No abstract

High-resolution terrain models of open-pit mine highwalls and benches are essential in developing new automated slope monitoring systems for operational optimization. This paper presents several contributions to the field of remote sensing in surface mines providing a practical framework for generating high-resolution images using low-trim Unmanned Aerial Vehicles (UAVs). First, a novel mobile application was developed for autonomous drone flights to follow mine terrain and capture high-resolution images of the mine surface. In this article, case study is presented showcasing the ability of developed software to import area terrain, plan the flight accordingly, and finally execute the area mapping mission autonomously. Next, to model the drone’s battery performance, empirical studies were conducted considering various flight scenarios. A multivariate linear regression model for drone power consumption was derived from experimental data. The model has also been validated using data from a test flight. Finally, a genetic algorithm for solving the problem of flight planning and optimization has been employed. The developed power consumption model was used as the fitness function in the genetic algorithm. The designed algorithm was then validated using simulation studies. It is shown that the offered path optimization can reduce the time and energy of high-resolution imagery missions by over 50%. The current work provides a practical framework for stability monitoring of open-pit highwalls while achieving required energy optimization and imagery performance.

Motivation Organization of the organoid models, imaged in 3D with a confocal microscope, is an essential morphometric index to assess responses to stress or therapeutic targets. In fact, differentiating malignant and normal cells is often difficult in monolayer cultures. But in 3D culture, colony organization can provide a clear set of indices for differentiating malignant and normal cells. The limiting factors are delineating each cell in a 3D colony in the presence of perceptual boundaries between adjacent cells and heterogeneity associated with cells being at different cell cycles. Results In a previous paper, we defined a potential field for delineating adjacent nuclei, with perceptual boundaries, in 2D histology images by coupling three deep networks. This concept is now extended to 3D and simplified by an enhanced cost function that replaces three deep networks with one. Validation includes four cell lines with diverse mutations, and a comparative analysis with the UNet models of microscopy indicates an improved performance with the F1-score of 0.83. Availability All software and annotated images are available through GitHub and Bioinformatics online. The software includes the proposed method, UNet for microscopy that was extended to 3D, and report generation for profiling colony organization. Supplementary information Supplementary data are available at Bioinformatics online and Github.

Reprogramming the tumor stroma is an emerging approach to circumventing the challenges of conventional cancer therapies. This strategy, however, is hampered by the lack of a specific molecular target. We previously reported that stromal fibroblasts (FBs) with high expression of CD36 could be utilized for this purpose. These studies are now expanded to identify the secreted factors responsible for tumor suppression. Methodologies included 3D colonies, fluorescent microscopy coupled with quantitative techniques, proteomics profiling, and bioinformatics analysis. The results indicated that the conditioned medium (CM) of the CD36+ FBs caused growth suppression via apoptosis in the triple-negative cell lines of MDA-MB-231, BT549, and Hs578T, but not in the ERBB2+ SKBR3. Following the bioinformatic analysis of the CM of CD36+ versus CD36− FBs, we determined KLF10 as one of the transcription factors responsible for growth suppression. We also identified FBLN1, SLIT3, and PENK as active ligands, where their minimum effective concentrations were determined. Finally, in MDA-MB-231, we showed that a mixture of FBLN1, SLIT3, and PENK could induce an amount of growth suppression similar to the CM of CD36+ FBs. In conclusion, our findings suggest that these ligands, secreted by CD36+ FBs, can be targeted for breast cancer treatment.

BackgroundNuclear segmentation is an important step for profiling aberrant regions of histology sections. If nuclear segmentation can be resolved, then new biomarkers of nuclear phenotypes and their organization can be predicted for the application of precision medicine. However, segmentation is a complex problem as a result of variations in nuclear geometry (e.g., size, shape), nuclear type (e.g., epithelial, fibroblast), nuclear phenotypes (e.g., vesicular, aneuploidy), and overlapping nuclei. The problem is further complicated as a result of variations in sample preparation (e.g., fixation, staining). Our hypothesis is that (i) deep learning techniques can learn complex phenotypic signatures that rise in tumor sections, and (ii) fusion of different representations (e.g., regions, boundaries) contributes to improved nuclear segmentation.ResultsWe have demonstrated that training of deep encoder-decoder convolutional networks overcomes complexities associated with multiple nuclear phenotypes, where we evaluate alternative architecture of deep learning for an improved performance against the simplicity of the design. In addition, improved nuclear segmentation is achieved by color decomposition and combining region- and boundary-based features through a fusion network. The trained models have been evaluated against approximately 19,000 manually annotated nuclei, and object-level Precision, Recall, F1-score and Standard Error are reported with the best F1-score being 0.91. Raw training images, annotated images, processed images, and source codes are released as a part of the Additional file 1.ConclusionsThere are two intrinsic barriers in nuclear segmentation to H&E stained images, which correspond to the diversity of nuclear phenotypes and perceptual boundaries between adjacent cells. We demonstrate that (i) the encoder-decoder architecture can learn complex phenotypes that include the vesicular type; (ii) delineation of overlapping nuclei is enhanced by fusion of region- and edge-based networks; (iii) fusion of ENets produces an improved result over the fusion of UNets; and (iv) fusion of networks is better than multitask learning. We suggest that our protocol enables processing a large cohort of whole slide images for applications in precision medicine.Electronic supplementary materialThe online version of this article (10.1186/s12859-018-2285-0) contains supplementary material, which is available to authorized users.