We have previously described the PACS configuration at Northwestern Memorial Hospital (NMH). As opposed to an imaging modality, PACS is an evolving system that continuously grows and changes to meet the needs of the institution. The NMH PACS has grown significantly in the past year and has undergone significant architectural enhancements. This growth and evolutionary change will be described and discussed. The system now contains over 339,000 studies consisting of over 13 million images. There are now two short-term RAID storage units that provide for twice as much fast storage. There are also two magneto-optical disk jukeboxes providing long-term archive. We have deployed a redundant database to improve reliability of the system in the event of database failure. The number of modalities connected to the system has increased and will be summarized. Statistics describing utilization of the PACS will be shown. Lastly, we will discuss our plans for exploiting the application service provider model in our PACS environment.

Planning PACS for a new hospital facility is an enormous undertaking with many varied user opinions. A pure objective-oriented planning approach was utilized along with a small PACS core planning team. The small PACS core team consisted of owner representation (administration), user representation (radiologists) and the PACS consultant. The overall PACS planning process was broken down into segments. Each segmented represented a key milestone for PACS planning with a set of clear objectives. The PACS core team followed a schedule beginning with preliminary PACS information moving forward to bid specification. An important element in PACS planning was the development of a PACS Component Location Matrix. The matrix was developed from the architectural plans for the new hospital and provided a guideline for everyone including architects and engineers. The matrix included CAD system identification number, location by architectural room number, room name, component network, remote diagnostics and storage. After a year of planning the PACS system and acceptance of vendor bid, the implementation process was ready to begin. The same small PACS core planning team remains intact for the implementation phase of the project.

In 1997, a large PACS was first introduced at Innsbruck University Hospital in the context of a new traumatology centre. In the subsequent years, this initial PACS setting covering only one department was expanded to most of the hospital campus, with currently some 250 viewing stations attached. Constantly connecting new modalities and viewing stations created the demand for several redesigns from the original PACS configuration to cope with the increasing data load. We give an account of these changes necessary to develop a multi hospital PACS and the considerations that lead us there. Issues of personnel for running a large scale PACS are discussed and we give an outlook to the new information systems currently under development for archiving and communication of general medical imaging data and for simple telemedicine networking between several large university hospitals.

The purpose of this study was to determine if the use of a PACS in ultrasound increased the number of images acquired per examination. The hypothesis that such an increase does occur was based on anecdotal information; this study sought to test the hypothesis. There was no significant difference in the total number of images per study for film and PACS.

Five aspects of grayscale calibration of display systems are discussed: (1) As has been shown before, it is very desirable that the discrete luminance levels on the calibrated display function are placed with high accuracy. The accuracy of a calibration with a commercial system is demonstrated. The effect of 8-bit versus 10-bit precision in computing the required luminance levels on contrast and noise in images is illustrated. (2) Display controllers of liquid crystal displays predominantly offer 8-bit precision. A subpixel modulation technique to increase the grayscale accuracy is presented. (3) LCDs may exhibit characteristic curves with non-monotonically changing first derivatives. To achieve a high degree of calibration accuracy, these curves must be represented by a sufficient number of strategically placed measurement points. (4) Photometers of commercial calibration systems for CRT monitors have wide acceptance angles. For display systems with non-uniform angular emission distribution, these photometers may indicate luminance values that do not represent the luminance perceived through the small acceptance angle of the eyes of observers. Techniques for adapting these photometers to luminance measurements for liquid crystal displays are proposed. (5) The suitability of built-in photo-sensors (attached to the tapered glass envelope of the CRT) for calibrating a monitor according to the DICOM Display Function Standard is investigated.

Four radiologists independently reviewed eighty individual thoracic computed radiography images. These consisted of 41 normal radiographs and 39 abnormal radiographs. Images were reviewed using 2,048 by 1,536 pixel monitors which varied widely in image quality as documented using a SMPTE pattern. Overall sensitivity for pathology ranged from .94 to .97 for the radiologists and varied minimally (0.95 to 1.0) for the four monitor quality levels tested while specificity varied to a much greater extent for the radiologists (0.68 to 0.93) and for the various monitors (0.76 to 0.93). The control (good quality) monitor demonstrated substantially higher specificity than the other three monitors (which varied from very poor to fair) but there were no significant differences among the non-control monitors despite their wide variance with regard to image quality. There was no significant difference in sensitivity for the monitors (including the control). Additionally no significant differences were found in the level of diagnostic confidence by the radiologists for the four monitors. These findings suggest that the actual effect of monitor quality degradation on clinical accuracy and confidence is complex and correlates poorly with traditional quality assurance metrics performed for these systems. Additional investigation into the monitor characteristics which most affect clinical confidence and performance is warranted.

The DICOM standard has been used successfully to exchange medical images for some time. However, the important aspect of image display consistency has mostly been disregarded in the past, which led to inconsistent implementations. This situation is changing now since DICOM has defined a set of services which guarantee the consistency of grayscale images for both hardcopy and softcopy. We have developed a test environment that helps to evaluate the conformance of an implementation of these new services. This environment consists of a set of test images and presentation states for softcopy displays as well as a number of test cases for hardcopy devices. The purpose of each test as well as the expected output are documented. Images and presentation states can also be visualized by means of a reference implementation, a DICOM presentation state and an image viewer that also allows to visualize print jobs. In our experience the services for image display consistency are implementable and work very well, however, there are some small details that can be subject to different interpretations. The test environment does not cover all possible cases but definitely helps making an implementation more conformant with the DICOM standard. The test suite is made available on the Internet as Open Source.

This paper retrospectively evaluates our experience with PACS, assessing referring physician and radiologist acceptance and satisfaction, as well as cost justification.

This paper describes the implementation experience in a community hospital setting to provide medical information at the point of decision instantaneously using digital and communication technologies. A unique partnership has been formed between the provider, the Saint John's Health Center, and the vendor to develop, design, and implement a hospital- wide PACS using a multi-phased approach.

The Hong Kong Polytechnic University has a Radiography Division under the Development of Optometry and Radiography. The Division trains both diagnostic and therapeutic radiographers with 60 students/year and offers a B.Sc. degree. In addition the Division together with the University Health Service operates a radiography clinic with radiology consultation from radiologists from other hospitals and clinics. This paper describers the implementation of a PACS in the Division for radiography training, and for clinical service.

Huadong hospital in Shanghai with 700 beds provides health care services for inpatients and outpatients, as well as special senior and VIP patients. In order to move to digital imaging based radiology practice, and also provide better intra-hospital clinical services for senior and VIP patients, we started designing and planning PACS implementation from September of 1999. Based on the radiology service model and current workflow in Huadong hospital, we implemented PACS in three steps.

A single point of hardware failure in PACS is at the PACS controller, or the main archive server. When it occurs, it renders the entire PACS inoperable and crippled until the problem is diagnosed and resolved. Many current PACS do not have a fault-tolerant design or adequate back-up system for the main archive server due to several issues including cost. Several large scale PACs utilize the Tandem or cluster design but are very costly and have not been critically tested for their degree of fault tolerance. This paper describes a novel, portable, and scalable fault-tolerant PACS controller design that is affordable for most PACS implementations.

For its new hospital, UCLA is redesigning the radiologist workplace based on new working paradigm and new technological advances that are becoming available. Given that serious space limitations are being imposed, the challenge is to provide more optimal working environment for radiologists in a crowded and busy working environment.

The Health Insurance Portability and Accountability Act (HIPAA) of 1996 effectively establishes a standard of due care for healthcare information security. One of the challenges of implementing policies, procedures, and practices consistent with HIPAA requirements in the Department of Defense Military Health System is the need for a method that can tailor the requirements to a variety of organizational contexts. This paper will describe a self- directed information security risk evaluation that will enable military healthcare providers to assess their risks and to develop mitigation strategies consistent with HIPAA guidelines.

Most medical images today are generated digitally before exposure on film. In hospitals that employ Picture Archiving and Communication Systems (PACS), the images are also stored and managed digitally. Indeed, film copies of images are still used at large, but the new generation of filmless hospitals tend to minimize the production of films unless deem necessary, or required by the patients or third parties. There are basically two main reasons for working with films in 'filmless' hospitals. One is that in fact, these are 'less film' hospitals due to the film-oriented environment where they operate. Environment which has not yet entered the PACS and DICOM era; Neither in operation, nor in intercommunication. The other reason is that films are needed for legal purposes as a sole indicator to the medical image evidence used during diagnosis. PACS offer numerous advantages, but a high entry cost which can be balanced with the savings in films production and handling. However, as long as films are mandatory, they do not help to lower the inhibitory cost of PACS, and the use of films prevails.

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, 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 a system of templates and matrices created for the purpose, P3WG identified gaps and discrepancies in DoD and service compliance with the proposed Health Insurance Portability and Accountability Act (HIPAA) Security Standard. P3WG represents an unprecedented attempt to coordinate policy review and revision across all military health services and the Office of Health Affairs. This method of policy reform can identify where changes need to be made to integrate health management policy and IT policy in to an organizational policy that will enable compliance with HIPAA standards. The process models how large enterprises may coordinate policy revision and reform across broad organizational and work domains.

The global scale, multiple units, diverse operating scenarios and complex authority structure of the Department of Defense Military Health System (MHS) create social boundaries that tend to reduce communication and collaboration about data security. Under auspices of the Defense Health Information Assurance Program (DHIAP), the Telemedicine and Advanced Technology Research Center (TATRC) is contributing to the MHS's efforts to prepare for and comply with the Health Insurance Portability and Accountability Act (HIPAA) of 1996 through organizational and technological innovations that bridge such boundaries. Building interdisciplinary (clinical, administrative and information technology) medical information security readiness teams (MISRT) at each military treatment facility (MTF) constitutes the heart of this process. DHIAP is equipping and training MISRTs to use new tools including 'OCTAVE', a self-directed risk assessment instrument and 'RIMR', a web-enabled Risk Information Management Resource. DHIAP sponsors an interdisciplinary, triservice workgroup for review and revision of relevant DoD and service policies and participates in formal DoD health information assurance activities. These activities help promote a community of proponents across the MHS supportive of improved health information assurance. The MHS HIPAA-compliance effort teaches important general lessons about organizational reform in large civilian or military enterprises.

We describe a query-by-content search engine that enables a radiologist to search a large database of diagnostically- proven (`benign' or `malignant') mammographic region of interest (ROIs). The database search is facilitated by a relational map which is a 2D display of all the ROIs in the database. Labeled points on the map represent ROIs in the database. The map is constructed from the output of a neural network that has been trained to cluster the ROIs in the database using a measure of perceptual similarity.

In 1998 three hospitals merged to form the Sahlgrenska University Hospital. The total radiology production became 325 000 examinations per year. Two different PACS and RIS with different and incompatible archiving solutions were used since 1996. One PACS had commercial origin and the other was developed inhouse. Together they managed 1/3 of the total production. Due to differences in standard compliance and system architecture the communication was unsatisfactory. In order to improve efficiency, communication and the service level to our customers the situation was evaluated. It was decided to build a transparent virtual radiology department based on a modular approach. A common RIS and a central DICOM image archive as the central nodes in a star configured system were chosen. Web technique was chosen as the solution for distribution of images and reports. The reasons for the decisions as well as the present status of the installation are described and discussed is this paper.

Recent standards and upcoming guidelines are beginning to address the issue of image display quality, and image representation consistency. The DICOM standard has recently been extended to address the consistency of the grayscale presentation of images (DICOM part 14, Grayscale Standard Display Function), and it also addresses the consistency of presentation of associated information (text, graphics, annotation, zoom, rotation) with the recently approved addition of the Softcopy Presentation State, i.e. DICOM Supplement 33). These two new standards provide image consistency between viewing stations and printers independent of the manufacturer, brand and/or type. These services have been demonstrated at ECR, RSNA, and are also part of the continuing IHE demonstrations. In addition to the DICOM standards, work is being done by AAPM Task Group 18 to address the issue of Quality Control and Quality Assurance for electronic displays.

The capacity to evaluate therapies in a multi-center clinical trial for a rare disease like X-linked adrenoleukodystrophy (ALD) can be improved by establishing a network for transmitting magnetic resonance imaging (MRI) data using current Internet and Next Generation Internet (NGI) technologies. A Digital Imaging and Communications in Medicine (DICOM) 3.0 application has been installed; it functions as a storage service class provider (SCP), query/retrieve SCP and central database. Sites with query/retrieve (Q/R) service class user (SCU) applications have access to the images. Using the DICOM Q/R SCU application, timing studies of image retrieval for a standard Internet connection and an NGI connection have been conducted. Standard Internet results indicate that performance is affected by bandwidth limitation and is constrained by network traffic, inhibiting the standard Internet as a useful tool for real-time therapy evaluation. NGI results indicate a higher and more consistent data throughput. The increased transmission speed along with the promise of improved quality of service offered by the NGI connection allows for physicians to discuss the images, correlate them with other disease findings, measure disease severity, and request additional MRI studies in real-time permitting a more efficient clinical evaluation strategy. The elimination of variability of transmission speed is no less significant, allowing the physicians to reserve the time needed for such consultations.

In the last few years more and more University Hospitals as well as private hospitals changed to digital information systems for patient record, diagnostic files and digital images. Not only that patient management becomes easier, it is also very remarkable how clinical research can profit from Picture Archiving and Communication Systems (PACS) and diagnostic databases, especially from image databases. Since images are available on the finger tip, difficulties arise when image data needs to be processed, e.g. segmented, classified or co-registered, which usually demands a lot computational power. Today's clinical environment does support PACS very well, but real image processing is still under-developed. The purpose of this paper is to introduce a parallel cluster of standard distributed systems and its software components and how such a system can be integrated into a hospital environment. To demonstrate the cluster technique we present our clinical experience with the crucial but cost-intensive motion correction of clinical routine and research functional MRI (fMRI) data, as it is processed in our Lab on a daily basis.

We have developed a new mechanism for the delivery of image processing functions to multi-modality PACS diagnostic viewing stations. The tools use the Java programming language. The core visible element of this system is a graphical user interface component that moves around the image(s) like a magnifying glass. Included with the component are controls that are capable of manipulating the image seen within it. The client-server architecture allows for dynamically adding and removing image processing functions as they are developed. In addition to several standard image-processing functions, the component has several novel functions.

With the advent of real-time image streaming, a new paradigm for development of image display and viewing systems that communicate with Picture Archiving and Communication (PACS) systems can be proposed. In this paradigm, the high bandwidth requirements of current systems can be significantly relaxed and security features can be seamlessly adopted and enforced. Based upon this paradigm RealTimeImage and OARG have developed a multi-tiered web-based image display and analysis system for teleradiology. The system architecture consisted of a backend module to communicate with the PACS system via direct file system access or standard DICOM protocols, an Image Server to stream image data to its clients using RealTimeImage Pixel-On-Demand^TM streaming technology and a web-based client to provide image display and analysis functionality. The system was used in a clinical research study that required analysis of several hundred images and included participants located at various remote geographical locations. Performance and maintainability of the system were objectively quantified. Usability issues were subjectively identified by the various users of the system. It was observed that the performance of such a system is comparable to that of today's systems over fast LAN, even if the user is connected via standard, dial-up connections. This level of performance was achieved without compromising the usability of the system required for the research study.

The wide spread of digital technology in the medical field has led to a demand for the high-quality, high-speed, and user-friendly digital image presentation system in the daily medical conferences. To fulfill this demand, we developed a presentation system for radiological and pathological images. It is composed of a super-high-definition (SHD) imaging system, a radiological image database (R-DB), a pathological image database (P-DB), and the network interconnecting these three. The R-DB consists of a 270GB RAID, a database server workstation, and a film digitizer. The P-DB includes an optical microscope, a four-million-pixel digital camera, a 90GB RAID, and a database server workstation. A 100Mbps Ethernet LAN interconnects all the sub-systems. The Web-based system operation software was developed for easy operation. We installed the whole system in NTT East Kanto Hospital to evaluate it in the weekly case conferences. The SHD system could display digital full-color images of 2048 x 2048 pixels on a 28-inch CRT monitor. The doctors evaluated the image quality and size, and found them applicable to the actual medical diagnosis. They also appreciated short image switching time that contributed to smooth presentation. Thus, we confirmed that its characteristics met the requirements.

In a simple, typical radiology workflow process, an order generates a single procedure, which in turn generates a single data set, from which, one radiology report is generated. There are, however, occasions when a single order consists of more than one procedure each with a separate report, yet the procedures are accomplished by one physical acquisition of data. The prototypical example of this is the request for computed tomographic evaluation of the chest, abdomen and pelvis. The study is accomplished, with modern day scanners, by a single helical acquisition, yet there are typically three codable and billable procedures involved, and these may be reported independently either for administrative or academic reasons. This grouping of procedures remained up to now a challenge to automate across integrated modalities, PACS and RIS. This paper discusses a number of other practical cases where this situation occurs and reviews the capabilities of the Presentation of Grouped Procedures IHE Integration Profile in solving this problem. The DICOM services used are evaluated as well as the strengths and weaknesses of this IHE Integration Profile. The implementation experience gained on both a CT and an MR for the IHE Demonstration at RSNA 2000 and HIMSS 2001 is also reviewed. In conclusion, the resulting clinical and operational benefits are discussed.

In a medical center performing an average of 50 upright chest examinations per day, a digital radiography dedicated chest device (General Electric) was compared to a conventional screen-film dedicated chest unit (Picker), in terms of workflow, technologist productivity, and overall speed-of-service from examination ordering to interpretation available. An assessment of ease-of-use and workflow of each device was collected via a technologist opinion survey. Productivity was measured as the rate of patient throughput from normalized timing studies. Measurements were made throughout a typical workday and workweek covering periods of constant as well as spurious activity.

DICOM Structured Reporting (SR) provides for encoding and interchanging structured information that may reference images, waveforms or other composite objects, in traditional reporting applications as well as for logs, measurements and CAD results. DICOM SR differs from generic content encoding approaches like XML, in that it supports coded entries, values that are strongly typed, and explicit relationships. DICOM structured reports (like images and waveforms) are composite objects that can be stored, transmitted and queried. The traditional DICOM binary encoding is used to encode structured reports. The structure and content of the SR tree should be accessible regardless of the internal or external representation. XML parsers and XSL-T tree transformation engines that can be used for data entry, presentation (display and printing) and trans-coding (to HL7 2.x and HL7 Clinical Document Architecture (CDA)) need to be interfaced with DICOM tools that support encoding, transmission, storage and retrieval. Issues associated with establishing the appropriate boundaries between tools are discussed, as are how and when to internalize a DICOM SR in an actual or virtual XML representation, the characteristics of such a representation, and the use of SAX events or the Document Object Model (DOM) to drive style-sheet driven tree transformation engines.

The US Department of Veterans Affairs VistA hospital information system (HIS) provides a complete infrastructure to support multimedia objects as part of its electronic medical record. Acquisition, exportation, storage, long-term archiving, retrieval, display, and clinical database integration capabilities are provided for DICOM objects. We have extended HIS support of DICOM to include endoscopy and ophthalmology. Both radiology and non-radiology studies have the same need for patient and study information to properly identify images, and can use the same DICOM services, but there are some major differences.

The availability of new digital detector technologies and high speed computer processing has led to the development of CAD (computer-aided diagnostic) tools that assist radiologists in detecting and characterizing mammographic lesions. To meet the challenge of developing and implementing algorithms that are computationally intensive, it is desirable to develop reusable components that can execute in a distributed environment. It is well know that the Common Object Request Broker Architecture (CORBA) provides an open solution in distributed computing. We have implemented a hybrid component model consisting of a CORBA server and a Contract Net Protocol (CNP) algorithm for distributing tasks to multiple computers for enhanced processing. Support classes were developed to wrap algorithms developed in C to operate within the distributed framework. CORBA provides communication between agents on different computers and computer platforms and the CNP algorithm is used to select the 'optimal' computer for performing a task. We have evaluated this framework with CAD processing applied to digitized mammograms by transparently scheduling and distributing multiple tasks on three server computers. We achieved a significant reduction in processing times compared to processing on a single computer.

UCLA is in the process of building a new acute-care hospital due to open in 2005 with the intent to operate fully digitally. The strategic planning for this hospital is based on a set of new paradigms: wider and more efficient access to all information sources, enterprise-wide data repository, usage of thin-client technology and wide usage generic information-appliances and wireless devices allowing access to information from anywhere in the hospital. These new paradigms required significant changes from traditional information technology architecture in particular in workflow management of large quantities of imaging data.

When faced with the enormous task of merging several rural and community hospitals with a university hospital, developers were faced with the problem of accessing radiology reports in an electronic radiology practice. Voice recognition systems were installed to handle diagnostic reporting, but not all the facilities had hospital information systems that could receive and disseminate reports. The project goal was to make radiology reports available to clinicians to facilitate treatment and for radiologists to use when comparing previous studies with current ones.

Applications of Picture Archiving and Communication Systems (PACS) in Pathology and Cytology are currently hardly used in the diagnostic process. Here we describe a system that, together with existing equipment, stores data from cervix smears, and aids the physician in the diagnostic process. The system comprises of a scanning system for the specimen and a multi-level storage system partly on disk and partly on optical media. The system adds additional benefits in the diagnostic process while at the same time provides a safe long-term storage and archive of the data. Due to the open nature and the cost-effectiveness of this solution applications of this system can be found not only in cervical cytology but also in pathology or other parts of medicine where image processing or storage is a major issue.

One of the goals of the National Cancer Institute cancer control program to reach more than 80% of the eligible women in mammography screening by the year 2000 has not been fully realized and yet remains as a challenge. That is primarily due to (1) the fact that examination process is a complex and lengthy one, and (2) it is not available to the majority of women who live in remote and urban sites. This problem can be solved using advanced networking technologies and signal processing algorithms.

Clinical evaluation and diagnosis of complex cardiac diseases require to extract, interpret and confront findings from different imaging modalities in a convenient manner for timely and efficient review. Multi-disciplinary conferences between cardiologists, radiologists, surgeons and referring physicians require concise presentations of clinical data and images. A typical presentation requires preparation of summary reports, results from the different investigations and imaging data. In today's settings this is a tedious and time consuming process that requires significant preparation and a long time to present image data from different media such as videotapes, cassettes, plane X-ray films as well as 35 mm cine films.

The purpose of this study was to determine if the interval between an examination being ordered by an Emergency Department physician and his or her review of the report and images could be shortened by notifying the physician that the results were available. Though some time shortening was shown, it was not clearly attributable to the notification process.

As the trend toward consolidation of hospitals continues, problems with merging radiology practices multiply. With no standardization of patient identifiers, increased demand for access to reports and medical records, and in some cases, too few radiologists to cover rural areas, the job of creating a unified system for electronic radiology practice becomes an important, frustrating, and time-consuming proposition. After acquiring several rural facilities and a community hospital, researchers and developers at Shands Medical System, Inc, centered at the University of Florida have worked integrate the various systems and create a unified system for image and report dissemination. Digital imaging equipment was installed at each institution and dedicated network lines were installed between rural locations. Since each hospital assigned medical record numbers, an institutional code was added to identify locations of patients and to assure unique identifiers. As radiology information systems (RIS) and hospital information systems (HIS) were implemented, they were interfaced to the PACS and voice recognition systems. A web-server provided wide access to clinical images and an interface to the voice recognition system provided reports when the HIS was not available or had not yet been installed.

This site has a contrast as a central data collection and image analysis center for a multilayer study involving four acquisition sites. Magnetic Resonance and Ultrasound studies are to be acquired at the sites and then transmitted via the Internet to the data collection center. This paper will describe the software architecture of a workstation designed to act as a store and forward node at a remote site. The software receives and stores images in DICOM format on the local hard drive. The workstation provides several different mechanisms for removing local identifying patient information and inserting patient and study identifiers which are specific to the multicenter study. After removing or modifying header information, the user may enqueue the data for transmission to the central repository.

A major contribution to improving interoperability of medical imaging systems is made by DICOM. This standard focuses on tasks related to the creation of the results of medical imaging procedures and supports facilities corresponding to those available to the recipient of a photographic negative, i.e. some possibility of post processing. End users increasingly expect to have access to medical images but normally require only facilities corresponding to those of the recipient of a photographic print. There is currently no simple format that supports clinical needs and is generally accepted as appropriate for use by end user systems. In this paper we compare the requirements for end user medical image access with those supported by the Basic Image Interchange Format standard (ISO/IEC 12087-5:1998(E)).

To date, the majority of Picture Archival and Communication Systems (PACS) have been utilized only for capture, storage, and display of radiology and in some cases, nuclear medicine images. Medical images for other subspecialty areas are currently stored in local, independent systems, which typically are not accessible throughout the healthcare enterprise and do not communicate with other hospital information or image management systems. It is likely that during the next few years, healthcare centers will expand PAC system capability to incorporate these multimedia data or alternatively, hospital-wide electronic patient record systems will be able to provide this function.

Studies of information and process modeling have demonstrated the importance and clinical impact of Picture Archiving and Communication Systems in the efficient operational management of imaging within the clinical setting. The appropriate identification of both clinical and technology requirements for the planning and deployment of such systems is essential to achieve cost-effectiveness in use. The understanding of the complexities of clinically viable network topologies and architectures for PACS can be achieved through realistic simulations and modeling. The purpose of this paper is to provide a methodology for modeling the DICOM session and application layer entities over Ethernet-based TCP/IP, by using the OPNET Modeler, and derive performance evaluation metrics for different PACS network topologies.

Object-oriented DICOM decoding library was developed as a type of DLL for MS-Windows environment development. It supports all DICOM standard Transfer Syntaxes, multi-frame images, RLE decoding and window level adjusting. Image library for medical application was also developed as a type of DLL and ActiveX Control using proposed DICOM library. It supports display of DICOM image, cine mode and basic manipulations. For an application of a proposed image library, a couple of DICOM viewers were developed. One can be used as an off-line DICOM Workstation, and the other can be used for browsing the local DICOM files.

