The current diagnostic process at hospitals is mainly based on reviewing and comparing images coming from multiple time points and modalities in order to monitor disease progression over a period of time. However, for ambiguous cases the radiologist deeply relies on reference literature or second opinion. Although there is a vast amount of acquired images stored in PACS systems which could be reused for decision support, these data sets suffer from weak search capabilities. Thus, we present a search methodology which enables the physician to fulfill intelligent search scenarios on medical image databases combining ontology-based semantic and appearance-based similarity search. It enabled the elimination of 12% of the top ten hits which would arise without taking the semantic context into account.

In the current health-care environment, the time available for physicians to browse patients' scans is shrinking due to the rapid increase in the sheer number of images. This is further aggravated by mounting pressure to become more productive in the face of decreasing reimbursement. Hence, there is an urgent need to deliver technology which enables faster and effortless navigation through sub-volume image visualizations. Annotating image regions with semantic labels such as those derived from the RADLEX ontology can vastly enhance image navigation and sub-volume visualization. This paper uses random regression forests for efficient, automatic detection and localization of anatomical structures within DICOM 3D CT scans. A regression forest is a collection of decision trees which are trained to achieve direct mapping from voxels to organ location and size in a single pass. This paper focuses on comparing automated labeling with expert-annotated ground-truth results on a database of 50 highly variable CT scans. Initial investigations show that regression forest derived localization errors are smaller and more robust than those achieved by state-of-the-art global registration approaches. The simplicity of the algorithm's context-rich visual features yield typical runtimes of less than 10 seconds for a 512³ voxel DICOM CT series on a single-threaded, single-core machine running multiple trees; each tree taking less than a second. Furthermore, qualitative evaluation demonstrates that using the detected organs' locations as index into the image volume improves the efficiency of the navigational workflow in all the CT studies.

When the first quarter of 2010 Department of Radiology statistics were provided to the Section Chiefs, the authors (SH, BC) were alarmed to discover that Ultrasound showed a decrease of 2.5 percent in billed examinations. This seemed to be in direct contradistinction to the experience of the ultrasound faculty members and sonographers. Their experience was that they were far busier than during the same quarter of 2009. The one exception that all acknowledged was the month of February, 2010 when several major winter storms resulted in a much decreased Hospital admission and Emergency Department visit rate. Since these statistics in part help establish priorities for capital budget items, professional and technical staffing levels, and levels of incentive salary, they are taken very seriously. The availability of a desktop, Web-based RIS database search tool developed by two of the authors (WK, WB) and built-in database functions of the ultrasound miniPACS, made it possible for us very rapidly to develop and test hypotheses for why the number of billable examinations was declining in the face of what experience told the authors was an increasing number of examinations being performed. Within a short time, we identified the major cause as errors on the part of the company retained to verify billable Current Procedural Terminology (CPT) codes against ultrasound reports. This information is being used going forward to recover unbilled examinations and take measures to reduce or eliminate the types of coding errors that resulted in the problem.

Fractures are common injuries, some complicated fractures may require a surgical intervention. When such an operation is planned it can be beneficial to have access to similar past cases including follow ups to compare, which method might be the most adapted one in a particular situation. At the orthopaedic service of the University hospitals of Geneva a database of past cases including pre- and post-operative images and case descriptions has been created over the past years with the goal to support clinical decision making. Images play an important role in the decision making process and the judgment of a fracture, but visual image content is currently not directly accessible for search. At the moment, search is mainly via a classification system of the fractures or in the patient record itself only by patient ID. In this paper we propose a solution that combines visual information from several images in a case to calculate similarity between cases and allow thus an access to visually similar cases. Such a system can complement the text- or classification-based search that has been used so far. In a preliminary study, we used pixel-grid-based salient-point features to build a first prototype of case-based visual retrieval of fracture cases. Cases belonging to different fracture classes were beforehand often confused due to the similar bone structures in the various images. In this article, a multi-scale approach is used in order to perform similarity measures at both large and small scales. When compared to the first prototype, the introduction of scale and spatial information allowed improving the performance of the system. Cases containing similar bone structures but with dissimilar fractures are generally ranked lower whereas more relevant cases are returned. The system can thus be expected to perform sufficiently well for use in clinical practice and particularly for teaching.

