Agency: Department of Health and Human Services | Branch: National Institutes of Health | Program: SBIR | Phase: Phase I | Award Amount: 149.10K | Year: 2015
DESCRIPTION provided by applicant Our research is focused on the use of advanced ontological models as a foundation for computerized Clinical Decision Support CDS systems that link hospitalized patient data routinely captured in electronic medical records EMRs with medical knowledge to effectively influence timely awareness and treatment choices by clinicians Our phase goal is to demonstrate how our advanced CDS technology has the potential to improve preventable mortality outcomes associated with sepsis in ICU patients We distinguish two types of CDS algorithms available today for detecting and or predicting sepsis using EMR data evidence based andquot knowledge drivenandquot detection algorithms and data driven andquot predictiveandquot algorithms based on machine learning ML techniques Recent studies indicate currently available CDS tools do not reduce risk of death in hospitalized patients We believe this may be because diseases such as sepsis are time sensitive complex syndromes and also due to the challenges of computerized reuse of unstructured EMR data Our sepsis ontology models this complexity to a provide enhanced knowledge driven sepsis risk stratified cohort classifications that help guide interventions b support accurate natural language processing NLP of free text clinical notes to enhance real time sepsis risk detection and c improve the accuracy of data driven prediction models when used in conjunction with ML training algorithms Our CDS is based on proprietary cluster computing technology we call andquot VFusionandquot designed to efficiently deal with the generation and practical use of large application domain specific ontologies Our sepsis ontology employs a family of upper level ontologies combined with granular evidence based domain ontologies configurable rule sets e g first order logic based and required components of reference terminologies Our research will use openly available ICU patient data to establish statistical detection prediction performance metrics using this ontology in modes of use as a knowledge based screening tool to detect subtle signs of sepsis in individualized hospitalized patients used in conjunction with ML to improve data driven predictive performance We will measure specificity sensitivity and positive negative predictive power of our hybrid ontology based technology to demonstrate dramatically improved performance over existing CDS algorithms In Phase II we plan a retrospective demonstration with a much larger sample of patients to include non ICU patients in collaboration with a major healthcare system Our product vision is an early inpatient sepsis detection algorithm with high accuracy embodied as a andquot plug in applicationandquot compatible with any modern EMR platform in use at a client hospital effective in both ICU and non ICU care settings PUBLIC HEALTH RELEVANCE Even with the advent of powerful new computer medical record technologies used in hospitals the risk of death has not been reduced in hospitalized patients The goal of this research effort is to combine medical record systems with advanced cognitive technologies that are commonly used today in products such as the iPhone andquot Siriandquot to improve clinician awareness and decisions that will dramatically reduce preventable mortality Our initial use case will be the management of sepsis the leading cause of death in non coronary intensive care units in hospitals
Computer Associates | Date: 2013-02-08
A method and system for identifying a machine used for an online session with an online provider includes executing a lightweight fingerprint code from a provider interface during an online session to collect and transmit machine and session information; generating and storing a machine signature or identity including a machine effective speed calibration (MESC) which may be used to identify the machine when the machine is used in a subsequent online session by a method of matching the machine signature and MESC to a database of machine identities, analyzing a history of the machines online sessions to identify one or more response indicators, such as fraud indicators, and executing one or more responses to the response indicators, such as disabling a password or denying an online transaction, where the response and response indicator may be provider-designated.
Agency: NSF | Branch: Standard Grant | Program: | Phase: Secure &Trustworthy Cyberspace | Award Amount: 50.00K | Year: 2014
The Learning from Authoritative Security Experiment Results (LASER) workshop series focuses on learning from and improving cyber security experiment results. The workshop explores both positive and negative results, the latter of which are not often published. The workshop strives to provide a highly interactive, collegial environment for discussing and learning from experiment design, issues, and outcomes. Ultimately, it seeks to foster a dramatic change in the paradigm of cyber security research and experimentation, improving the overall quality and reporting of practiced science.
Computer Associates | Date: 2014-08-12
A method, system and program product comprise communicating, to a server, a users request for a report comprising a name associated with a portfolio of patent applications, a first value indicating a time frame, and a second value indicating a period of time. The server is operable for using the name for extracting patent application data from databases to a second database and filtering the extracted data using the first value. The filtered data is organized by offices of filing, patent families, classifications, and status. Costs for patent applications associated with the filtered data are determined using the second value. Filing patterns are determined using the offices of filing and the patent families. Filing activity are determined using the classifications. Times to grant for the offices of filing are determined. At least one report is generated and received over the network.
Computer Associates | Date: 2012-01-20
According to one embodiment, a method for managing one or more virtual machines includes generating a request for at least one performance characteristic for at least one virtual machine, the at least one virtual machine being associated with a processing group, the processing group including one or more processing modules; receiving a response to the generated request for at least one performance characteristic for the at least one virtual machine; automatically determining whether an increase in the number of processing modules included in the processing group is required, by analyzing the received response to the generated request; and, in response to a determination that an increase in the number of processing modules included in the processing group is required, automatically adding at least one processing module to the processing group.