Congo (DRC), and South Sudan. In contrast to GLW, which uses dasymmetric modeling applied to census data to produce time-stratified estimates of livestock counts and spatial density, our work uses complex survey data and distinct modeling methods to generate a time-series of livestock distribution, defining livestock density as the ratio of animals to humans. In addition to favorable cross-validation results and general agreement with national density estimates derived from external data on national human and livestock populations, our results demonstrate extremely good agreement with GLW-3 estimates, supporting the validity of both efforts. Our results furthermore offer a high-resolution time series result and employ a definition of density which is particularly well-suited to the study of livestock-origin zoonoses.

Incorporating external trial data to improve survival extrapolations: a pilot study of the COU-AA-301 trial

Survival extrapolation plays a key role within cost effectiveness analysis and is often subject to substantial uncertainty. Use of external data to improve extrapolations has been identified as a key research priority. We present findings from a pilot study using data from the COU-AA-301 trial of abiraterone acetate for metastatic castration-resistant prostate cancer, to explore how external trial data may be incorporated into survival extrapolations. External trial data were identified via a targeted search of technology assessment reports. Four methods using external data were compared to simple parametric models (SPMs): informal reference to external data to select appropriate SPMs, piecewise models with, and without, hazard ratio adjustment, and Bayesian models fitted with a prior on the shape parameter(s). Survival and hazard plots were compared, and summary metrics (point estimate accuracy and restricted mean survival time) were calculated. Without consideration of external data, several SPMs may have been selected as the ‘best-fitting’ model. The range of survival probability estimates was generally reduced when external data were included in model estimation, and external hazard plots aided model selection. Different methods yielded varied results, even with the same data source, highlighting potential issues when integrating external trial data within model estimation. By using external trial data, the most (in)appropriate models may be more easily identified. However, benefits of using external data are contingent upon their applicability to the research question, and the choice of method can have a large impact on extrapolations.

Network models of prostate cancer immune microenvironments identify ROMO1 as heterogeneity and prognostic marker

Prostate cancer (PCa) is the fifth leading cause of death from cancer in men worldwide. Its treatment remains challenging due to the heterogeneity of the tumor, mainly because of the lack of effective and targeted prognostic markers at the system biology level. First, the data were retrieved from TCGA dataset, and valid samples were obtained by consistent clustering and principal component analysis; next, key genes were analyzed for prognosis of PCa using WGCNA, MEGENA, and LASSO Cox regression model analysis, while key genes were screened based on disease-free survival significance. Finally, TIMER data were selected to explore the relationship between genes and tumor immune infiltration, and GSCAlite was used to explore the small-molecule targeted drugs that act with them. Here, we used tumor subtype analysis and an energetic co-expression network algorithm of WGCNA and MEGENA to identify a signal dominated by the ROMO1 to predict PCa prognosis. Cox regression analysis of ROMO1 was an independent influence, and the prognostic value of this biomarker was validated in the training set, the validated data itself, and external data, respectively. This biomarker correlates with tumor immune infiltration and has a high degree of infiltration, poor prognosis, and strong correlation with CD8+T cells. Gene function annotation and other analyses also implied a potential molecular mechanism for ROMO1. In conclusion, we putative ROMO1 as a portal key prognostic gene for the diagnosis and prognosis of PCa, which provides new insights into the diagnosis and treatment of PCa.

Building Reliable Radiomic Models Using Image Perturbation

Radiomic model reliability is a central premise for its clinical translation. Presently, it is assessed using test-retest or external data, which, unfortunately, is often scarce in reality. Therefore, we aimed to develop a novel image perturbation-based method (IPBM) for the first of its kind toward building a reliable radiomic model. We first developed a radiomic prognostic model for head-and-neck cancer patients on a training (70%) and evaluated on a testing (30%) cohort using C-index. Subsequently, we applied the IPBM to CT images of both cohorts (Perturbed-Train and Perturbed-Test cohort) to generate 60 additional samples for both cohorts. Model reliability was assessed using intra-class correlation coefficient (ICC) to quantify consistency of the C-index among the 60 samples in the Perturbed-Train and Perturbed-Test cohorts. Besides, we re-trained the radiomic model using reliable RFs exclusively (ICC>0.75) to validate the IPBM. Results showed moderate model reliability in Perturbed-Train (ICC:0.565, 95%CI:0.518-0.615) and Perturbed-Test (ICC:0.596, 95%CI:0.527-0.670) cohorts. An enhanced reliability of the re-trained model was observed in Perturbed-Train (ICC:0.782, 95%CI:0.759-0.815) and Perturbed-Test (ICC:0.825, 95%CI:0.782-0.867) cohorts, indicating validity of the IPBM. To conclude, we demonstated capability of the IPBM toward building reliable radiomic models, providing community with a novel model reliability assessment strategy prior to prospective evaluation.

