Information Technology Frontiers for a New Millenium Human Centered Systems

	Introduction Knowledge and Distributed Intelligence (KDI) Knowledge Networking (KN) STIMULATE Digital Libraries (DL) Initiative, Phase Two "Virtual Los Angeles": A realtime visualization system for large scale urban environments Virtual reality technology for simulating medical procedures Robotic surgery NASA's "software scalpel" and virtual reality tools NASA's computerized breast cancer diagnostic tool Battlefield awareness Mobile autonomous robot software (MARS) Information visualization Visualization and virtual reality for collaboration and manufacturing Virtual Worlds NOAA collaborations using NCSA's Habanero NOAAServer Speech recognition technology Spanish language interfaces Systems Integration for Manufacturing Applications (SIMA) MMC Collaboratory Biomedical technology collaboratories Electronic Notebooks for collaborative environments Unified Medical Language System (UMLS) Medline Plus The Visible Human project Universal access Usability Fingerprint and mugshots standards Regional Technology in Education Consortia (RTEC) Regional Educational Laboratories National Institute on Disability and Rehabilitation Research (NIDRR) Web Accessibility Initiative (WAI)
Introduction	HuCS R&D focuses on improving technologies that enable humans, computing systems, and information resources to work together more effectively and transparently. Federal investments in HuCS research lead to increased accessibility and usability of computing systems and communications networks. HuCS researchers endeavor to: Improve the ability to create and use worldwide information systems with enormous amounts of diverse content Increase the effectiveness and comfort with which people can exploit computers Extend computer availability to everyone in our society Create environments where sensors and mobile equipment adapt to humans, resulting in more rapid and effective completion of work tasks Scientists, engineers, physicians, educators, students, the workforce, and the general public are all potential beneficiaries of HuCS technologies.   FY 2000 focus areas include: Knowledge repositories, information agents, and digital libraries that enable people to better collect, manage, analyze, summarize, and present data and images from complex databases Increased research data collection and annotation for training systems and for experimental evaluation Multimodal human-computer interfaces -- including speech recognition tools, audio interfaces, and optic and haptic (touch-sensitive) devices -- that help provide universal access to information technologies and enable people, regardless of their physical ability, education, or culture, to communicate with computers by sight, sound, touch, and gesture Visualization, virtual reality, and 3-D imagery tools that enhance human perception and understanding in simulation and problem-solving environments Collaboratories that facilitate knowledge sharing, group decision-making, control of remote instruments, and data sharing across large distributed systems Multilingual technologies that enhance foreign document and speech-to-speech translation, interpretation, and understanding, including collaboration across national boundaries
Knowledge and Distributed Intelligence (KDI)	Recent advances in computing power and Internet connectivity are reshaping relationships among people and organizations and transforming the processes of discovery, learning, and communication. These advances are creating unprecedented opportunities to provide rapid and efficient access to enormous amounts of knowledge and information, to study vastly more complex systems, and to increase our understanding of learning and intelligence in these systems.   To help exploit these opportunities, NSF's Foundation-wide Knowledge and Distributed Intelligence (KDI) program seeks to accelerate technological innovation and the adoption of technologies into society. Anticipated benefits of KDI research include: Deep, far-reaching scientific discovery Increases in scientific productivity and in the timeliness and quality of the results Increased ability to handle problems of greater complexity, scale, and structure The creation of new scientific and engineering communities to exploit novel discoveries Enhancements in science and engineering education through development of richer learning tools, technologies, and environments, and more universal access to these resources and tools Enhanced understanding of the processes and results of learning A more complete understanding of the fundamental processes of distributed intelligence in natural and artificial systems An understanding of the legal, ethical, and societal implications of the increased capability to gather and access information Enhanced ability to communicate and transfer new understanding and technological innovations to society Advances in statistical data reduction, data visualization, data mining, and data organization for retrieval to use vast stores of data Improved methods to express, compute with, and evaluate different types of uncertainties in real-world data
Knowledge Networking (KN)	In FY 1999, KDI focused on Knowledge Networking (KN), Learning and Intelligent Systems (LIS), and New Computational Challenges (NCC). HuCS R&D was primarily conducted in KN. NSF's LIS efforts are described in the ETHR section of this book, while NCC efforts are described in the HECC section.   The goals of NSF's KN efforts are to (1) understand the fundamental processes through which knowledge is created, communicated, validated, and valued in distributed information systems, and (2) improve the technical, social, educational, and economic performance of knowledge generation and use, collaborative computation, and remote interaction.   KN supports multidisciplinary research to develop and employ the next generation of communication networks, associated information repositories, collaborative technologies, and knowledge management techniques to gather, create, distribute, use, and evaluate knowledge in new and more secure ways. These activities include research on the human, behavioral, social, and ethical dimensions of knowledge networking.
