The IHE (Integrating the Healthcare Enterprise) initiative has recently developed into a multi-national effort that addresses the specific needs of healthcare systems in North America, many European countries as well as Japan. The introduction of IHE in Europe is a particular challenge because the structure of the healthcare sector deviates significantly from country to country, with different legislation, language, medical and organizational procedures and, not the least, different vendors: HIS/RIS systems are typically tailored towards a specific market, and the majority of vendors is only present in one or a few countries. The first multi-national "European Connect-a-thon" was performed in April, 2002 with 57 participating systems from 33 companies, in preparation of four different public demonstrations in three countries. The systems were grouped into a French, German and Italian section (according to the national extensions supported by each system), with certain overlaps, i. e. systems participating in multiple sections in parallel. In conclusion, experiences in Europe show that IHE cannot simply be copied, but has to be adapted to the specificities of national healthcare systems. The good news is, however, that the specifics are only of minor nature compared to the overall complexity of the transactions involved.

The enterprise distribution of radiology images should be integrated into the same application that physicians obtain other clinical information about their patients. Over the past year the Roudebush Veterans Affairs Medical Center has provided enterprise access to radiology images after integrating a commercial web-based image distribution system (Stentor, Brisbane, CA) with the Department of Veterans Affairs internally developed Computerized Patient Record System (CPRS). The application, CPRS, serves as the foundation for the VHA to implement an electronic medical record (EMR). We developed the necessary program communications between the CPRS application and the image distribution application to link the request for a report to a request for the corresponding images. When a physician selects a given radiology report in CPRS the text of the report displays in CPRS and the image application loads the corresponding image study. We found that the requests for film jackets decreased over fifty percent six months after full implementation of the enterprise image distribution system. We have found the integration of the image access and display application into an existing patient information system to be very successful.

DICOM is a success for radiology and cardiology and it is now beginning to be used for other clinical specialties. The US Department of Veterans Affairs has been instrumental in promoting this technological advancement. We have worked with a number of non-radiology imaging vendors over the past several years, encouraging them to support DICOM, providing requirement specifications, validating their implementations, installing their products, and integrating their systems with the VA healthcare enterprise. We require each new non-radiology vendor to support the DICOM Modality Worklist and Storage services, as specified in the IHE Technical Framework, and insist that they perform validation testing with us over the Internet before installing at a VA site. Three years ago we began working with commercial DICOM image acquisition applications in ophthalmology and endoscopy. Today we are interfacing with six vendors in ophthalmology, six in dental, and two in endoscopy. Getting imaging modality vendors to support DICOM is only part of the story, however. We have also developed the capabilities of the VistA hospital information system to properly handle DICOM interfaces to the different clinical specialties. The workflow in the clinical specialties is different than that of radiology, and is much more diverse. We designed the VistA DICOM image acquisition and display interface to use the generic order entry, result entry, result reporting, and appointment scheduling applications of our hospital information system, which are common to other hospital information systems, in order to maintain existing clinical workflow, minimize operational disruptions, simplify training, and win user acceptance. This software is now being field tested with dental and ophthalmology systems at a large number of VA medical centers. We have learned several things from this field test. The DICOM Modality Worklist and Storage services can be successfully used for image acquisition in the clinical specialties, although the specifications for some of the clinical specialty image types need to be enhanced. Special consideration needs to be given to the healthcare provider workflow in order to support DICOM requirements and to minimize change. The clinical specialties handle a large number of different kinds of requests, and imaging procedures may comprise only a small subset, which may need to be isolated out for efficient operation of DICOM Modality Worklist. The clinical specialties will acquire a large volume of images. Our goal is to incorporate all of the patient’s data into the electronic medical record and DICOM is making this easier for everyone. The work involved in extending DICOM to the clinical specialties and integrating them with the hospital information systems continues to be an ongoing and worthwhile challenge.

Korea is one of the leading countries in PACS implementation, and over 15% of all hospitals has been introducing and running in PACS. With the support of the Ministry of Health and Welfare, the National Computerization of Agency and National Cancer Center had a plan to try integration of PACS with a purpose of sharing medical image information. The target hospitals have been selected with over 500 beds, and the distance between hospitals from 40km and to 250km. As the vendors of PACS and HIS that had implemented in target hospitals were different, the 'sharing host' has been developed for the purpose of their integration, which enables communication through DICOM and HL7. In order to monitor the communication among the sharing hosts, the 'sharing center' also has been developed. This project was completed by November 2002. We expected that approximate of 100 doctors including 50 radiologists would use this project, high patient’s satisfaction and the decrease in national insurance fee for test and evaluation period. This project is the first attempt that the government has tried to integrate the independent PACS and HIS. On the model of this project, the government will try to expand it through all nation-wide.

Increasing production and use of digital medical imagery are driving new approaches to information storage and management. Traditional, centralized approaches to image communication, storage and archiving are becoming increasingly expensive to scale and operate with high levels of reliability. Multi-site, geographically-distributed deployments connected by limited-bandwidth networks present further scalability, reliability, and availability challenges. A grid storage architecture built from a distributed network of low cost, off-the-shelf servers (nodes) provides scalable data and metadata storage, processing, and communication without single points of failure. Imaging studies are stored, replicated, cached, managed, and retrieved based on defined rules, and nodes within the grid can acquire studies and respond to queries. Grid nodes transparently load-balance queries, storage/retrieval requests, and replicate data for automated backup and disaster recovery. This approach reduces latency, increases availability, provides near-linear scalability and allows the creation of a geographically distributed medical imaging network infrastructure. This paper presents some key concepts in grid storage and discusses the results of a clinical deployment of a multi-site storage grid for cancer care in the province of British Columbia.

Determining the performance bottleneck of a PACS system is a challenging task. System performance is dependent on several variables such as the workstation, network, servers, type of data, and different loading conditions. This makes planning difficult to ensure the system capacity will deliver fast access to images throughout the enterprise of a hospital even during rush periods. The rules of thumb that most vendors use for the number of workstations per server are based upon heuristic experience and may not apply from institution to institution where usage and infrastructures are different. Rules of thumb can be problematic and usually cannot predict the impact when new technology is introduced like Gigabit Ethernet or distributed architectures. We have developed a Monte Carlo Model in an attempt to develop a more accurate model to predict loading on a system at peak “rush hour” times. The focus of the model was on user metrics of performance such as the latency and throughput of images to their workstation. Analysis demonstrates that “traffic jams” can occur and dissipate in a matter of minutes and be relatively irreproducible to the PACS administrator.

For the replacement of its legacy Picture Archiving and Communication Systems (approx. annual workload of 300,000 procedures), UCLA Medical Center has evaluated and adopted an off-site data-warehousing solution based on an ASP financial with a one-time single payment per study archived. Different financial models for long-term data archive services were compared to the traditional capital/operational costs of on-site digital archives. Total cost of ownership (TCO), including direct and indirect expenses and savings, were compared for each model. Financial parameters were considered: logistic/operational advantages and disadvantages of ASP models versus traditional archiving systems. Our initial analysis demonstrated that the traditional linear ASP business model for data storage was unsuitable for large institutions. The overall cost markedly exceeds the TCO of an in-house archive infrastructure (when support and maintenance costs are included.) We demonstrated, however, that non-linear ASP pricing models can be cost-effective alternatives for large-scale data storage, particularly if they are based on a scalable off-site data-warehousing service and the prices are adapted to the specific size of a given institution. The added value of ASP is that it does not require iterative data migrations from legacy media to new storage media at regular intervals.

Last year we presented a Fault-Tolerant Backup Archive using an Application Service Provider (ASP) model for disaster recovery. The purpose of this paper is to update and provide clinical experiences related towards implementing the ASP model archive solution for short-term backup of clinical PACS image data as well as possible applications other than disaster recovery. The ASP backup archive provides instantaneous, automatic backup of acquired PACS image data and instantaneous recovery of stored PACS image data all at a low operational cost and with little human intervention. This solution can be used for a variety of scheduled and unscheduled downtimes that occur on the main PACS archive. A backup archive server with hierarchical storage was implemented offsite from the main PACS archive location. Clinical data from a hospital PACS is sent to this ASP storage server in parallel to the exams being archived in the main server. Initially, connectivity between the main archive and the ASP storage server is established via a T-1 connection. In the future, other more cost-effective means of connectivity will be researched such as the Internet 2. We have integrated the ASP model backup archive with a clinical PACS at Saint John's Health Center and has been operational for over 6 months. Pitfalls encountered during integration with a live clinical PACS and the impact to clinical workflow will be discussed. In addition, estimations of the cost of establishing such a solution as well as the cost charged to the users will be included. Clinical downtime scenarios, such as a scheduled mandatory downtime and an unscheduled downtime due to a disaster event to the main archive, were simulated and the PACS exams were sent successfully from the offsite ASP storage server back to the hospital PACS in less than 1 day. The ASP backup archive was able to recover PACS image data for comparison studies with no complex operational procedures. Furthermore, no image data loss was encountered during the recovery. During any clinical downtime scenario, the ASP backup archive server can repopulate a clinical PACS quickly with the majority of studies available for comparison during the interim until the main PACS archive is fully recovered.

We summarize research and development for the extraction and distribution of biomedical information from a collection of 17,000 spine x-ray images collected by the second National Health and Nutrition Examination Survey (NHANES II). We present a history of the technical milestones of this work, including the data collection as film, digitization, quality control, archiving technology, database organization, medical expert content evaluation, and Web data distribution. We conclude by presenting our current work in content-based image retrieval (CBIR) to exploit the information content of these images directly by using image processing. We provide an overview and current research results from this CBIR work, which includes: extensive segmentation research, focusing on Active Shape Modeling and Active Contour methods; alternative techniques for shape representation, including invariant moments, simple polygon approximation, and Fourier descriptors; neural network classification of shapes into biomedical categories, such as “anterior osteophytes present/not present”; and the implementation of a prototype CBIR system for the vertebrae that supports hybrid text/image queries using MATLAB and the MySQL relational database system.

The main goal of content based image retrieval is to efficiently retrieve images that are visually similar to a query image. In this paper we will focus on content based image retrieval from large medical databases, outline the problems specific to this area, and describe the recent advances in the field. We will also present some of the more significant results obtained with ASSERT (Automatic Search and Selection Engine with Retrieval Tools), the content based image retrieval system developed in our laboratory.

Content-based image retrieval has been proposed as a viable alternative to text-based image retrieval. Mainstream content-based image retrieval methods, however, do not satisfy the complex demands created by biomedical images. The success of content-based biomedical image retrieval hinges on the consideration of domain-specific image characteristics. We strive for image retrieval with a high-level of content understanding, a high degree of query completion and integral user-computer interaction. Departing from the view that user interaction is mandatory for query completion and that query completion is mandatory for content-understanding, we explore the prospects of incrementally and interactively learning concepts during image retrieval for precise formalization of the user's perception of image information. As concepts closely relate to a user's preferences and subjectivity, and in addition can be considered as generalizations of a query instance, they are expected to allow more accurate and reliable image retrieval. In this paper, we discuss a method for concept-based image retrieval by population-based incremental learning of multi-feature image segmentations using query-by-example and relevance feedback. Image retrieval is demonstrated using digitized vertebral X-ray images from the NHANES II database.

Picture archiving and communication systems (PACS) aim to efficiently provide the radiologists with all images in a suitable quality for diagnosis. Modern standards for digital imaging and communication in medicine (DICOM) comprise alphanumerical descriptions of study, patient, and technical parameters. Currently, this is the only information used to select relevant images within PACS. Since textual descriptions insufficiently describe the great variety of details in medical images, content-based image retrieval (CBIR) is expected to have a strong impact when integrated into PACS. However, existing CBIR approaches usually are limited to a distinct modality, organ, or diagnostic study. In this state-of-the-art report, we present first results implementing a general approach to content-based image retrieval in medical applications (IRMA) and discuss its integration into PACS environments. Usually, a PACS consists of a DICOM image server and several DICOM-compliant workstations, which are used by radiologists for reading the images and reporting the findings. Basic IRMA components are the relational database, the scheduler, and the web server, which all may be installed on the DICOM image server, and the IRMA daemons running on distributed machines, e.g., the radiologists’ workstations. These workstations can also host the web-based front-ends of IRMA applications. Integrating CBIR and PACS, a special focus is put on (a) location and access transparency for data, methods, and experiments, (b) replication transparency for methods in development, (c) concurrency transparency for job processing and feature extraction, (d) system transparency at method implementation time, and (e) job distribution transparency when issuing a query. Transparent integration will have a certain impact on diagnostic quality supporting both evidence-based medicine and case-based reasoning.

Radiotherapy (RT) requires information and images from both diagnostic and treatment equipment. Standards for radiotherapy information have been ratified with seven DICOM-RT objects and their IODs (Information Object Definitions). However, the contents of these objects require the incorporation of the RT workflow in a logical sequence. The first step is to trace the RT workflow. The second step now is to direct all images and related information in their corresponding DICOM-RT objects into a DICOM RT Server and then ultimately to an RT application server. Methods: In our design, the RT DICOM Server was based on a PACS data model. The data model can be translated to web-based technology server and an application server built on top of the Web server for RT. In the process, the contents in each of the DICOM-RT objects were customized for the RT display windows. Results: Six display windows were designed and the data model in the RT application server was developed. The images and related information were grouped into the seven DICOM-RT Objects in the sequence of their procedures, and customized for the seven display windows. This is an important step in organizing the data model in the application server for radiation therapy. Conclusion: Radiation therapy workflow study is a pre-requisite for data model design that can enhance image-based healthcare delivery.

Due to the large volume and density of the medical images data, it is necessary the use of suitable database systems to facilitate their storage and management, interacting with the PACS (Picture Archiving and Communication Systems). This paper presents an architecture designed for acquisition and storage of the extracted data related to medical images, emphasizing the importance of experts in acquisition of consistent data. This work also presents the division of the information contained in the medical images into levels such as: low level, segmentation level, interpretation level, semantic level and related information. The levels work as a basis to the database schema represented by ER (entity relationship). This architecture has been validated by a content-based image retrieval system for Neonatology support.

Many educational courses have been designed for training radiologists and allied healthcare providers to operate PACS workstations. There are yet tools available for educational training of PACS concepts and workflow analysis. We have designed and implemented a RIS/PACS Simulator for this purpose. The RIS/PACS Simulator consists of six key components simulating a typical clinical RIS/PACS: RIS simulator, acquisition modality Simulator (AMS), DICOM gateway, PACS controller (UNIX-based), clinical viewing workstation, and network infrastructure with a 100mbits/sec Ethernet switch connecting to all these components. A generic RIS and a generic DICOM compliant PACS software package are used to simulate normal clinical data flow. Using this simulator, trainees can: 1. Observe clinical RIS/PACS operation, component by component 2. Trace image flow through each component 3. Identify PACS data flow bottle neck 4. Induce failure in a component to observe its impact on the PACS workflow and operation RIS/PACS simulator is a valuable tool for participants to gain knowledge of the complexity of RIS/PACS data flow with hands-on experience. As a stand-alone system, it also becomes a good test bed for evaluation of medical imaging applications without interrupting clinical workflow.

We have developed and evaluated a novel image-matching method for medical images. This method allows the radiologist to search through - in a matter of seconds - large medical databases containing thousands of patients. To illustrate the usefulness of this method in a clinical setting, we have employed this method as a diagnostic support tool for pediatric brain diseases. To this aim, we have assembled a database containing Magnetic Resonance (MR) brain images of 2500 patients between ages 0 and 18 with known brain lesions. As the images are added to the database, they are registered to a global coordinate system. In addition, regions of interests (ROI) are labeled, and sophisticated image processing techniques are used to extract image parameters from the ROIs and from the entire MR image. To perform a clinically realistic search through this database, we have established a training testbed at Childrens Hospital Los Angeles for acquiring MR images from our PACS server of patients with unknown lesions. We have matched these images with the images in the pediatric brain MR database containing known lesions using our image-matching method. An expert pediatric neuroradiologist evaluated the search results. We found that in most cases, our image-matching method is able to retrieve images with relevant diagnostic content, making it highly attractive as a diagnostic support tool.

Many regions face a crucial shortage of radiologists, especially in rural areas. In 2001 the Navajo Area Indian Health Service issued an RFP for teleradiology services. The University of Arizona department of Radiology received the contract and worked with Navajo Area service units to create an effective teleradiology service while overcoming challenges of communications infrastructure and multiple organizational boundaries. Department personnel worked with Navajo Area to design and implement new high-speed communications infrastructure on Navajo lands to support teleradiology services. This deployment was completed in the Spring of 2002. Each Area service unit is essentially an independent organization and maintains separate information about patients. This creates a complex, multi-organizational information environment. The case volume for teleradiology, including three sites other than the Navajo Areas, is at approximately 2,000 cases per month. Teleradiology is a routine part of the work flow at the university and is increasingly becoming integrated into the work flow at the rural sites. We have found teleradiology to be extremely effective in addressing the problems of medically underserved areas. Multi-organizational operation presents challenges for electronic integration requiring collaboration from appropriate clinical and technical personnel. The multi-organizational factor also benefits from an evolutionary approach with gradually increasing integration.

Mass proliferation of IP networks and the maturity of standards has enabled the creation of sophisticated image distribution networks that operate over Intranets, Extranets, Communities of Interest (CoI) and even the public Internet. Unified monitoring, provisioning and management of such systems at the application and protocol levels represent a challenge. This paper presents a web based monitoring and management tool that employs established telecom standards for the creation of an open system that enables proactive management, provisioning and monitoring of image management systems at the enterprise level and across multi-site geographically distributed deployments. Utilizing established standards including ITU-T M.3100, and web technologies such as XML/XSLT, JSP/JSTL, and J2SE, the system allows for seamless device and protocol adaptation between multiple disparate devices. The goal has been to develop a unified interface that provides network topology views, multi-level customizable alerts, real-time fault detection as well as real-time and historical reporting of all monitored resources, including network connectivity, system load, DICOM transactions and storage capacities.

Tele-imaging consultation requires high performance network, which can be fulfilled by Internet2 technology. International partnership allows institutions around the world to join Internet2 outside of North America. Once this partnership has been formed, there are three major issues of connectivity: commercial issue, cost of international link and performance issue. Those issues are worthy to be considered before establishing the International Internet2 connection. Since the international link involves the collaborative efforts from vendors, there exists infrastructure looping inside US, which degrade the performance and raise the cost.

Image workflow in today's Picture Archiving and Communication Systems (PACS) is controlled from fixed Display Workstations (DW) using proprietary control interfaces. A remote access to the Hospital Information System (HIS) and Radiology Information System (RIS) for urgent patient information retrieval does not exist or gradually become available. The lack for remote access and workflow control for HIS and RIS is especially true when it comes to medical images of a PACS on Department or Hospital level. As images become more complex and data sizes expand rapidly with new image techniques like functional MRI, Mammography or routine spiral CT to name a few, the access and manageability becomes an important issue. Long image downloads or incomplete work lists cannot be tolerated in a busy health care environment. In addition, the domain of the PACS is no longer limited to the imaging department and PACS is also being used in the ER and emergency care units. Thus a prompt and secure access and manageability not only by the radiologist, but also from the physician becomes crucial to optimally utilize the PACS in the health care enterprise of the new millennium. The purpose of this paper is to introduce a concept and its implementation of a remote access and workflow control of the PACS combining wireless, Internet and Internet2 technologies. A wireless device, the Personal Digital Assistant (PDA), is used to communicate to a PACS web server that acts as a gateway controlling the commands for which the user has access to the PACS server. The commands implemented for this test-bed are query/retrieve of the patient list and study list including modality, examination, series and image selection and pushing any list items to a selected DW on the PACS network.

A PACS mobile terminal has applications in ward round, emergency room and remote teleradiology consultation. Personal Digital Assistants (PDAs) have the highest mobility and are used for many medical applications. However, their roles are limited in the field of radiology due to small screen size. In this study, we built a wireless PACS terminal using a hand-held tablet-PC. A tablet PC (X-pilot, LEO systems, Taiwan) running the WinCE operating systems was used as our mobile PACS terminal. This device is equipped with 800×600 resolution 10.4 inch TFT monitor. The network connection between the tablet PC and the server was linked via wireless LAN (IEEE 802.11b).

As Military Medical Information Assurance organizations face off with modern pressures to downsize and outsource, they battle with losing knowledgeable people who leave and take with them what they know. This knowledge is increasingly being recognized as an important resource and organizations are now taking steps to manage it. In addition, as the pressures for globalization (Castells, 1998) increase, collaboration and cooperation are becoming more distributed and international. Knowledge sharing in a distributed international environment is becoming an essential part of Knowledge Management. This is a major shortfall in the current approach to capturing and sharing knowledge in Military Medical Information Assurance. This paper addresses this challenge by exploring Risk Information Management Resource (RIMR) as a tool for sharing knowledge using the concept of Communities of Practice. RIMR is based no the framework of sharing and using knowledge. This concept is done through three major components - people, process and technology. The people aspect enables remote collaboration, support communities of practice, reward and recognize knowledge sharing while encouraging storytelling. The process aspect enhances knowledge capture and manages information. While the technology aspect enhance system integration and data mining, it also utilizes intelligent agents and exploits expert systems. These coupled with supporting activities of education and training, technology infrastructure and information security enables effective information assurance collaboration.

Health data security, characterized in terms of data privacy, authenticity, and integrity, is a vital issue when digital images and other patient information are transmitted through public networks in telehealth applications such as teleradiology. Mandates for ensuring health data security have been extensively discussed (for example The Health Insurance Portability and Accountability Act, HIPAA) and health informatics guidelines (such as the DICOM standard) are beginning to focus on issues of data continue to be published by organizing bodies in healthcare; however, there has not been a systematic method developed to ensure data security in medical imaging Because data privacy and authenticity are often managed primarily with firewall and password protection, we have focused our research and development on data integrity. We have developed a systematic method of ensuring medical image data integrity across public networks using the concept of the digital envelope. When a medical image is generated regardless of the modality, three processes are performed: the image signature is obtained, the DICOM image header is encrypted, and a digital envelope is formed by combining the signature and the encrypted header. The envelope is encrypted and embedded in the original image. This assures the security of both the image and the patient ID. The embedded image is encrypted again and transmitted across the network. The reverse process is performed at the receiving site. The result is two digital signatures, one from the original image before transmission, and second from the image after transmission. If the signatures are identical, there has been no alteration of the image. This paper concentrates in the method and evaluation of the digital image envelope.

Confidentiality, integrity verification and access control of medical imagery and associated metadata is critical for the successful deployment of integrated healthcare networks that extend beyond the department level. As medical imagery continues to become widely accessed across multiple administrative domains and geographically distributed locations, image data should be able to travel and be stored on untrusted infrastructure, including public networks and server equipment operated by external entities. Given these challenges associated with protecting large-scale distributed networks, measures must be taken to protect patient identifiable information while guarding against tampering, denial of service attacks, and providing robust audit mechanisms. The proposed framework outlines a series of security practices for the protection of medical images, incorporating Transport Layer Security (TLS), public and secret key cryptography, certificate management and a token based trusted computing base. It outlines measures that can be utilized to protect information stored within databases, online and nearline storage, and during transport over trusted and untrusted networks. In addition, it provides a framework for ensuring end-to-end integrity of image data from acquisition to viewing, and presents a potential solution to the challenges associated with access control across multiple administrative domains and institution user bases.

This paper analyzes the method and training of a self directed risk assessment methodology entitled OCTAVE (Operationally Critical Threat Asset and Vulnerability Evaluation) at over 170 DOD medical treatment facilities. It focuses specifically on how OCTAVE built interdisciplinary, inter-hierarchical consensus and enhanced local capabilities to perform Health Information Assurance. The Risk Assessment Methodology was developed by the Software Engineering Institute at Carnegie Mellon University as part of the Defense Health Information Assurance Program (DHIAP). The basis for its success is the combination of analysis of organizational practices and technological vulnerabilities. Together, these areas address the core implications behind the HIPAA Security Rule and can be used to develop Organizational Protection Strategies and Technological Mitigation Plans. A key component of OCTAVE is the inter-disciplinary composition of the analysis team (Patient Administration, IT staff and Clinician). It is this unique composition of analysis team members, along with organizational and technical analysis of business practices, assets and threats, which enables facilities to create sound and effective security policies. The Risk Assessment is conducted in-house, and therefore the process, results and knowledge remain within the organization, helping to build consensus in an environment of differing organizational and disciplinary perspectives on Health Information Assurance.

Health information management policies usually address the use of paper records with little or no mention of electronic health records. Information Technology (IT) policies often ignore the health care business needs and operational use of the information stored in its systems. Representatives from the Telemedicine & Advanced Technology Research Center (TATRC), TRICARE and Offices of the Surgeon General of each Military Service, collectively referred to as the Policies, Procedures and Practices Work Group (P3WG) examined military policies and regulations relating to computer-based information systems and medical records management. Using an interdisciplinary and interservice QA approach they compared existing military policies with the Health Insurance Portability and Accountability Act (HIPAA) Security Rule to identify gaps and discrepancies. The final report, including a plain English explanation of the individual standards and relevance to the Department of Defense (DoD), a comparative analysis and recommendations, will feed in to the security management process and HIPAA implementation efforts at multiple levels within the DoD. In light of High Reliability Theory, this process models how large enterprises may coordinate policy revision and reform across broad organizational and work domains, building consensus on key policy reforms among military stakeholders across different disciplines, levels of command hierarchy and services.

