![]() Our work targets nucleus segmentation workflows as key part of this open challenge. Nevertheless, development of robust and efficient computerized image analysis workflows to reliably extract imaging features from WSIs remains an open challenge. A number of projects have developed tissue image analysis methods and shown that quantitative image characterizations from pathology images can be used to predict outcome and treatment responses. In addition, the Food and Drug Administration (FDA) has recently approved the use of digitized tissue images for diagnostic purposes, both recognizing the value of whole slide tissue imaging in clinical settings and paving the way for routine use of tissue imaging, which we expect will lead to significant increases in the number and volume of WSI datasets for imaging studies. ![]() Whole slide tissue images enable quantitative and reproducible analysis of tissue morphology-the importance of improving precision and reducing inter-observer variability in pathology studies is well recognized. Advances in digital microscopy scanners have made it possible to capture tissue images at very high resolutions state-of-the-art scanners can capture images at 100,000 × 100,000 square pixel resolutions and can automatically scan hundreds of tissue slides rapidly, thanks to sophisticated auto-focusing mechanisms. Manual examination of tissue specimens, however, has had limited use in biomedical research because it is a labor-intensive and time-consuming process. Diseased tissue shows changes in tissue morphology, which are indicators of disease onset and progress and provide rich information with which to study disease biology at the subcellular level. Microscopic examination of whole slide tissue specimens by pathologists has long been considered a de facto standard for disease diagnosis and prognosis. We propose and experimentally evaluate a software platform that integrates a suite of methods and tools to enable automatic parameter tuning in analysis algorithms that segment nuclei in digitized images of tissue specimens fixed on glass slides, also called Whole Slide Tissue Images (WSIs). Our results using three real-world image segmentation workflows demonstrate that the proposed solution is able to (1) search a small fraction (about 100 points) of the parameter space, which contains billions to trillions of points, and improve the quality of segmentation output by × 1.20, × 1.29, and × 1.29, on average (2) decrease the execution time of a segmentation workflow by up to 11.79× while improving output quality and (3) effectively use parallel systems to accelerate parameter tuning and segmentation phases. These capabilities are packaged in a Docker container for easy deployment and can be used through a friendly interface extension in 3D Slicer. In addition, the methodology is developed to execute on high-performance computing systems to reduce the execution time of the parameter tuning phase. It implements several optimization methods to search the parameter space efficiently. Our software platform adjusts the parameters of a nuclear segmentation algorithm to maximize the quality of image segmentation results while minimizing the execution time. ![]() This is a time-consuming and computationally expensive process automating this step facilitates more robust image segmentation workflows and enables more efficient application of image analysis in large image datasets. Input parameters in many nucleus segmentation workflows affect segmentation accuracy and have to be tuned for optimal performance. The shape, size, and texture features of nuclei in tissue are important biomarkers for disease prognosis, and accurate computation of these features depends on accurate delineation of boundaries of nuclei. ![]() The goal of our work is to provide an approach for improving the accuracy of nucleus/cell segmentation pipelines by tuning their input parameters. We propose a software platform that integrates methods and tools for multi-objective parameter auto-tuning in tissue image segmentation workflows. ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |