Browsing by Author "Lin, Edward"

Now showing 1 - 2 of 2
  • 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
    Prospective randomized controlled trial of an injectable esophageal prosthesis versus a sham procedure for endoscopic treatment of gastroesophageal reflux disease
    (Springer Verlag, 2010-03-03) Fockens, Paul; Cohen, Lawrence; Edmundowicz, Steven A.; Binmoeller, Kenneth; Rothstein, Richard I.; Smith, Daniel; Lin, Edward; Nickl, Nicholas; Overholt, Bergein; Kahrilas, Peter J.; Vakil, Nimish; Aziz Hassan, Ayman M. Abdel; Lehman, Glen A.; Medicine, School of Medicine
    Background This study aimed to assess whether endoscopic implantation of an injectable esophageal prosthesis, the Gatekeeper Reflux Repair System (GK), is a safe and effective therapy for controlling gastroesophageal reflux disease (GERD). Methods A prospective, randomized, sham-controlled, single-blinded, international multicenter study planned final enrollment of 204 patients in three groups: up to 60 lead-in, 96 GK, and 48 sham patients. The sham patients were allowed to cross over to the GK treatment arm or exit the study at 6 months. The primary end points were (1) reduction in serious device- and procedure-related adverse device effects compared with a surgical composite complication rate and (2) reduction in heartburn symptoms 6 months after the GK procedure compared with the sham procedure. The secondary end point was improved esophageal pH (total time pH was <4) 6 months after the GK procedure compared with baseline.Results A planned interim analysis was performed after 143 patients were enrolled (25 lead-in, 75 GK, and 43 sham patients), and the GK study was terminated early due to lack of compelling efficacy data. Four reported serious adverse events had occurred (2 perforations, 1 pulmonary infiltrate related to a perforation, and 1 severe chest pain) at termination of the study with no mortality or long-term sequelae. Heartburn symptoms had improved significantly at 6 months compared with baseline in the GK group (p < 0.0001) and the sham group (p < 0.0001), but no significant between-group difference in improvement was observed (p = 0.146). Esophageal acid exposure had improved significantly at 6 months compared with baseline in the GK group (p = 0.021) and the sham group (p = 0.003), but no significant between-group difference in improvement was observed (p = 0.27). Conclusions The GK procedure was associated with some serious but infrequent complications. No statistically significant difference in outcomes was observed between the treatment and control groups at 6 months compared with baseline.