This project demonstrates use of OCTAVE, an information security risk assessment method, as an approach to the safe design and planning of a teleradiology system. By adopting this approach to project planning, we intended to provide evidence that including information security as an intrinsic component of project planning improves information assurance and that using information assurance as a planning tool produces and improves the general system management plan. Several considerations justify this approach to planning a safe teleradiology system. First, because OCTAVE was designed as a method for retrospectively assessing and proposing enhancements for the security of existing information management systems, it should function well as a guide to prospectively designing and deploying a secure information system such as teleradiology. Second, because OCTAVE provides assessment and planning tools for use primarily by interdisciplinary teams from user organizations, not consultants, it should enhance the ability of such teams at the local level to plan safe information systems. Third, from the perspective of sociological theory, OCTAVE explicitly attempts to enhance organizational conditions identified as necessary to safely manage complex technologies. Approaching information system design from the perspective of information security risk management proactively integrates health information assurance into a project’s core. This contrasts with typical approaches that perceive “security” as a secondary attribute to be “added” after designing the system and with approaches that identify information assurance only with security devices and user training. The perspective of health information assurance embraces so many dimensions of a computerized health information system’s design that one may successfully deploy a method for retrospectively assessing information security risk as a prospective planning tool. From a sociological perspective, this approach enhances the general conditions as well as establishes specific policies and procedures for reliable performance of health information assurance.

Purpose: The purpose of this survey was to identify and characterize available programs for viewing DICOM images on personal computers. Methods: To determine the most commonly used software packages for viewing DICOM images and to seek out less well-known ones, recommendations from colleagues, Internet searches, and searches of databases were carried out. Available software was downloaded and run on the hardware recommended by the developer. Features, such as DICOM information object types, image processing capabilities included, and the ability to export images in other formats were tested and compared. Results: A number of “freeware” and “shareware” programs for viewing DICOM images are available. They range from comprehensive offerings that are very similar to commercial workstation software to simple-to-use programs to open DICOM image files and display the images on personal computers. Surprisingly, some of the more capable software is also “freeware”. Breakthrough work: While no scientific breakthroughs resulted from this work, at least one of the DICOM image software packages was not well known among the author’s colleagues who were familiar with other systems, and this particular software was among the most flexible in terms of exporting images in other forms. Conclusion: DICOM viewing software is readily available at no, or low, cost. These programs vary in ease of use, features, and output capability. The results of this survey, while necessarily a “snapshot” in the fast-moving world of software development, should be useful to those who desire to open DICOM images on personal computers or export images from PACS for use in typical office or educational applications.

During the past several years, the using the Web technology and Web server as a means to access PACS image data are being considered and implemented with different technologies architectures. Here, we presented our design and implementation about developing component-based image display module, and then, gave a method to integrate this image display processing (DP) component into a Web-based image distribution server to enable users using Web browsers to access, view and manipulate PACS images. First, we designed the component software architecture of the image display. Second, we developed a Web-based PACS image distribution server based on component architecture, and integrated the DP components and other three PACS components into the Web architecture. Third, we added the new interface supporting http communication to the DP component by using the WinINet API (application program interface) developed by Microsoft, so that, the DP component can be plug-in to Web browsers to interact with the component-based Web server to display and manipulate DICOM images sent from any PACS archiving server. The performance evaluation on the diagnostic display workstation and the component-based Web server shows that the image distribution and display performance from the Web server to browser clients is similar with that of the image loading and displaying procedure of the diagnostic workstation, as more browser clients accessing to the Web server at same time.

Some institutions have adopted digital media such as CD ROMs to provide patients and referring physicians with results of radiological procedures. These systems require a computer to review the images and lack the supporting explanations and guidance for patients to properly understand the results. We developed a hybrid DVD encoding format that combines DICOM-compliant image storage format with video streams viewable on any consumer DVD players. The addition of the video material allows radiologists to provide explanations, disclaimers and guidance to the patients regarding the results of the study. The diagnostic report is also included on the DVD as a PDF file and as a set of video frames that can be viewed on a DVD player. The native high-resolution images are also included in DICOM format on the DVD and can be accessed by any workstation equipped with a DICOM viewer. The DVD that can be reviewed on any consumer DVD player overcomes the limitation of CD ROMS that require the use of a personal computer. The results of the radiological procedures become more accessible patients that could be reluctant or unable to use a computer. Also, live video clips add a more personalized note and allow radiologists to convey important messages to the patients. This is particularly useful for self-referred screening procedures where results should always be accompanied with education material and explanations of the findings.

We evaluated a telemammography system for reviewing and rating screening mammography in a clinical setting. Three remote sites transmitted 306 exams to a central site. Films were digitized at 50 micron pixel dimensions and compressed at a 50:1 ratio. At the central site images were displayed on a workstation with two high-resolution monitors. Five radiologists reviewed and rated the screens without the availability of prior images or additional information indicating: 1) if additional procedures were needed, 2) which breast was involved, and 3) when appropriate, the recommended additional procedures. During the actual clinical interpretation 13.7% (42 cases) of the patients were recalled for additional procedures. During the retrospective review radiologists 1, 2, 3, 4, and 5 recommended additional procedures for 26.1%, 29.1%, 36.3%, 45.1%, and 54.2% of the cases, respectively. The agreements between the clinical interpretation and radiologists 1, 2, 3, 4, and 5 were 77.8%, 76.1%, 69.0%, 62.7%, and 53.6%, respectively. The exceedingly high percentage of recommended additional procedures using the workstation was attributed to lack of prior images or additional information, the knowledge that case management was not affected, and the observers’ expectation for an enriched case mix.

The purpose of this project is to develop a software system to provide feedback to radiologists and other clinicians from interventional procedures in which they participate. Using Health Level Seven (HL7) traffic between the anatomy/pathology information system and other major information systems, we were able to develop a semi-automatic 'tickler' system that can notify clinicians of pathology results as well as the absence of pathology results after a specified time interval. By using this system, radiologists can get more rapid feedback concerning their interpretations and thereby learn to distinguish false positive from true positive cases. Potentially, fewer patients would fall through the follow-up cracks when using our system versus a paper-based method. The system demonstrates, among other things, how HL7 information can be a powerful tool at an institution when used for purposes for which it was not intentionally designed.

The migration of image data from different generations of legacy archive systems represents a technical challenge and in incremental cost in transitions to newer generations of PACS. UCLA medical center has elected to completely replace the existing PACS infrastructure encompassing several generations of legacy systems by a new commercial system providing enterprise-wide image management and communication. One of the most challenging parts of the project was the migration of large volumes of legacy images into the new system. Planning of the migration required the development of specialized software and hardware, and included different phases of data mediation from existing databases to the new PACS database prior to the migration of the image data. The project plan included a detailed analysis of resources and cost of data migration to optimize the process and minimize the delay of a hybrid operation where the legacy systems need to remain operational. Our analysis and project planning showed that the data migration represents the most critical path in the process of PACS renewal. Careful planning and optimization of the project timeline and resources allocated is critical to minimize the financial impact and the time delays that such migrations can impose on the implementation plan.

In January 2002, the Bavarian Statutory Health Care Administration ("Kassenarztliche Vereinigung Bayerns", KVB) started a recertification programme for quality assurance and quality improvement in mammography reading. All accredited radiologists and gynaecologists are asked to prove their qualification every 1-2 years. The recertification programme requires the physicians to read 50 cases randomly selected from a larger collection of high-quality test cases. The portion of malignant and benign cases corresponds to the requirements of the German National Association of Statutory Health Insurance Physicians ("Kassenarztliche Bundesvereinigung", KBV). In order to read the mammograms on a softcopy device the images are digitised with a high-quality scanner and converted to DICOM Digital Mammography format. The workstation software has been implemented according to the particular requirements of this programme. To verify the applicability of digitised mammograms for recertification purposes, a comparative study with 32 trained radiologists and gynaecologists has been performed. As a result the study showed that there was no significant difference in the error rate of the reported findings between conventional film and softcopy reading. The first intermediate results of this quality initiative are promising. The introduction of a corresponding federal German recertification programme is intended.

We developed a multimedia presentation platform that allows retrieving data from any digital and analog modalities and to prepare a script of a clinical presentation in an XML format. This system was designed for cardiac multi-disciplinary conferences involving different cardiology specialists as well as cardiovascular surgeons. A typical presentation requires preparation of summary reports of data obtained from the different investigations and imaging techniques. An XML-based scripting methodology was developed to allow for preparation of clinical presentations. The image display program uses the generated script for the sequential presentation of different images that are displayed on pre-determined presentation settings. The ability to prepare and present clinical conferences electronically is more efficient and less time consuming than conventional settings using analog and digital documents, films and videotapes. The script of a given presentation can further be saved as part of the patient record for subsequent review of the documents and images that supported a given medical or therapeutic decision. This also constitutes a perfect documentation method for surgeons and physicians responsible of therapeutic procedures that were decided upon during the clinical conference. It allows them to review the relevant data that supported a given therapeutic decision.

This paper justifies and explains current efforts in the Military Health System (MHS) to enhance information assurance in light of the sociological debate between “Normal Accident” (NAT) and “High Reliability” (HRT) theorists. NAT argues that complex systems such as enterprise health information systems display multiple, interdependent interactions among diverse parts that potentially manifest unfamiliar, unplanned, or unexpected sequences that operators may not perceive or immediately understand, especially during emergencies. If the system functions rapidly with few breaks in time, space or process development, the effects of single failures ramify before operators understand or gain control of the incident thus producing catastrophic accidents. HRT counters that organizations with strong leadership support, continuous training, redundant safety features and “cultures of high reliability” contain the effects of component failures even in complex, tightly coupled systems. Building highly integrated, enterprise-wide computerized health information management systems risks creating the conditions for catastrophic breaches of data security as argued by NAT. The data security regulations of the Health Insurance Portability and Accountability Act of 1996 (HIPAA) implicitly depend on the premises of High Reliability Theorists. Limitations in HRT thus have implications for both safe program design and compliance efforts. MHS and other health care organizations should consider both NAT and HRT when designing and deploying enterprise-wide computerized health information systems.

This project is intended to capture and interactively display non-image information routinely generated by imaging modalities. This information relates to the device's performance of the individual procedures and is not necessarily available in other information streams such as DICOM headers. While originally intended for use in servicing the modalities, this information can also be presented to radiologists and administrators within the department for both micro- and macro-management purposes. This data can help hospital administrators and radiologists manage available resources and discover clues to indicate what modifications in hospital operations might significantly improve its ability to provide efficient patient care. Data is collected from a departmental CT scanner. The data consists of a running record of exams followed by a list of processing records logged over a 24-hour period. MineScan extracts information from these records and stores it into a database. A statistical program is run once a day to collect relevant metrics. MineScan can be accessed via a Web browser or through an advanced prototype PACS workstation. This information, if provided in real-time, can be used to manage operations in a busy department. Even when provided historically, the data can be used to assess current activity, analyze trends and plan future operations.

A new intelligent software tool for PACS systems called 'iScout' has been developed that constructs and displays an overview of large series or studies before downloading the set of images to a PACS workstation. The overview consists of two orthogonal cross-section images that allow the user to select and download a subset of images, avoiding long delays that can occur while downloading hundreds or even thousands of images. The iScout also provides a navigational tool, allowing the user to click on anatomical regions and view the relevant slices, while displaying the anatomical location of the image currently being displayed by the PACS workstation software. The construction of an iScout can be done on either a workstation or a server with only minimal overhead that does not significantly affect the speed of loading. A working iScout tool has been integrated with multi-modality PACS workstation software (McKesson Medical Imaging Solutions), and it was found that the iScout can be generated on the workstation with a maximum added overhead of only 3.4 seconds while downloading a study containing 433 512x512 CT images. The iScout is flexible and can generate scouts for virtually all types of CT and MR images, as well as 3D Ultrasound.

The management of medical images and their metadata is complicated by the continued use of different file formats (e.g., DICOM, ANALYZE, MINC) as well as the use of different conventions within the same standard (e.g., different DICOM metadata values assigned by different scanners). The conversion between image file formats often involves a complicated and non-unique value mapping that is usually influenced by the specific needs of users who interpret the meaning of each metadata value. We developed a general translation engine that (1) decodes common medical image file formats and organizes their metadata into trees, (2) determines the source that produced the image data using values in the metadata trees, (3) regroups image data from the same source, (4) maps values from a source metadata tree to a designated target metadata tree, and (5) encodes the image data from the target metadata tree into its corresponding file format. We also developed a graphical user interface that allows users to construct a visual representation of the mappings. The resulting application (the Debabeler) is a flexible tool for inspecting the contents of medical image files and defining user-specific translations between image file formats and conventions.