In clinical decision processes, relevant scientific publications and their associated medical images can provide valuable and insightful information. However, effectively searching through both text and image data is a difficult and arduous task. More specifically in the area of image search, finding similar images (or regions within images) poses another significant hurdle for effective knowledge dissemination. Thus, we propose a method using local regions within images to perform and refine medical image retrieval. In our first example, we define and extract large, characteristic regions within an image, and then show how to use these regions to match a query image to similar content. In our second example, we enable the formulation of a mixed query based upon text, image, and region information, to better represent the end user's search intentions. Given our new framework for region-based queries, we present an improved set of similar search results.

Articles in the literature routinely describe advances in Content Based Image Retrieval (CBIR) and its potential for improving clinical practice, biomedical research and education. Several systems have been developed to address particular needs, however, surprisingly few are found to be in routine practical use. Our collaboration with the National Cancer Institute (NCI) has identified a need to develop tools to annotate and search a collection of over 100,000 cervigrams and related, anonymized patient data. One such tool developed for a projected need for retrieving similar patient images is the prototype CBIR system, called CervigramFinder, which retrieves images based on the visual similarity of particular regions on the cervix. In this article we report the outcomes from a usability study conducted at a primary meeting of practicing experts. We used the study to not only evaluate the system for software errors and ease of use, but also to explore its "user readiness", and to identify obstacles that hamper practical use of such systems, in general. Overall, the participants in the study found the technology interesting and bearing great potential; however, several challenges need to be addressed before the technology can be adopted.

Multiple sclerosis (MS) is a debilitating autoimmune disease of the central nervous system that damages axonal pathways through inflammation and demyelination. In order to address the need for a centralized application to manage and study MS patients, the MS e-Folder - a web-based, disease-specific electronic medical record system - was developed. The e-Folder has a PHP and MySQL based graphical user interface (GUI) that can serve as both a tool for clinician decision support and a data mining tool for researchers. This web-based GUI gives the e-Folder a user friendly interface that can be securely accessed through the internet and requires minimal software installation on the client side. The e-Folder GUI displays and queries patient medical records--including demographic data, social history, past medical history, and past MS history. In addition, DICOM format imaging data, and computer aided detection (CAD) results from a lesion load algorithm are also displayed. The GUI interface is dynamic and allows manipulation of the DICOM images, such as zoom, pan, and scrolling, and the ability to rotate 3D images. Given the complexity of clinical management and the need to bolster research in MS, the MS e-Folder system will improve patient care and provide MS researchers with a function-rich patient data hub.

Indeterminate incidental findings pose a challenge to both the radiologist and the ordering physician as their imaging appearance is potentially harmful but their clinical significance and optimal management is unknown. We seek to determine if it is possible to automate detection of adrenal nodules, an indeterminate incidental finding, on imaging examinations at our institution. Using PRESTO (Pathology-Radiology Enterprise Search tool), a newly developed search engine at our institution that mines dictated radiology reports, we searched for phrases used by attendings to describe incidental adrenal findings. Using these phrases as a guide, we designed a query that can be used with the PRESTO index. The results were refined using a modified version of NegEx to eliminate query terms that have been negated within the report text. In order to validate these findings we used an online random date generator to select two random weeks. We queried our RIS database for all reports created on those dates and manually reviewed each report to check for adrenal incidental findings. This survey produced a ground- truth dataset of reports citing adrenal incidental findings against which to compare query performance. We further reviewed the false positives and negatives identified by our validation study, in an attempt to improve the performance query. This algorithm is an important step towards automating the detection of incidental adrenal nodules on cross sectional imaging at our institution. Subsequently, this query can be combined with electronic medical record data searches to determine the clinical significance of these findings through resultant follow-up.

With the current emphasis on healthcare reform and cost effectiveness, methods to increase healthcare efficiency while improving outcomes are paramount. With reference to breast cancer, delay in diagnosis can cause significant morbidity and mortality, as well as increased long term health care costs. Assessment with short interval mammographic follow-up of BI-RADS category 3 lesions has been shown to increase detection of a small number of breast cancers at an early stage. Because of the importance of timely follow-up for these patients, we propose a novel computer application that identifies patients due for short-term mammographic follow-up, thus reducing costly hours spent by personnel, reducing human error, and improving patient compliance. Our web-based application mines radiology reports and scheduling information to generate lists of patients due for short-term mammographic follow-up of BI-RADS category 3 results. The results can be placed in a worklist that can be used by a staff member to contact patients to schedule follow-up appointments. Additional analytic features of the application can identify referral characteristics that may serve as potential sources for improvement of patient follow-up. We believe that an automated system can be designed to improve patient care and compliance with follow-up of BI-RADS category 3 results.

Medical image based biomarkers are being established for therapeutic cancer clinical trials, where image assessment is among the essential tasks. Large scale image assessment is often performed by a large group of experts by retrieving images from a centralized image repository to workstations to markup and annotate images. In such environment, it is critical to provide a high performance image management system that supports efficient concurrent image retrievals in a distributed environment. There are several major challenges: high throughput of large scale image data over the Internet from the server for multiple concurrent client users, efficient communication protocols for transporting data, and effective management of versioning of data for audit trails. We study the major bottlenecks for such a system, propose and evaluate a solution by using a hybrid image storage with solid state drives and hard disk drives, RESTfulWeb Services based protocols for exchanging image data, and a database based versioning scheme for efficient archive of image revision history. Our experiments show promising results of our methods, and our work provides a guideline for building enterprise level high performance medical image management systems.

Telemedical systems are not practical for use in a clinical workflow unless they are able to communicate with the Picture Archiving and Communications System (PACS). On the other hand, there are many medical imaging applications that are not developed as telemedical systems. Some medical imaging applications do not support collaboration and some do not communicate with the PACS and therefore limit their usability in clinical workflows. This paper presents a general architecture based on a three-tier architecture model. The architecture and the components developed within it, transform medical imaging applications into collaborative PACS-based telemedical systems. As a result, current medical imaging applications that are not telemedical, not supporting collaboration, and not communicating with PACS, can be enhanced to support collaboration among a group of physicians, be accessed remotely, and be clinically useful. The main advantage of the proposed architecture is that it does not impose any modification to the current medical imaging applications and does not make any assumptions about the underlying architecture or operating system.

Exploitation of advanced, PACS-centric image analysis and interpretation pipelines provides well-developed storage, retrieval, and archival capabilities along with state-of-the-art data providence, visualization, and clinical collaboration technologies. However, pursuit of integrated medical imaging analysis through a PACS environment can be limiting in terms of the overhead required to validate, evaluate and integrate emerging research technologies. Herein, we address this challenge through presentation of a high-throughput bundled resource imaging system (HUBRIS) as an extension to the Philips Research Imaging Development Environment (PRIDE). HUBRIS enables PACS-connected medical imaging equipment to invoke tools provided by the Java Imaging Science Toolkit (JIST) so that a medical imaging platform (e.g., a magnetic resonance imaging scanner) can pass images and parameters to a server, which communicates with a grid computing facility to invoke the selected algorithms. Generated images are passed back to the server and subsequently to the imaging platform from which the images can be sent to a PACS. JIST makes use of an open application program interface layer so that research technologies can be implemented in any language capable of communicating through a system shell environment (e.g., Matlab, Java, C/C++, Perl, LISP, etc.). As demonstrated in this proof-of-concept approach, HUBRIS enables evaluation and analysis of emerging technologies within well-developed PACS systems with minimal adaptation of research software, which simplifies evaluation of new technologies in clinical research and provides a more convenient use of PACS technology by imaging scientists.

Cortical activation maps estimated from MEG data fall prey to variability across subjects, trials, runs and potentially MEG centers. To combine MEG results across sites, we must demonstrate that inter-site variability in activation maps is not considerably higher than other sources of variability. By demonstrating relatively low inter-site variability with respect to inter-run variability, we establish a statistical foundation for sharing MEG data across sites for more powerful group studies or clinical trials of pathology. In this work, we analyze whether pooling MEG data across sites is more variable than aggregating MEG data across runs when estimating significant cortical activity. We use data from left median nerve stimulation experiments on four subjects at each of three sites on two runs occurring on consecutive days for each site. We estimate cortical current densities via minimum-norm imaging. We then compare maps across machines and across runs using two metrics: the Simpson coefficient, which admits equality if one map is equal in location to the other, and the Dice coefficient, which admits equality if one map is equal in location and size to the other. We find that sharing MEG data across sites does not noticeably affect group localization accuracy unless one set of data has abnormally low signal power.