STIMULATE	The Speech, Text, Image, and MULtimedia Advanced Technology Effort (STIMULATE) is a multiagency collaboration among researchers funded by NSF, DARPA, and NSA that seeks to improve human-computer interaction. Researchers focus on creating new, innovative computer interfaces using multiple languages and modalities, such as gestures, facial expressions, handwriting, images, and video. Multimodal devices developed through STIMULATE will help people with disabilities, language challenges, and reading deficiencies gain access to computing and communications technologies.     Human-computer interface researchers at Rutgers University have developed unique force feedback transducers to enable computer control and monitoring of graded exercises for extremities recovering from traumatic injury. The transducers, embedded in tight-fitting gloves for hands and boots for feet, apply programmed forces and movement to the injured extremities for prescribed durations. Medical specialists supervise rehabilitation remotely over computer networks, and patients have access to a library of computerized rehabilitation exercises. The photograph on the left illustrates the Rutgers University force feedback system. The figures on the right show a patient undergoing remote rehabilitation. This research is supported under the NSF STIMULATE initiative and the Aid to Persons with Disabilities program.
Digital Libraries (DL) Initiative, Phase Two	The original Digital Libraries Initiative (DLI), begun in 1995, was a joint four year collaborative program involving NSF, DARPA, and NASA. Its goal was to advance methods to collect, store, organize, and use widely distributed knowledge resources that contain diverse types of information and content stored in a variety of electronic forms.   Currently in Phase Two, DL is now a multiagency effort of NSF, DARPA, NASA, NLM, the Library of Congress, and the National Endowment for the Humanities conducted in partnership with the National Archives and Records Administration, the Smithsonian Institution, and the Institute of Museum and Library Sciences. DL Phase Two will develop the next generation of digital libraries in such areas as education, engineering and design, Earth and space sciences, biosciences, geography, economics, and the arts and humanities; advance the use and usability of globally-distributed networked information resources; and encourage new and existing communities to focus on innovative applications areas. DL Phase Two will also address the digital libraries life cycle from information creation, access, and use, to archiving and preservation, and will study the long term social, behavioral, and economic effects of digital libraries in human activities such as research, education, commerce, defense, health services, and recreation.   A special feature highlighting the Digital Libraries Phase Two Initiative begins on page 80.
"Virtual Los Angeles": A realtime visualization system for large scale urban environments	NSF-funded researchers on UCLA's Urban Simulation Team have developed "Virtual Los Angeles" -- a long term urban simulation project that combines custom simulation software and realtime database technologies with efficient modeling methods. The simulation is derived from aerial photographs of Los Angeles with street level imagery and 3-D geometry, city engineering maps, internally generated vegetation libraries, and numerous site visits. Accurate to the level of graffiti on walls and signs in windows, Virtual Los Angeles is a realistic 3-D visual model of an extremely dense urban environment, containing data covering more than 4,000 square miles. It allows users to fly or walk through the virtual city, complete with details like trees, street signs, and other visual landmarks.   The Virtual Los Angeles database is a long term project that can be used to help address a multitude of real-world civil engineering, urban design, and urban planning problems, including the development of a "Virtual World" data server displaying a realtime virtual model of the entire Los Angeles Basin. Researchers are working on applications including a 3-D in-car navigation system, a helicopter flight training system, and other transit and traffic management tools. The Los Angeles Fire Department and other emergency response teams will use these tools as an aid to emergency response -- for example, to learn whether adjoining buildings are higher or lower than available rescue equipment can reach.     Taking a walk downtown in "Virtual Los Angeles." The goal of this NSF-supported project at UCLA is to help solve civil engineering and urban design- and urban planning problems.
Virtual reality technology for simulating medical procedures	In medical procedures, the ability to visualize large amounts of data is often crucial. In clinical care, the ability to "see" areas deep inside the body -- especially the brain -- can determine whether a surgical procedure or other therapeutic intervention can be successfully performed with minimum patient trauma. These fundamental requirements are increasingly being aided by virtual reality technology that provides realistic visualization of data in real time. NIH/NCRR researchers use virtual reality technologies to simulate operations and other medical procedures, and to interface with high resolution instruments like atomic force microscopes. They are also building and evaluating surgical collaboratories that allow remote users to gain high speed access to scientific instruments for basic research, molecular visualization, and surgical and other therapeutic interventions such as radiation treatment planning.   In FY 1999, NCRR conducted research in augmented reality, a form of virtual reality, to support image-guided neurosurgery, initiating eight collaborative demonstrations at resource centers covering a wide range of technology R&D. Building on these efforts in FY 2000, NCRR will conduct additional R&D focusing on new uses for emerging virtual reality technologies and environments.
Robotic surgery	NSF-supported research at Johns Hopkins University focuses on Computer-Integrated Surgery (CIS), a rapidly expanding field where advances in computing and engineering technologies are helping to overcome the limitations of traditional surgery. By extending human surgeons' ability to plan and implement surgical interventions more accurately and less invasively, CIS systems address a vital national need to reduce surgical costs and improve both the clinical outcomes and the efficiency of healthcare delivery.   According to the November 1998 Scientific American, "coming changes in hospital technology may within a few years make television dramas...less exciting but should help operating rooms become healthier places for patients. The tense crowd gathered around the operating table is likely to be replaced by a silent robot cutting and suturing through an incision less than an inch long. The surgeon doing the operation will be seated at a console peering deep into the body through a virtual reality headset that combines the actual image of the patient with up-to-date diagnostic images showing the precise location of the problem. He may even have rehearsed the operation in cyberspace before the incision was made. The manipulators that control the surgical tools will be capable of the tiniest, most precise motions -- and they will compensate for any tremors in the surgeon's hand."   To foster advances in robotic surgery, NSF is establishing a new Engineering Research Center in Computer-Integrated Surgical Systems and Technology. Headed by Johns Hopkins University, this collaborative project involves Johns Hopkins School of Medicine, Johns Hopkins Applied Physics Laboratory, MIT's Project on Image Guided Surgery, the Surgical Planning Group at Harvard University's Brigham and Women's Hospital, and Carnegie Mellon University and its affiliate, Shadyside Hospital. The Center draws upon experts in computer science and robotics; electrical, mechanical, and biomedical engineering; and physicians specializing in fields such as radiology, neurosurgery, urology, orthopedics, and ophthalmology.   NSF-funded HuCS research projects at Johns Hopkins include robotic steady hand assistance for neurosurgery and microsurgery; robotic camera and instrument holding and monitoring for laparoscopic surgical procedures; treatment planning for and automation of pattern placement therapy for minimally invasive cancer treatment; image processing techniques for the next generation RoboDoc system for hip replacement surgery; automation of percutaneous needle insertion for kidney surgery; and modeling and instrument tracking for orthopedic and craniofacial procedures. MIT research includes using medical scans to model a patient's internal anatomy. One Carnegie Mellon project is the HipNav computer-based surgical assistant that helps more accurately plan and place the socket portion of a hip implant. The goals of this research are to develop a family of systems that extend the human ability to operate on very small structures inside the body without large incisions, assist in routine surgical tasks as "robotic residents" operating under a surgeon's supervision, help surgeons rehearse and optimize surgical plans based on preoperative patient-specific data, and permit surgeons to assist in procedures conducted in remote areas.
NASA's software scalpel and virtual reality tools	Researchers at the NASA Ames Center for Bioinformatics are developing a "software scalpel" that -- combined with clear, accurate, 3-D images of the human head -- is helping doctors practice complex facial reconstructive surgery and visualize its outcome more accurately. Using the new Virtual Surgery Cutting Tool software, a physician wearing 3-D glasses can see an image of a patient's head from all angles on a computer monitor or on the surface of a large "immersive virtual reality workbench." This computerized reconstruction can also be applied to mastectomy patients who require breast reconstruction and children who need reconstructive surgery to correct head and facial deformities.   In the future, virtual reality will allow surgeons to rehearse many complex procedures before operating, and will serve as a powerful teaching tool for medical students. A digital library of computerized "virtual patients" will be created so that physicians can share information about infrequently performed medical procedures.
NASA's computerized breast cancer diagnostic tool	NASA researchers are developing and testing a smart probe for use in breast cancer detection and analysis. A spin-off from a computerized robotic brain surgery "assistant," this probe is designed to see a lump, determine by its features if it is cancerous, and predict how the disease may progress, aiding doctors in expediting diagnoses and helping them to suggest individualized treatment. The predictive capability uses trainable neural net software to look for telltale characteristics or patterns.
Battlefield awareness	DARPA battlefield awareness R&D develops and demonstrates technologies to extract information from imagery, understand news broadcasts, and integrate information from heterogeneous data sources.   As part of DARPA's Text, Radio, Video, and Speech (TRVS) program, researchers are developing, demonstrating, and evaluating automated news information processing techniques that will: Determine the structure of and automatically index data from broadcast information sources with different speakers, topics, and distractions with 85 percent reliability Extract facts from diverse media with 90 percent accuracy and transcribe spoken language with less than a 20 percent word error rate Transcribe audio portions of radio and TV news broadcasts in multiple languages with a goal of handling two languages, Mandarin and Spanish, with less than a 35 percent error rate in FY 1999 Researchers are developing a plug-and-play framework to extract content from diverse TRVS inputs to provide for effective interaction with the extracted information. DARPA's long term goal is to intercept spoken language communications and automatically process the data to extract information from enemy sources. By processing the data through analysis algorithms, the military can extract key facts about the enemy's plans and infer enemy intent.
Mobile autonomous robot software (MARS)	DARPA is funding R&D in mobile autonomous robot software (MARS) that enables mobile robots to perform tasks such as the safe, reliable, realtime operation of unmanned military systems in dynamic, realistic, and unstructured operational environments. These tasks may include platform mobility, navigation, obstacle avoidance, payload operation, and human-robot interaction without dependence on a human operator. MARS techniques will aid DoD in domains such as reconnaissance, surveillance and target acquisition, countermine and explosive ordnance disposal, force protection and physical security, and logistics support operations.   Robots employing this software will exhibit: Reprogrammable, goal-directed behaviors that exploit symbolic, human-supplied information Reflexive, sensor-mediated behaviors that are stable in complex, uncertain, and dynamic environments A high degree of adaptability, including a facility for learning The ability to operate safely and reliably in close proximity to humans MARS will develop reusable software for autonomous mobile systems that encompasses: A software framework that lets robots synthesize deliberative (symbol mediated) and reactive (sensor mediated) control Software composition methods that incorporate both hand-coded programming and automated learning-driven coding so robots can function better in unpredictable environments
Information visualization	Information visualization is the process of transforming inherently spatial data and information into a visual form, allowing users to understand the information better. As part of its HuCS R&D, NIST is examining the utility and feasibility of 3-D visualization and computer graphic techniques to access, manipulate, and exchange complex information.   In FY 1999, NIST is developing evaluation methods and test corpora for measuring the usability and scalability of interfaces for visualizing a statistical text retrieval system that accesses large collections of data. NIST researchers are also developing a visualization system that monitors user navigation paths through a Web site, helping usability engineers to determine the effectiveness of the site's user interface. NIST plans to investigate the use of large screen visualization of documents and related information to support effective human-computer interaction in an electronic "smart" meeting room.     NIST is examining the usefulness and feasibility of employing 3-D visualization and computer graphic techniques to access, manipulate, and exchange complex information. In the global 3-D model to the left, clusters of conceptually similar documents, and the relationships among them, are portrayed.
Visualization and virtual reality for collaboration and manufacturing	NIST is collaborating with industrial partners in applying 3-D visualization to accelerate the manufacturing process. NIST researchers are analyzing the usability and performance capability of a number of virtual environment and Web visualization and collaboration tools, including Virtual Reality Modeling Language (VRML) to visualize distributed collaborative interactions in high performance "smart" spaces, facilitating design and engineering analyses for remote participants. These also include commercial off-the-shelf software to create visualizations of the factory floor assembly line and to support parts design and assembly.     Used to model manufacturing assembly operations, simulation tools will help improve the productivity of engineers who plan manufacturing operations. The graphic on the left demonstrates an ergonomic modeling software system where a virtual human is assembling the base of a miter saw using assembly instructions specified in a NIST-developed computer language. Ergonomic simulation environments will be used to validate assembly processes, perform ergonomic analyses, and train workers on actual assembly operations.   The Virtual Machine Shop is being used to develop, test, and evaluate interfaces to manufacturing simulation environments. The graphic on the right shows a NIST-developed machine shop simulation where raw materials are prepared, cutting tools are set up and loaded into machine tool magazines, parts are machined, and materials are handled. NIST is developing interfaces between manufacturing simulation software and software that supports engineering and manufacturing activities.
