Robert Snelick

Large-scale evaluation of multimodal biometric authentication using state-of-the-art systems

We examine the performance of multimodal biometric authentication systems using state-of-the-art commercial off-the-shelf (COTS) fingerprint and face biometric systems on a population approaching 1,000 individuals. The majority of prior studies of multimodal biometrics have been limited to relatively low accuracy non-COTS systems and populations of a few hundred users. Our work is the first to demonstrate that multimodal fingerprint and face biometric systems can achieve significant accuracy gains over either biometric alone, even when using highly accurate COTS systems on a relatively large-scale population. In addition to examining well-known multimodal methods, we introduce new methods of normalization and fusion that further improve the accuracy.

Multimodal biometrics: issues in design and testing

Experimental studies show that multimodal biometric systems for small-scale populations perform better than single-mode biometric systems. We examine if such techniques scale to larger populations, introduce a methodology to test the performance of such systems, and assess the feasibility of using commercial off-the-shelf (COTS) products to construct deployable multimodal biometric systems. A key aspect of our approach is to leverage confidence level scores from preexisting single-mode data. An example presents a multimodal biometrics system analysis that explores various normalization and fusion techniques for face and fingerprint classifiers. This multimodal analysis uses a population of about 1000 subjects, a number ten-times larger than seen in any previously reported study. Experimental results combining face and fingerprint biometric classifiers reveal significant performance improvement over single-mode biometric systems.

Synthetic-perturbation tuning of MIMD programs

Synthetic-perturbation tuning (SPT) is a novel technique for assaying and improving the performance of programs on MIMD systems. Conceptually, SPT brings the powerful, mathematical perspective of statistically designed experiments to the interdependent, sometimes refractory aspects of MIMD program tuning. Practically, synthetic perturbations provide a much needed quick-change mechanism for what otherwise would be ad hoc, hand-configured experiment setups. Overall, the technique identifies bottlenecks in programs directly as quantitative effects upon a measured response. SPT works on programs for both shared and distributed memory and it scales well with increasing system size.

Synthetic-perturbation techniques for screening shared memory programs

The synthetic‐perturbation screening (SPS) methodology is based on an empirical approach; SPS introduces artificial perturbations into the MIMD program and captures the effects of such perturbations by using the modern branch of statistics called design of experiments. SPS can provide the basis of a powerful tool for screening MIMD programs for performance bottlenecks. This technique is portable across machines and architectures, and scales extremely well on massively parallel processors. The purpose of this paper is to explain the general approach and to extend it to address specific features that are the main source of poor performance on the shared memory programming model. These include performance degradation due to load imbalance and insufficient parallelism, and overhead introduced by synchronizations and by accessing shared data structures. We illustrate the practicality of SPS by demonstrating its use on two very different case studies: a large image understanding benchmark and a parallel quicksort.

Towards Interoperable Healthcare Information Systems: The HL7 Conformance Profile Approach

Typical healthcare organizations have many proprietary heterogeneous information systems that must exchange data reliably. Seamlessly sharing information among systems is complex. The widely adopted HL7 version 2 messaging standard has helped the process of systems integration. However, using the HL7 standard alone does not ensure system interoperability. The HL7 standard offers a wide range of options. Trading partners, without prior agreement, are not likely to implement options that are compatible. As a result, interoperability is stifled and organizations are left to employ their own ad hoc solutions. Conformance message profiles provide the solution to this problem. Message profiles define a standard template that provides a precise definition of the data exchanged between applications in a common format. Defining a set of message profiles for controlling message exchanges establishes a well defined communications interface among organizations and facilitates interoperability. We present a methodology based on message profiles for defining, implementing, and testing HL7 interfaces. We demonstrate how the use of message profiles improve system interoperability and present a collection of tools that we have developed to facilitate the use of message profiles.

Conformance Profiling Tools

This chapter provides a survey of tools for creating and maintaining conformance profiles and implementation guides. The set of tools is a sample to give the reader an overview of their purpose and capabilities. Conformance profiling tools aid in the specification of requirements. As described, some data exchange standards also provides a framework that aids in constraining the standard for a specific use case. The standards provide “templates” (e.g., a base message) and conformance constructs. Profiling tools can take advantage of these elements to provide a structured and consistent approach for specifying and constraining requirements.

Principles of Conformance Testing

This chapter focuses on the fundamental principles of conformance testing. Although the concepts of conformance testing are the key focal points addressed here, they can’t be discussed in isolation; thus, we include related concepts when appropriate. A testing life cycle and process are presented along with their relationship to the standards development life cycle. An important principle in standards development is to integrate testing early in the process in order to obtain feedback for the authors of the standard. A testing methodology framework is introduced that provides a process for developing, organizing, and managing tests, as well as conducting testing and analyzing the results. A detailed description and example of a Test Plan is given. Since our focus is on communication between distributed applications, we discuss how sending and receiving applications can be tested, using a laboratory test results case study as the context. Next, we offer a set of basic principles for developing Test Plans and Test Cases. Finally, a comparison between capability testing and site-testing is presented. The conformance testing principles discussed in this chapter are applied in Chap. 13 and are examined in relation to test tool implementations in Chap. 14.

Healthcare Data Exchange Standards

The standards landscape from an organizational and scope perspective was presented in Chap. 3. In this chapter we will examine data exchange standards from a technical viewpoint, with a focus on the conformance approach that is used for each standard. This chapter presents a sampling of the most commonly used and well-known healthcare data exchange standards from a variety of categories. It begins with an overview of Version 2.x®, and FHIR® (Fast Healthcare Interoperability Resources) from the family of products specified by HL7 International. This chapter continues with EDIFACT as a European data exchange standard that is also employed in healthcare. ebXML as the foundation for the IHE (Integrating the Healthcare Enterprise) ITI (Information Technology Infrastructure) XDS integration profile is described next. To complete this sampling, details about the HL7 CTS2 (Common Terminology Services, Rel.2) and ISO/FDIS 13120 ClaML (Classification Markup Language) are presented.

Principles of Specifying Conformance

The focus of this chapter is to provide a review of and general guidance for specifying conformance in data exchange standards. The guidance is grouped into three notions. Definition of conformance keywords Conformance clauses Specification of conformance requirements

Testing and Certification Programs

The final step in the cycle of implementation of standards is the certification process. Certification provides assurances to stakeholders that a HIT software product meets the requirements indicated in the specification Certification is a method that formally verifies conformance claims made by a vendor. Certification can be advantageous to the vendor, as it offers a certain degree of confidence to the purchasers. In some cases, use of certified HIT may be required in order for a health care organization or provider to participate in federal programs, such as qualifying to receive Meaningful Use incentive payments (in the US).