Images are an integral part of medical practice for diagnosis, treatment planning and teaching. Image retrieval has gained in importance mainly as a research domain over the past 20 years. Both textual and visual retrieval of images are essential. In the process of mobile devices becoming reliable and having a functionality equaling that of formerly desktop clients, mobile computing has gained ground and many applications have been explored. This creates a new field of mobile information search & access and in this context images can play an important role as they often allow understanding complex scenarios much quicker and easier than free text. Mobile information retrieval in general has skyrocketed over the past year with many new applications and tools being developed and all sorts of interfaces being adapted to mobile clients. This article describes constraints of an information retrieval system including visual and textual information retrieval from the medical literature of BioMedCentral and of the RSNA journals Radiology and Radiographics. Solutions for mobile data access with an example on an iPhone in a web-based environment are presented as iPhones are frequently used and the operating system is bound to become the most frequent smartphone operating system in 2011. A web-based scenario was chosen to allow for a use by other smart phone platforms such as Android as well. Constraints of small screens and navigation with touch screens are taken into account in the development of the application. A hybrid choice had to be taken to allow for taking pictures with the cell phone camera and upload them for visual similarity search as most producers of smart phones block this functionality to web applications. Mobile information access and in particular access to images can be surprisingly efficient and effective on smaller screens. Images can be read on screen much faster and relevance of documents can be identified quickly through the use of images contained in the text. Problems with the many, often incompatible mobile platforms were discovered and are listed in the text. Mobile information access is a quickly growing domain and the constraints of mobile access also need to be taken into account for image retrieval. The demonstrated access to the medical literature is most relevant as the medical literature and their images are clearly the largest knowledge source in the medical field.

Acute intra-cranial hemorrhage (AIH) may result from traumatic brain injury (TBI). Successful management of AIH depends heavily on the speed and accuracy of diagnosis. Timely diagnosis in emergency environments in both civilian and military settings is difficult primarily due to severe time restraints and lack of resources. Often, diagnosis is performed by emergency physicians rather than trained radiologists. As a result, added support in the form of computer-aided detection (CAD) would greatly enhance the decision-making process and help in providing faster and more accurate diagnosis of AIH. This paper discusses the implementation of a CAD system in an emergency environment, and its efficacy in aiding in the detection of AIH.

Intensity-modulated radiation therapy (IMRT) has gained popularity in the treatment of cancers because of its excellent local control with decreased normal tissue complications. Yet, computer planning for the treatment relies heavily on human inspection of resultant radiation dose distribution within the irradiated region of the body. Even for experienced planners, comparison of IMRT plans is definitely cumbersome and not error-free. To solve this problem, a computer-aided decision-support system was built for automatic evaluation of IMRT plans based on the DICOM standard. A DICOM based IMRT plan with DICOM and DICOM-RT objects including CT images, RT Structure Set, RT Dose and RT Plan were retrieved from the Treatment Planning System for programming. Utilizing the MATLAB program language, the decoding-encoding software applications were developed on the basis of the DICOM information object definitions. After tracing the clinical workflow and understanding the needs and expectations from radiation oncologists, a set of routines were written to parse key data items such as isodose curves, region of interests, dose-volume histogram from the DICOM-RT objects. Then graphical user interfaces (GUIs) were created to allow planners to query for parameters such as overdose or underdose areas. A total of 30 IMRT plans were collected in a Department of Clinical Oncology for systematic testing of the DICOM-based decision-support system. Both structural and functional tests were implemented as a major step on the road to software maturity. With promising test results, this decision-support system could represent a major breakthrough in the routine IMRT planning workflow.

Clinical decisions for improving motor function in patients both with disability as well as improving an athlete's performance are made through clinical and movement analysis. Currently, this analysis facilitates identifying abnormalities in a patient's motor function for a large amount of neuro-musculoskeletal pathologies. However definitively identifying the underlying cause or long-term consequences of a specific abnormality in the patient's movement pattern is difficult since this requires information from multiple sources and formats across different times and currently relies on the experience and intuition of the expert clinician. In addition, this data must be persistent for longitudinal outcomes studies. Therefore a multimedia ePR system integrating imaging informatics data could have a significant impact on decision support within this clinical workflow. We present the design and architecture of such an ePR system as well as the data types that need integration in order to develop relevant decision support tools. Specifically, we will present two data model examples: 1) A performance improvement project involving volleyball athletes and 2) Wheelchair propulsion evaluation of patients with disabilities. The end result is a new frontier area of imaging informatics research within rehabilitation engineering and biomechanics.