Virtual Worlds	NOAA's Pacific Marine Environmental Laboratory (PMEL) and Old Dominion University are continuing their partnership to provide scientists with general purpose Java tools, enabling them to combine graphical objects created from commercial and customized routines, such as GIS and Matlab, to produce their own integrated, customized Virtual Worlds for interactive exploration on the Web.   Virtual Worlds constructed by scientists with this Java toolkit will include graphical objects (e.g., contour slices, vectors, bathymetry, topography), 3-D domains, specification of pre-defined viewpoints, proximity sensor triggers, user touch sensor triggers, and audio and video capabilities as appropriate. PMEL and Old Dominion will also explore realtime generation of Virtual Worlds from archived data.     NOAA/PMEL and Old Dominion University have developed three dimensional representations of NOAA El Niño data in VRML and a prototype, customized, complex Virtual World containing these 3-D graphical objects. Users can interact with the graphical objects and with the virtual world via the Web. Depicted above is the Equatorial Pacific sea level colored with ocean temperatures from NOAA's network of El Niño buoys in the tropical Pacific Ocean. These 3-D VRML renderings show El Niño (December 1997) and La Niña (December 1998), along the equator from 8N to 8S, viewed from the Andes mountains in Peru.
NOAA collaborations using NCSA's Habanero	NOAA is using advanced distributed network collaboration tools, such as NCSA's Habanero, to enhance research collaborations needing access to NOAA's data holdings. Habanero is a collaborative framework and set of applications that can be used to create and work in shared applications from remote locations over the Internet. It enables developers of groupware applications to build powerful collaborative software in a reduced amount of time by allowing them to create or convert existing applications into collaborative applications. Useful Habanero tools include a whiteboard, a collaborative text editor, a Java graph, and various chat utilities.   Tools are evolving to provide on-the-fly production of integrated browser images from profile datasets stored at two or more geographically-distributed data archives. This allows two or more scientists at different locations to view these images and have annotations simultaneously appear on all collaborating scientists' desktops. All collaborators will be able to interactively post messages and change the data being displayed.
NOAAServer	NOAA has implemented a prototype application that contains robust collaborative features such as text and line annotations, highlighting of individual profiles, display of data values, and support for multiple users, and plans to expand the prototype to include user feedback and access to multiple NOAAServer data servers. This project leverages the collaborative tools developed at NCSA and the ongoing development of graphics that integrate data from multiple archives at different locations under the NOAAServer project.
Speech recognition technology	NIST collaborates with DARPA and NSA on speech technologies by developing benchmark tests to track technology development and measures of progress. In its work with DARPA, NIST researchers focus on continuous speech recognition and on understanding the large and difficult vocabulary used in radio and TV news broadcasts. For NSA, NIST is implementing test protocols for transcribing conversational telephone-line speech and speaker recognition technologies. These tests involve scientists in the United Kingdom, France, and Germany.   In FY 1999, NIST is working on a testbed to develop and test spoken-language based information retrieval technologies. Research focuses on user interactions and on experiments with dialogue management protocols. FY 1999 and FY 2000 speech technology research projects focus on automatic speech recognition and transcription, spoken document retrieval, topic detection and tracking, and new information extraction measures and tests.
Spanish language interfaces	To help support Spanish-speaking individuals in the emerging information society, researchers at the Oregon Graduate Institute's Center for Spoken Language Understanding (CSLU) and researchers at the Universidad de las Americas-Puebla -- jointly funded by NSF, DARPA, Office of Naval Research (ONR), Fonix, and Intel -- collaborated on a Spanish version of a spoken language toolkit for human computer interaction.   The CSLU Speech Toolkit is a comprehensive software environment for research, development, and education of spoken language systems that integrates speech recognition, speech synthesis, facial animation, and speaker recognition, and features authoring and analysis tools enabling quick and easy development of desktop- and telephone-based speech applications. The software is available free of charge for research and education at not-for-profit institutions.   The Spanish language CSLU Speech Toolkit, which incorporates Spanish speech recognition and synthesis, enables faculty, staff, and students to develop Spanish-spoken language systems and to teach courses in Mexico, accelerating work in speech technology infrastructure, research, system development, and education.