This paper discusses our initial efforts to design and develop a digital mammography data warehouse to facilitate clinical and research activities. Data warehouse is a complete and consistent integration of data from many information sources. It enables users to explore the warehouse for various analysis and decision support purposes. We are designing an infra-structural information system by incorporating various kinds of breast imaging data, from a diversity of existing clinical systems, into a digital data warehouse. Various types of breast imaging data, including patient demographics, family history, digital mammography and radiological reports, will be acquired for the University of California San Francisco digital mammography PACS modules, as well as Radiological Information System.

There are about 30/mon consultation meetings held in Huadong Hospital in Shanghai for senior and VIP patients. In order to move to digital imaging based radiology practice in the radiology department, and also provide more efficient clinical services, we built a PACS connecting to the DICOM conformance modalities. The specially designed teleconsultation workstations were installed in the radiology department and clinical departments for better consultation services. The images generated from the modalities were sent from PACS server to these workstations following normal PACS image data flow.

Content-based image retrieval (CBIR) provides a flexible means of searching a digital image library based on the description of the desired image. In this paper, we integrate CBIR, RIS, and HIS in PACS to allow retrieving images of similar features. Once the system finds the related images, the embedded CBIR retrieves the radiological reports and medical records of the output images, which can be used to increase diagnostic accuracy. The CBIR system is implemented on a separate server based on multi-resolution image matching. To reduce the retrieval loading on the server and network shanks, a procedure to use copies of images that are temporarily located in some workstations in the PACS is applied. These copies are stored on a temporary database space created on the different workstations. A new image retrieval management server contains image IDs in the database and the IP addresses of the workstations containing temporary image copies. Data on the management server are continuously updated with each addition or retrieval operation. When a display workstation needs a specific image, it sends a request with the required image ID to the management server, which in turn replies with the IP of the workstation containing the inquired image ID.

Although DICOM compliant computed tomography has been prevailing in medical fields nowadays, there are some incompliant ones, from which we could hardly get the raw data and make an apropos interpretation due to the proprietary image format. Under such condition, one usually uses frame grabbers to capture CT images, the results of which could not be freely adjusted by radiologists as the original CT number array could. To alleviate the inflexibility, a new method is presented in this paper to reconstruct the array of CT number from several gray-level images acquired under different window settings. Its feasibility is investigated and a few tips are put forward to correct the errors caused respectively by 'Border Effect' and some hardware problems. The accuracy analysis proves it a good substitution for original CT number array acquisition. And this method has already been successfully used in our newly developing PACS and accepted by the radiologists in clinical use.

Colossal Storage Inc. has patents on new ways of non-contact reading and writing with non destructive reading of information to a ferroelectric molecule. The se methods will be used to develop the worlds first 2D/3D Area/Volume Holographic mass storage device.

We described a web-based data warehousing method for retrieving and analyzing neurological multimedia information. The web-based method supports convenient access, effective search and retrieval of clinical textual and image data, and on-line analysis. To improve the flexibility and efficiency of multimedia information query and analysis, a three-tier, multimedia data warehouse for epilepsy research has been built. The data warehouse integrates clinical multimedia data related to epilepsy from disparate sources and archives them into a well-defined data model.

In this work we present a practical model of telemedicine usage based on a HIS/PACS system that was developed and is being tested at the Heart Institute of Sao Paulo (Brazil). The objective of this project is to allow hospitals that are distant of specialized medical centers to have access to the services of other medical institutions using a low cost telemedicine solution supported by an appropriate architecture of storage and management of medical information. The services that can be accessed using this solution are, for example, second medical opinion, medical images databases and reports of clinical exams. With a simple architecture and easy operation, this project showed to be an efficient way to make a bridge between modern medicine centers and others, localized in places not provided by specialized medical assistance. The system described is still a prototype in experimental phase operating at the Heart Institute of Sao Paulo with good results and will shortly equip other hospitals (auxiliary units of the Heart Institute).

We present a web-based collaborated diagnosis system developed using Java programming language. The system allows more than two physicians to look at the same images, to discuss through a chat box, and to make diagnosis collaboratively. The system provides tools such as window and value setting that physicians generally need. Beside the tools, we are implementing different tools for the system such as a volume rendering algorithm and an animation playback method. To make the system easily accessed, all the physician need is a web browser.