The paper provides a description of methodologies and techniques required for a Training System Development in the field of Senology (TSDS), based n the exploitation of senologic images (primarily mammograms but also echographic images or MRI) and their related clinical files. The aim of such a system is to help breast cancer screening in education. This system will help assist junior radiologists in routine critical use. Development of such a TSDS requires understanding of users’ needs (expertise and pedagogy), model design and system implementation. Specifications have been derived from the experience of the senologists from the Department of Radiology of the Necker Hospital (Paris, France), Department where the training system will be implemented. To be compliant with commercial systems for digital and CAD mammograms, terminological systems used by the TSDS to describe and index data must be based on DICOM and BI-RADS dictionaries. A detailed discussion of the choice of such a method and technique is provided and their respective contribution is described.

We propose a lossless watermarking scheme, in a sense that the original image can be exactly recovered from the watermarked one, with the purpose of verifying the integrity and authenticity of medical images in PACS. Our embedding method includes imposing invertible modulations that induce low distortion level, as well as losslessly compressing the status of the original image and carrying such information in the watermarked image. Digital signature of the whole image is embedded for verifying the integrity of the image. An identifier also presented in electronic patient record (EPR) is embedded for verifying the authenticity by simultaneously processing of the watermarked image and the EPR. Combining with fingerprint system, patient’s fingerprint information is further embedded into several image slices and then extracted for verifying the authenticity. No visual quality degradation is detected in the watermarked image.

Our goal was to develop an inexpensive, high-quality, multi-site telemammography system, implemented with low-level data connections that provided a communication link for an “almost real-time” response from a radiologist (central site) to remote “underserved” sites. The remote sites digitize mammographic films using high-resolution, laser digitizers. Images are automatically cropped, compressed (wavelet-based), and encrypted prior to transmission. At the central site images are decrypted, decompressed, unsharp masked, and displayed using automatically determined LUTs. The sites communicate instantly via a “chat box.” Remote sites 1, 2, and 3 are 15, 20, and 90 miles from the central site, respectively, and connected by POTS (sites 1 and 2) and LAN (site 3). Only minimal noticeable difference at compression levels of 50:1 and 75:1 could be identified unless magnified to extreme levels. Two experienced observers rated the LUTs for 200 images as “acceptable” to “excellent.” Average cycle times to digitize, transmit and receive cases (four films each) at 75:1 compression were 5.97, 6.85, and 5.77 min/case from sites 1, 2, and 3, respectively. Unique data-handling schemes significantly decrease the image file size and allow successful transmission in a reliable, timely manner. Over 1000 cases have been transmitted to date. Messaging was found to be easy to use.

An innovative DICOM meta information database has been developed to support PACS archive servers. Contrary to mainstream designs using a relational database with one table for each information level (namely: patient, study, series, image), we present a boolean-based design that supports the same information model with one table. In this table, the data are stored with the DICOM tags and auxiliary indices required for the information model. Each representative value in a DICOM data set is stored as a record in the table. When a new SOP Class is added, it is unnecessary to rebuild the entire database because no new columns are needed in the table. The new tag, i.e. new information, is added into the table as an additional record. With our current implementation of this model, the same query is typically almost as fast as the 4-table design. In tests using simulated data with a server holding 10,000 patients, each with 2 studies and 2 series of images, it takes less than 6 seconds to query all the patient names starting with “B” on our new database. It takes 5 seconds on a comparable server using the 4 table design. Searching for a value at image level requires a similar amount of time, depending on the amount of data returned.

In this paper, we propose a block-based conditional entropy coding scheme for medical image compression using the 2-D integer Haar wavelet transform. The main motivation to pursue conditional entropy coding is that the first-order conditional entropy is always theoretically lesser than the first and second-order entropies. We propose a sub-optimal scan order and an optimum block size to perform conditional entropy coding for various modalities. We also propose that a similar scheme can be used to obtain a sub-optimal scan order and an optimum block size for other wavelets. The proposed approach is motivated by a desire to perform better than JPEG2000 in terms of compression ratio. We hint towards developing a block-based conditional entropy coder, which has the potential to perform better than JPEG2000. Though we don't indicate a method to achieve the first-order conditional entropy coder, the use of conditional adaptive arithmetic coder would achieve arbitrarily close to the theoretical conditional entropy. All the results in this paper are based on the medical image data set of various bit-depths and various modalities.

An important telemedicine application is the perusal of CT scans (digital format) from a central server housed in a healthcare enterprise across a bandwidth constrained network by radiologists situated at remote locations for medical diagnostic purposes. It is generally expected that a viewing station respond to an image request by displaying the image within 1-2 seconds. Owing to limited bandwidth, it may not be possible to deliver the complete image in such a short period of time with traditional techniques. In this paper, we investigate progressive image delivery solutions by using JPEG 2000. An estimate of the time taken in different network bandwidths is performed to compare their relative merits. We further make use of the fact that most medical images are 12-16 bits, but would ultimately be converted to an 8-bit image via windowing for display on the monitor. We propose a windowing progressive RoI technique to exploit this and investigate JPEG 2000 RoI based compression after applying a favorite or a default window setting on the original image. Subsequent requests for different RoIs and window settings would then be processed at the server. For the windowing progressive RoI mode, we report a 50% reduction in transmission time.

In this paper, we presented a new security approach to provide security measures and features in both healthcare information systems (PACS, RIS/HIS), and electronic patient record (EPR). We introduced two security components, certificate authoring (CA) system and patient record digital signature management (DSPR) system, as well as electronic envelope technology, into the current hospital healthcare information infrastructure to provide security measures and functions such as confidential or privacy, authenticity, integrity, reliability, non-repudiation, and authentication for in-house healthcare information systems daily operating, and EPR exchanging among the hospitals or healthcare administration levels, and the DSPR component manages the all the digital signatures of patient medical records signed through using an-symmetry key encryption technologies. The electronic envelopes used for EPR exchanging are created based on the information of signers, digital signatures, and identifications of patient records stored in CAS and DSMS, as well as the destinations and the remote users. The CAS and DSMS were developed and integrated into a RIS-integrated PACS, and the integration of these new security components is seamless and painless. The electronic envelopes designed for EPR were used successfully in multimedia data transmission.

