Browsing by Subject "performance evaluation"

Now showing 1 - 3 of 3
  • Thumbnail Image
    Item
    Application of Edge-to-Cloud Methods Toward Deep Learning
    (IEEE, 2022-10) Choudhary, Khushi; Nersisyan, Nona; Lin, Edward; Chandrasekaran, Shobana; Mayani, Rajiv; Pottier, Loic; Murillo, Angela P.; Virdone, Nicole K.; Kee, Kerk; Deelman, Ewa; Library and Information Science, School of Computing and Informatics
    Scientific workflows are important in modern computational science and are a convenient way to represent complex computations, which are often geographically distributed among several computers. In many scientific domains, scientists use sensors (e.g., edge devices) to gather data such as CO2 level or temperature, that are usually sent to a central processing facility (e.g., a cloud). However, these edge devices are often not powerful enough to perform basic computations or machine learning inference computations and thus applications need the power of cloud platforms to generate scientific results. This work explores the execution and deployment of a complex workflow on an edge-to-cloud architecture in a use case of the detection and classification of plankton. In the original application, images were captured by cameras attached to buoys floating in Lake Greifensee (Switzerland). We developed a workflow based on that application. The workflow aims to pre-process images locally on the edge devices (i.e., buoys) then transfer data from each edge device to a cloud platform. Here, we developed a Pegasus workflow that runs using HTCondor and leveraged the Chameleon cloud platform and its recent CHI@Edge feature to mimic such deployment and study its feasibility in terms of performance and deployment.
  • Thumbnail Image
    Item
    Ensure that constructive feedback is truly constructive
    (IBJ, 2022-06-17) Westerhaus-Renfrow, Charlotte; Kelley School of Business
  • Thumbnail Image
    Item
    Multi-center evaluation of analytical performance of the Beckman Coulter AU5822 chemistry analyzer
    (Elsevier, 2015-09) Zimmerman, Michelle K.; Friesen, L. R.; Nice, A.; Vollmer, P. A.; Dockery, E. A.; Rankin, J. D.; Zmuda, K.; Wong, S. H.; Department of Pathology and Laboratory Medicine, IU School of Medicine
    Objectives Our three academic institutions, Indiana University, Northwestern Memorial Hospital, and Wake Forest, were among the first in the United States to implement the Beckman Coulter AU5822 series chemistry analyzers. We undertook this post-hoc multi-center study by merging our data to determine performance characteristics and the impact of methodology changes on analyte measurement. Design and methods We independently completed performance validation studies including precision, linearity/analytical measurement range, method comparison, and reference range verification. Complete data sets were available from at least one institution for 66 analytes with the following groups: 51 from all three institutions, and 15 from 1 or 2 institutions for a total sample size of 12,064. Results Precision was similar among institutions. Coefficients of variation (CV) were < 10% for 97%. Analytes with CVs > 10% included direct bilirubin and digoxin. All analytes exhibited linearity over the analytical measurement range. Method comparison data showed slopes between 0.900-1.100 for 87.9% of the analytes. Slopes for amylase, tobramycin and urine amylase were < 0.8; the slope for lipase was > 1.5, due to known methodology or standardization differences. Consequently, reference ranges of amylase, urine amylase and lipase required only minor or no modification. Conclusion The four AU5822 analyzers independently evaluated at three sites showed consistent precision, linearity, and correlation results. Since installations, the test results had been well received by clinicians from all three institutions.