The popularity of digital imaging devices and PACS installations has increased during the last years. Still, images are analyzed and diagnosed using conventional techniques. Our research group begun to study the requirements for digital image diagnostic methods to be applied together with PACS systems. The research was focused on various image analysis procedures (e.g., segmentation, volumetry, 3D visualization, image fusion, anatomic atlas, etc.) that could be useful in medical diagnosis. We have developed Image Analysis software (www.medimag.net) to enable several image-processing applications in medical diagnosis, such as volumetry, multimodal visualization, and 3D visualizations. We have also developed a commercial scalable image archive system (ActaServer, supports DICOM) based on component technology (www.acta.fi), and several telemedicine applications. All the software and systems operate in NT environment and are in clinical use in several hospitals. The analysis software have been applied in clinical work and utilized in numerous patient cases (500 patients). This method has been used in the diagnosis, therapy and follow-up in various diseases of the central nervous system (CNS), respiratory system (RS) and human reproductive system (HRS). In many of these diseases e.g. Systemic Lupus Erythematosus (CNS), nasal airways diseases (RS) and ovarian tumors (HRS), these methods have been used for the first time in clinical work. According to our results, digital diagnosis improves diagnostic capabilities, and together with PACS installations it will become standard tool during the next decade by enabling more accurate diagnosis and patient follow-up.

PACS is a system need for great investment and high technology, its development in China should give consideration to the national conditions of China. We propose the component PACS that rely on the component technology and can be developed synergicly, implement flexibly. We implement PACS by distributed component technology, use Internet browser to construct client-side, and design an open programming interface. Component PACS reduces the total cost in development and application.

In this paper, we present our approach in order to implement a Medical Image Database (MIDB) for archiving mammograms and their related information in the Department of Radiology of the Necker Hospital (Paris). The aim of such a database is to help breast cancer screening in clinics, research and education. As implementation of such a MIDB requires the understanding of users' needs, we have analyzed requirements by using the Crews-l'Ecritoire (Cooperative REquirements With Scenarios) approach developed in our laboratory. This approach is based on the 'Requirement Engineering' concept. It helps understanding users' needs using a semi-automatic analysis of textual scenarios, i.e. scenarios written in natural language. This approach mixes concepts of goals and of scenarios into the notion of 'Requirement Chunk'. Authored scenarios and goal discovery are guided by rules, which lead to a structured network of scenarios. Our analysis results in 58 Requirements Chunks gathering 72 authored scenarios and 300 goals which represent MIDB services requested by radiologists in the course of their daily practice.

Digital radiographs obtained using a prototype Digital Radiography System (Stingray) were compared with those obtained using conventional screen-film. Forty adult volunteers each had two identical radiographs taken at the same level of radiation exposure, one using screen-film and the other the digital detector. Each digital image was processed by hand to ensure that the printed quality was optimal. Ten radiologists compared the diagnostic image quality of the digital images with the corresponding film radiographs using a seven point ranking scheme.

With the development of PACS system more and more medical images are stored in the database, retrieving images in such database becomes more and more difficult. that is why we introduce the way based on image content. In this paper the method of retrieving medical image is presented, which is based on both histogram and correlation algorithm. Histogram algorithm has characteristics of rotation invariance, position invariance and scale invariance. However, the limitation of histogram algorithm is obvious, it can't represent the spatial information of image. Hence, we introduce algorithm of correlation into our method to solve this problem. The method presented in this paper has characteristics of exactness and quickness in retrieving medical images, and which can retrieve image directly without processing image through image segmentation. So it can be realized easily. There was no any relative report about using this method to retrieve images before. Our work shows the method of correlation has a better performance in retrieving medical images, therefore it is a promising approach in this field.

This paper proposes pathological microscopic image compression schemes that suit lossless and progressive transmission. Because pathological microscopic images require very high resolution, they create heavy storage requirements and long transmission times. Image compression is desired to reduce these problems. First, we propose a lossless Karhunen-Loeve Transform (KLT) based on ladder networks. The proposed lossless KLT reduces inter-color redundancies which increases coding performance. Next we propose a progressive transmission algorithm by combining the lossless KLT and set partitioning in hierarchical trees (SPIHT) with the S+P transform. SPIHT is adopted to encode individual color-transformed components. By considering coding efficiency, the transmission bit rates of each encoded component are determined. The resulting algorithm gives high coding performance and has progressive transmission capability. When all transmitted data are decoded, decoding yields the original image. We demonstrate the performance of the proposed algorithm when applied to super high definition pathological microscopic images. All the images used in our tests have 2048x2048 pixels and 24 bits per pixel. It is shown that the coding performance of the proposed algorithm is superior to that of DCT-based JPEG with the RGB/YUV transform.