The purpose of this study is to investigate and apply a new, intuitive and space-conscious visualization framework to facilitate efficient data presentation and exploration of large-scale data warehouses. We have implemented the DBMap framework for the UCSF Brain Research Registry. Such a novel utility would facilitate medical specialists and clinical researchers in better exploring and evaluating a number of attributes organized in the brain research registry. The current UCSF Brain Research Registry consists of a federation of disease-oriented database modules, including Epilepsy, Brain Tumor, Intracerebral Hemorrphage, and CJD (Creuzfeld-Jacob disease). These database modules organize large volumes of imaging and non-imaging data to support Web-based clinical research. While the data warehouse supports general information retrieval and analysis, there lacks an effective way to visualize and present the voluminous and complex data stored. This study investigates whether the TreeMap algorithm can be adapted to display and navigate categorical biomedical data warehouse or registry. TreeMap is a space constrained graphical representation of large hierarchical data sets, mapped to a matrix of rectangles, whose size and color represent interested database fields. It allows the display of a large amount of numerical and categorical information in limited real estate of computer screen with an intuitive user interface. The paper will describe, DBMap, the proposed new data visualization framework for large biomedical databases. Built upon XML, Java and JDBC technologies, the prototype system includes a set of software modules that reside in the application server tier and provide interface to backend database tier and front-end Web tier of the brain registry.

The development of multislice helical CT scanners that can image the lungs in a short time has created interest in the use of mass screening for lung cancer. In the mass screening process, helical CT images are acquired for the entire lung area. Therefore, mass screening based on helical CT images requires a considerable number of images to be read. It is this time-consuming step that makes the use of helical CT for mass screening impractical at present. To overcome this problem, we have developed a newly PACS for a computer-aided diagnosis (CAD) system to automatically detect suspicious regions in chest CT images that can store the huge amounts of data obtained in multi-slice CT studies and permit image comparison without films, and have developed a newly mobile CT system for CAD system. This newly developed CAD system is provided with a comparison review support function employing an automatic slice-matching algorithm that can calculate the difference in slice number between the present and past CT images. This makes it much easier to read images, since the corresponding CT slices from the present and past studies can be displayed simultaneously on the CRT monitor screen. The experimental results indicate that the CAD system can be effectively used in clinical practice to increase the speed and accuracy of routine diagnosis.

The transmission and storage became a major dilemma due to the large size of medical imaging data. Clinical CT scanners typically generate sequential data in each patient. Because there are great correlations between continuous slices in these data, we could investigate a significant breakthrough in data compression applying motion estimation and compensation. With motion vector and motion compensated frame difference, we propose a new approach to encode a compressed sequential stream into DICOM image data structure and it may also be possible to efficiently redesign a multiframe header approach, which could improve compression ratios due to the avoidance of duplicative header information of each DICOM file in the same patient. Motion estimation could be retrieved by block matching approaches on 8x8 blocks of image information. For each 8x8 block, a single motion vector was found by minimizing the mean-square error. The compensation of the estimated motion was presented to smooth out the artifacts along block boundaries. In result, we were able to show the higher compression ratios relative to the output file size and effectively transfer a series of CT images at lower bit rates.

The implementation of the Object Management Group Healthcare Domain Task Force specification is very complex and time consuming. Most healthcare institutions lack professionals with the necessary time, knowledge or experience to understand and adapt the specification to the needs of the institution. In this paper we propose a framework that facilitates the development of Clinical Image Access Service (CIAS) and Clinical Observation Access Service (COAS) using Java. The framework hides the inherent complexity of ORB initialization, object reference and object persistence, allowing the implementers to focus on the access to the information that will be made available by the HDTF services. The framework is based on a delegation scheme for the main components of the CIAS and COAS specification. The framework handles tasks such as object initialization, registration, activation and deactivation which, otherwise, would be responsibility of the implementers. The framework has been implemented using JacORB 1.4 although it can support any ORB. The framework is being used to integrate different systems at the Heart Institute of Sao Paulo. The use of this framework can ease the implementation and integration of distributed information systems. It will improve the quality of healthcare services and allow interoperability between healthcare providers by the use of OMG growing standard in medicine.

Modern communication standards such as Digital Imaging and Communication in Medicine (DICOM) include non-image data for a standardized description of study, patient, or technical parameters. However, these tags are rather roughly structured, ambiguous, and often optional. In this paper, we present a mono-hierarchical multi-axial classification code for medical images and emphasize its advantages for content-based image retrieval in medical applications (IRMA). Our so called IRMA coding system consists of four axes with three to four positions, each in {0,...9,a,...,z}, where "0" denotes "unspecified" to determine the end of a path along an axis. In particular, the technical code (T) describes the imaging modality; the directional code (D) models body orientations; the anatomical code (A) refers to the body region examined; and the biological code (B) describes the biological system examined. Hence, the entire code results in a character string of not more than 13 characters (IRMA: TTTT - DDD - AAA - BBB). The code can be easily extended by introducing characters in certain code positions, e.g., if new modalities are introduced. In contrast to other approaches, mixtures of one- and two-literal code positions are avoided which simplifies automatic code processing. Furthermore, the IRMA code obviates ambiguities resulting from overlapping code elements within the same level. Although this code was originally designed to be used in the IRMA project, other use of it is welcome.

This case study details the experience of system engineers of the Imaging Science and Information Systems Center, Georgetown University Medical Center (ISIS) and radiologists from the department of Radiology in the implementation of a new Teleradiology system. The Teleradiology system enables radiologists to view medical images from remote sites under those circumstances where a resident radiologist needs assistance in evaluating the images after hours and during weekends; it also enables clinicians access to patients’ medical images from different workstations within the hospital. The Implementation of the Teleradiology project was preceded by an evaluation phase to perform testing, gather users feedback using a web site and collect information that helped eliminate system bugs, complete recommendations regarding minimum hardware configuration and bandwidth and enhance system’s functions, this phase included a survey-based system assessment of computer configurations, Internet connections, problem identification, and recommendations for improvement, and a testing period with 2 radiologists and ISIS engineers; The second phase was designed to launch the system and make it available to all attending radiologists in the department. To accomplish the first phase of the project a web site was designed and ASP pages were created to enable users to securely logon and enter feedback and recommendations into an SQL database. This efficient, accurate data flow alleviated networking, software and hardware problems. Corrective recommendations were immediately forwarded to the software vendor. The vendor responded with software updates that better met the needs of the radiologists. The ISIS Center completed recommendations for minimum hardware and bandwidth requirements. This experience illustrates that the approach used in collecting the data and facilitating the teamwork between the system engineers and radiologists was instrumental in the project’s success. Major problems with the Teleradiology system were discovered and remedied early by linking the actual practice experience of the physicians to the system improvements.