Systems Integration for Manufacturing Applications (SIMA)	At NIST, SIMA researchers are developing technologies to improve the interfaces between engineering and manufacturing software systems and between people and the applications themselves. Through the development of authoritative, Internet-accessible repositories of scientific and engineering knowledge and the development of computerized collaboration and remote operation tools, NIST supports interoperability among advanced manufacturing systems applications.     The NIST Ceramics WebBook provides scientists and engineers with Internet access to an extensive, authoritative information database. This image of a large-grain defect in ceramic material is located in the flaw catalog in the WebBook's Fractography section. The strength of many advanced ceramics reflects flaws present in the material and the material's intrinsic fracture toughness. Strength-test results must be interpreted in the context of these flaws whether this testing has been performed for quality control, materials development, or design purposes.   To assist businesses in plant and factory design, NIST R&D is enhancing commercial modeling and simulation software systems with programmable human ergonomic models, new human modeling programming languages, and the integration of physics-based models of machine and object behavior. These technologies will enable businesses to build "simulated factories" with prototyped plant layouts, material and component routings, and human factors prior to committing significant investments in plant redesigns or new factories. The results of these efforts are captured in consensus standards efforts conducted through standards organizations like the International Organization for Standardization.   NIST is continuing its work on a manufacturing collaboratory to provide manufacturers, distributors, and researchers with the structured methods, guidelines, and standards needed to implement collaboratory technologies in business applications. This has already helped research in robotic arc welding, which requires asynchronous and synchronous collaboration for local and remote support of diverse data formats such as video, still images, audio, text, database records, and image annotations. One product of this work will be a collaborative tool prototype with annotation capabilities that synchronize various time-indexed data streams, such as audio, video, and log data.
MMC Collaboratory	Linking instrumentation at six DOE user facilities nationwide, DOE's Materials Microcharacterization Collaboratory (MMC) enables scientists and researchers to access state-of-the-art tools remotely to study the structure and composition of metals, ceramics, and alloys at the atomic and molecular level. Scientists at other DOE and university labs can operate two electron microscopes in the National Center for Electron Microscopy at Lawrence Berkeley National Laboratory in real time on the Internet. For example, an aluminum-lead alloy can be remotely heated to lead's melting point, allowing observation of phase transformation using a Kratos EM-1500 transmission electron microscope -- which uses the highest accelerating voltage available in the U.S. -- with four angstrom resolution. (One angstrom is one ten-billionth of a meter. Most atoms are approximately two angstroms apart.) This work is helping materials scientists across the country better understand such manufacturing processes as soldering, welding, and brazing. The uniqueness of these demonstrations lies in the ability to perform dynamic realtime experiments with the microscope over the Internet by bringing high performance computing close to the instrument site for online analysis, and by using such analysis for closed-loop control.
Biomedical technology collaboratories	NCRR has initiated eight collaboratory testbeds in NCRR-supported resource centers. These testbeds are designed to take full advantage of emerging Internet capabilities, such as increased bandwidth, security, access, and reliability, while testing technologies including instrument control from remote locations (NanoManipulator, electron and optical microscopes, and National Medical Resource [NMR] spectroscopes), collaborative software design, complete crystallographic experiments over the Internet, remote access to sophisticated visualization systems, and collaborative drug design and protein engineering.
Electronic Notebooks for collaborative environments	DOE is continuing its R&D on Electronic Notebooks used in collaborative environments. These notebooks -- which require interoperability, integration of multiple technologies, and use of heterogeneous computing and communications systems -- aid researchers in recording and sharing experiments that use remote instruments, serving as a medium for collaborative scientific inquiry and engineering design.   Security technology is also necessary to help detect possible tampering with data when notebooks are used as patent records for inventions or proof of conformance to procedures. Future R&D will focus on ease of use, privacy, security, and tools to coordinate access to multiple notebooks.