Multiple sclerosis (MS) is a demyelinating disease of the central nervous system. The chronic nature of MS necessitates multiple MRI studies to track disease progression. We have presented an imaging informatics decision-support system, called MS eFolder, designed to integrate patient clinical data with MR images and a computer-aided detection (CAD) component for automatic white matter lesion quantification. The purpose of the MS eFolder is to comprehensively present MS patient data for clinicians and radiologists, while providing a lesion quantification tool that can be objective and consistent for MS tracking in longitudinal studies. The MS CAD algorithm is based on the K-nearest neighbor (KNN) principles and has been integrated within the eFolder system. Currently, the system has been completed and the CAD algorithm for quantifying MS lesions has undergone the expert evaluation in order to validate system performance and accuracy. The evaluation methodology has been developed and the data has been collected, including over 100 MS MRI cases with various age and ethnic backgrounds. The preliminary results of the evaluation are expected to include sensitivity and specificity of lesion and non-lesion voxels in the white matter, the effectiveness of different probability thresholds for each voxel, and comparison between CAD quantification results and radiologists' manual readings. The results aim to show the effectiveness of a MS lesion CAD system to be used in a clinical setting, as well as a step closer to full clinical implementation of the eFolder system.

The role of computers in medical image display and analysis continues to be one of the most computationally demanding tasks facing modern computers. Recent advances in GPU architecture have allowed for a new programming paradigm which utilized the massively parallel computational capacity of GPUs for general purpose computing. These parallel processors provide substantial performance benefits in image analysis and manipulation. Automated segmentation algorithms gain the most benefit from incorporation of GPU computing into the image processing workflow. There are also new visualization paradigms, such as stereoscopic 3D, which have been made possible by the continued increase in computational capacity of GPUs. These two key functions of modern GPUs will enable medical imagers to keep pace with the increasing size of scan data sets while allowing for new and innovative analysis and interaction paradigms.

In addition to the primary care context, medical images are often useful for research projects and community healthcare networks, so-called "secondary use". Patient privacy becomes an issue in such scenarios since the disclosure of personal health information (PHI) has to be prevented in a sharing environment. In general, most PHIs should be completely removed from the images according to the respective privacy regulations, but some basic and alleviated data is usually required for accurate image interpretation. Our objective is to utilize and enhance these specifications in order to provide reliable software implementations for de- and re-identification of medical images suitable for online and offline delivery. DICOM (Digital Imaging and Communications in Medicine) images are de-identified by replacing PHI-specific information with values still being reasonable for imaging diagnosis and patient indexing. In this paper, this approach is evaluated based on a prototype implementation built on top of the open source framework DCMTK (DICOM Toolkit) utilizing standardized de- and re-identification mechanisms. A set of tools has been developed for DICOM de-identification that meets privacy requirements of an offline and online sharing environment and fully relies on standard-based methods.

Breast cancer is the most common type of non-skin cancer in women. 2D mammography is a screening tool to aid in the early detection of breast cancer, but has diagnostic limitations of overlapping tissues, especially in dense breasts. 3D mammography has the potential to improve detection outcomes by increasing specificity, and a new 3D screening tool with a 3D display for mammography aims to improve performance and efficiency as compared to 2D mammography. An observer study using a mammography phantom was performed to compare traditional 2D mammography with this ne 3D mammography technique. In comparing 3D and 2D mammography there was no difference in calcification detection, and mass detection was better in 2D as compared to 3D. There was a significant decrease in reading time for masses, calcifications, and normals in 3D compared to 2D, however, as well as more favorable confidence levels in reading normal cases. Given the limitations of the mammography phantom used, however, a clearer picture in comparing 3D and 2D mammography may be better acquired with the incorporation of human studies in the future.

Modern Medical diagnoses are more and more based upon interactivity with different kinds of data. Images are at the very base of these diagnosis policies and require high degrees of interaction, a requirement that most compression standards do not meet since for achieving this, high granularity levels are needed. JPEG2000 (J2K) has lately arisen as a compression standard that tackles with these challenges, allowing appropriate compression rates and efficient access to data, i.e. random spatial access at any resolution and with any desired quality. Based on the J2K standard method, this article presents a 3D compression method which adapts the J2K simplicity handling of 2D data and includes the 3D information with no modification of the structures used in 2D implementations. The proposed method was compared with a conventional J2K implementation in 2D, using 3D and 4D data, showing that the 3D strategy saves around a 12 % of hard disk space when compared to the conventional 2D implementation.