Prediction of Bike Share Demand by Machine Learning

In the fourth industrial revolution period, multinational companies and start-ups have applied a sharing economy concept to their business and have attempted to better serve customer demand by integrating demand prediction results into their business operations. For survival amongst today’s fierce competition, companies need to upgrade their prediction model to better predict customer demand in a more accurate manner. This study explores a new feature for bike share demand prediction models that resulted in an improved RMSLE score. By applying this new feature, the number of daily vehicle accidents reported in the Washington, D.C. area, to the Random Forest, XGBoost, and LightGBM models, the RMSLE score results improved. Many previous studies have primarily focused on feature engineering and regression techniques within given dataset. However, this study is meaningful because it focuses more on finding a new feature from an external data source.

How Can Remote Sensing Reduce Required Human Intervention in Robotic Forest Regeneration

In this article, we introduce an alternative solution for forest regeneration based on unmanned ground vehicles (UGV) and describe requirements for external data, which could significantly increase the level of automation. Over the past few decades, the global forested area has decreased, and there is a great need to restore and regenerate forests. Challenges such as the lack of labor and high costs demand innovative approaches for forest regeneration. Mechanization has shown satisfactory results in terms of time-efficient planting, although its usage is limited by high operational costs. Innovative technologies must be cost-efficient and profitable for large scale usage. Automation could make mechanized forest regeneration feasible. Forest regeneration operations can be automated using a purpose built unmanned platform. We developed a concept to automate forest planting operations based on mobility platform. The system requires external data for efficient mobility in clear-cut areas. We developed requirements for external data, analyzed available solutions, and experimented with the most promising option, the SfM (structure from motion) technique. Earth observation data are useful in the planning phase. A DEM (digital terrain model) for UGV planter operations can be constructed using ALS (airborne laser scanning), although it may be restricted by the cost. Low-altitude flights by drones equipped with digital cameras or lightweight laser scanners provided a usable model of the terrain. This model was precise (3–20 cm) enough for manually planning of the trajectory for the planting operation. This technique fulfilled the system requirements, although it requires further development and will have to be automated for operational use.

Poor prognosis of stage I lung adenocarcinoma patients determined by elevated expression over pre/minimally invasive status of COL11A1 and THBS2 in the focal adhesion pathway

Around 20% stage I lung adenocarcinoma (LUAD) patients die within five years after surgery, and efforts for developing gene-expression based models for risk-tailored post-surgery treatment are largely unsatisfactory due to overfitting-related lack of validation and extrapolation. Because patients with adenocarcinomas in situ (AIS) and minimally invasive (MIA) LUAD are completely curable by surgical resection, we hypothesize that poor-prognosis stage I patients may exhibit key molecular characteristics deviating from AIS/MIA. We first found focal adhesion (FA) as the only pathway significantly perturbed at both genomic and transcriptomic levels by comparing 98 AIS/MIA and 99 invasive LUAD patients. Then, we identified two FA pathway genes (COL11A1 and THBS2) strongly upregulated from AIS/MIA to stage I while expressed steadily from normal to AIS/MIA. Furthermore, unsupervised clustering separated stage I patients into two molecularly and prognostically distinct subtypes (S1 and S2) based solely on the expression levels of COL11A1 and THBS2 (FA2). Subtype S1 looked like AIS/MIA, whereas S2 exhibited more somatic alterations, elevated expression of COL11A1 and THBS2, and more activated cancer-associated fibroblast (CAF). The prognostic performance of the knowledge-driven and overfitting-resistant FA2 model was validated with 12 external data sets and may help reliably identify high-risk stage I patients for more intensive post-surgery treatment.

An Elevation Ambiguity Resolution Method Based on Segmentation and Reorganization of TomoSAR Point Cloud in 3D Mountain Reconstruction

Tomographic Synthetic Aperture Radar (TomoSAR) is a breakthrough of the traditional SAR, which has the three-dimentional (3D) observation ability of layover scenes such as buildings and high mountains. As an advanced system, the airborne array TomoSAR can effectively avoid temporal de-correlation caused by long revisit time, which has great application in high-precision mountain surveying and mapping. The 3D reconstruction using TomoSAR has mainly focused on low targets, while there are few literatures on 3D mountain reconstruction. Due to the layover phenomenon, surveying in high mountain areas remains a difficult task. Consequently, it is meaningful to carry out the research on 3D mountain reconstruction using the airborne array TomoSAR. However, the original TomoSAR mountain point cloud faces the problem of elevation ambiguity. Furthermore, for mountains with complex terrain, the points located in different elevation periods may intersect. This phenomenon increases the difficulty of solving the problem. In this paper, a novel elevation ambiguity resolution method is proposed. First, Density-Based Spatial Clustering of Applications with Noise (DBSCAN) and Gaussian Mixture Model (GMM) are combined for point cloud segmentation. The former ensures coarse segmentation based on density, and the latter allows fine segmentation of the abnormal categories caused by intersection. Subsequently, the segmentation results are reorganized in the elevation direction to reconstruct all possible point clouds. Finally, the real point cloud can be extracted automatically under the constraints of the boundary and elevation continuity. The performance of the proposed method is demonstrated by simulations and experiments. Based on the airborne array TomoSAR experiment in Leshan City, Sichuan Province, China in 2019, the 3D model of the surveyed mountain is presented. Moreover, three kinds of external data are applied to fully verify the validity of this method.