Unified Medical Language System (UMLS)	With the rapidly growing number of computerized database resources offering bibliographic, full text, and factual data via the Internet, it is often difficult for users to locate and process needed information due to unforgiving access protocols and retrieval languages that differ between databases. In biomedicine, the disparity in biomedical terminology used to describe related concepts in different machine readable files prevents practitioners and researchers from retrieving and integrating biomedical information from sources such as the biomedical literature, clinical records, medical data banks, and expert knowledge bases.   NLM is developing its Unified Medical Language System (UMLS), a long term project to compensate for dissimilar ways that related information is classified in different automated systems. Intelligent-agent-mediated gateways will provide users with a single point of access to information, freeing them from having to master multiple information sources.   NLM researchers are continuing to develop and deploy new capabilities for automatic source selection and for retrieving and sorting information from multiple databases available via NLM's Web site and its Internet Grateful Med (IGM), PubMed, and TOXNET retrieval services. NLM will focus on facilitating access for health professionals, researchers, and the public. FY 2000 plans include R&D in the use of UMLS knowledge sources and integrating multiple multimedia information sources into computer-based patient record systems.
Medline Plus	NLM has developed Medline Plus, an Internet site with information on common diseases and conditions, as well as links to Web sites, medical dictionaries, and reference tools used by medical librarians. Medline Plus developers are now packaging biomedical information and references so they are useful to people without medical degrees or backgrounds. Medline Plus contains information on approximately 50 health topics including AIDS, cancer, diabetes, eating disorders, Parkinson's Diseases, smoking, and tuberculosis. NLM is increasing Medline Plus coverage to some 400 health-related topics and is enlisting the aid of 207 public libraries to help make the site more user-friendly.
The Visible Human project	New computer-based technologies provide unprecedented opportunities to supplement traditional 2-D biological images with dynamic 3-D images that users can view, rotate, and reversibly dissect. As part of its Visible Human project, NLM is building and evaluating digital image libraries of human anatomical structures by integrating advanced computing and communications technologies with medical imaging systems for computer tomography (CT) and magnetic resonance (MR) imaging. NLM is investigating both compression technologies for storing such massive datasets and communications technologies for transmitting them faster over the Internet.     Researchers at Columbia University are generating a "gallery" of 3-D photographic quality color models of anatomical structures obtained from the Visible Human datasets. The 3-D visualizations are fully interactive and can be used and reused for undergraduate and graduate medical teaching, patient education, and in industry. The knowledge base design builds on work done at Columbia using NLM's UMLS. Pictured is a snapshot from a Quicktime movie that displays a rotational 3-D image of the bladder, prostate, and urethra.   NLM's FY 1999 activities in this ongoing project include segmenting and labeling male and female anatomical structures; converting the thorax sections from the Visible Human image file format into a national 3-D image file format; experimenting with interfacing the visualization subsystem with industrial image rendering software; initiating development of a platform-independent radiological atlas for epidemiological and general biomedical research; and developing tools to organize and statistically analyze information from text and X-ray image data-bases.   FY 2000 plans include beta testing the Visible Human anatomical database -- including the 3-D file format -- over the Internet, and completing the development and testing of an online, Java-based, platform-independent radiological atlas for biomedical research.
Universal access	In FY 1999, NSF began a multi-year research effort to help individuals find, manipulate, and use information in an efficient and comprehensive manner. The primary objective of this universal access research is to empower people with disabilities to participate in the emerging information society. The research also aims to benefit the entire Nation by advancing technologies that allow all people to become more productive through better and more effective use of multimodal information resources.   To achieve these goals, NSF plans to develop new models and architectures that emphasize interface usability and speed, define semantic structures for multimedia information to support cross-modal I/O, develop specific solutions to address the special needs of large disabled communities, and conduct experimental studies to evaluate the success of attempts to provide access. Research topics include alternatives to the traditional desktop interface, access to auditory information in textual form for hearing impaired users, new I/O techniques for motor impaired users, and large scale tactile access to graphical information for visually impaired users.   NSF's High Capability Applications for the Individual will address society's need for universal, easy to use access to information resources, powerful methods of presenting information for ease of understanding, and customized "information space" for personal use. Examples include digital libraries and medical information servers. These and other applications will drive the enabling research of computing systems, human centered systems, and networking, communications, and the convergence of computing and communication, and may lead to fundamentally different ways of solving problems. The applications come from the physical and biological sciences, geosciences, social and behavioral sciences, and engineering.
Usability	As more companies and institutions rely on Web sites for electronic commerce, information dissemination, and networked applications, the ability to design usable Web sites is critical for increasing productivity and sales and lowering costs and user frustration. NIST is developing tools to automate and support the design and evaluation of Web sites. These tools help Web designers analyze sites for potential usability problems during the design phase and assist usability professionals in expediting their evaluations. In FY 1999, NIST expanded its automated analysis capabilities based on numerous guidelines and current Web-usability research, improved instrumentation tools, and conducted research on the effects of cultural icons on usability.
Fingerprint and mugshots standards	In the late 1960s, NIST partnered with the FBI to create a computerized approach for automatically comparing and matching fingerprint images. This work has expanded to include developing standards and specifications for capturing and storing fingerprint and mugshot images, as well as creating fingerprint and mugshot image databases for testing fingerprint processing techniques. NIST's R&D resulted in standards and specifications for the quality, format, and transmission of electronic images and related data, such as identifying marks.   NIST has coordinated the development of official American National Standards Institute (ANSI) data and image interoperability standards, and has developed data exchange standards for interoperable systems and evaluation methods for fingerprint systems. NIST maintains the ANSI standard for the data exchange of FBI fingerprint and mugshot data and publishes Standard Reference Material CDs containing over 50,000 fingerprint image pairs and mugshot evaluation data used throughout the industry. NIST has also developed public domain Pattern Level Classification Automation System (PCASYS) and has distributed over 300 copies of this system.   In 1999, NIST began to develop a system in which a smartcard containing an individual's fingerprint and digital signature is used in combination with the individual's realtime scanned fingerprint image to verify authenticity of the creator of the electronic record. The smartcard, together with a registry of legitimate digital signatures and the electronic signing of the transaction, will help ensure that neither the authenticity of the record nor the data were corrupted.
Regional Technology in Education Consortia (RTEC)	In FY 1999, ED is conducting R&D to provide the education community with resources and skills to improve teaching and learning. ED continues to support the Regional Technology in Education Consortia (RTEC), which provide state and local education agencies with technical assistance in developing comprehensive educational technology plans, help integrate technology into the learning and teaching environments, and assist educators in acquiring, maintaining, and using new technologies.
Regional Educational Laboratories	ED's Regional Educational Laboratories assist educators and policy makers in implementing effective school reform strategies by testing new approaches to teaching and learning, providing training and technical assistance, and disseminating information. This is accomplished in part through educational technology services, including telecommunications for rural areas that lack modern technology, and videos and CD-ROMS that highlight exemplary teaching and learning methods.
National Institute on Disability and Rehabilitation Research (NIDRR)	ED's National Institute on Disability and Rehabilitation Research (NIDRR) investigates technologies to improve the lives of people with disabilities. In FY 1999, four of NIDRR's Rehabilitation Engineering Research Centers (RERCs) are focused on HuCS-related R&D: information technology access, communication enhancement, universal telecommunications access, and telerehabilitation.   NIDRR supports ABLEDATA, a database of assistive devices for individuals with disabilities, and the National Rehabilitation Information Center's REHABDATA, a rehabilitation bibliography database. In FY 2000, NIDRR proposes three initiatives in information technology and telecommunications accessibility: a demonstration center to bring together producers, researchers, and consumers; several industry consortia to help build accessibility into new products; and a technology transfer program to explore the use of existing technologies in new applications.
Web Accessibility Initiative (WAI)	The World Wide Web Consortium (W3C) is partnering with organizations around the world to implement the WAI to ensure that new Internet technologies are available to all individuals regardless of physical limitations. Sponsored by NSF, ED's NIDRR, IBM and Lotus Development Corporation, Microsoft, NCR, and the European Commission's TIDE Programme under Directorate Generale XIII, the WAI is coordinating efforts in five areas related to Web accessibility: technology, guidelines, tools, education and outreach, and R&D.   In the technology area, WAI is addressing issues such as HTML coding and style sheets, multimedia, graphics, mobile access, and other data formats and protocols. Researchers are also developing accessibility guidelines for browsers, authoring and content tools, and tools for evaluation, repair, and proxy conversions. The program office facilitates WAI technical activity and coordinates partnerships among stakeholders in Web accessibility including industry, disability organizations, governments, and research organizations.