The current trend in medical device industry towards modular systems architectures leads to a growing need of standards which guarantee interoperability between modules and systems. Unfortunately, there are several standards as well as proprietary solutions which might have an overlapping scope, and within standard-based systems there are sometimes different interpretations on how a standard shall be used. We propose to use the "Integrating the Healthcare Enterprise (IHE)" initiative as a platform for clarification the use of standards for the surgical domain. To show the feasibility of this approach, we sketched three Integration Profiles within the field of implantation planning: One for the description of surface meshes, one for Implant Templates and one for Implantation Plans. This paper focus is on the first IHE proposals based on the existing DICOM Supplements in the context of surgical implantation planning. The Integration Profiles are describing, how the corresponding DICOM data structures introduced by DICOM Supplements 131, 132 and 134 (Implant Template, Surface Mesh Segmentation and Implantation Plan) shall be used in order to accomplish the data transfer needed for the whole process chain from implant manufacturer via the manufacturer of an implantation planning software to the end user in the hospital.

Managing different registries and repositories within healthcare regions grows the risk of having almost the same information but with different status and with different content. This is due to the fact that when medical information is created it´s done in a dynamical process that will lead to that information will change its contents during lifetime within the "active" healthcare phase. The information needs to be easy accessible, being the platform for making the medical decisions transparent. In the Region Västra Götaland (VGR), Sweden, data is shared from 29 X-ray departments with different Picture Archive and Communication Systems (PACS) and Radiology Information Systems (RIS) systems through the Infobroker solution, that's acts as a broker between the actors involved. Request/reports from RIS are stored as DIgital COmmunication in Medicine (DICOM)-Structured Reports (SR) objects, together with the images. Every status change within this activities are updated within the Information Infrastructure based on Integrating the Healthcare Enterprise (IHE) mission. Cross-enterprise Document Sharing for Imaging (XDS-I) were the registry and the central repository are the components used for sharing medical documentation. The VGR strategy was not to apply one regional XDS-I registry and repository, instead VGR applied an Enterprise Architecture (EA) intertwined with the Information Infrastructure for the dynamic delivery to consumers. The upcoming usage of different Regional XDS registries and repositories could lead to new ways of carrying out shared work but it can also lead into "problems". XDS and XDS-I implemented without a strategy could lead to increased numbers of status/versions but also duplication of information in the Information Infrastructure.

We designed the image-enabled EHR sharing solution (i-EHR) for cross-enterprise and cross-domain with SOA architecture and combined the grid-based image management and distribution capability, which are compliant with IHE XDS-I/XCA integration profiles. We selected one districts with four hospitals and two hospital groups as image sharing pilot testing bed. Our approach presented in this presentation uses peer-to-peer mode to share and exchange image data cross enterprise PACSs and domains, which provides single point of services to local systems so it is easy to integrate with different vendor's PACS and easy to deploy to different hospitals to implement the i-EHR.

There are many Web-based image accessing technologies used in medical imaging area, such as component-based (ActiveX Control) thick client Web display, Zerofootprint thin client Web viewer (or called server side processing Web viewer), Flash Rich Internet Application(RIA) ,or HTML5 based Web display. Different Web display methods have different peformance in different network environment. In this presenation, we give an evaluation on two developed Web based image display systems. The first one is used for thin client Web display. It works between a PACS Web server with WADO interface and thin client. The PACS Web server provides JPEG format images to HTML pages. The second one is for thick client Web display. It works between a PACS Web server with WADO interface and thick client running in browsers containing ActiveX control, Flash RIA program or HTML5 scripts. The PACS Web server provides native DICOM format images or JPIP stream for theses clients.

The standard format for medical imaging storage and transmission is DICOM. openEHR is an open standard specification in health informatics that describes the management and storage, retrieval and exchange of health data in electronic health records. Considering that the integration of DICOM and openEHR is beneficial to information sharing, on the basis of XML-based DICOM format, we developed a method of creating a DICOM Imaging Archetype in openEHR to enable the integration of DICOM and openEHR. Each DICOM file contains abundant imaging information. However, because reading a DICOM involves looking up the DICOM Data Dictionary, the readability of a DICOM file has been limited. openEHR has innovatively adopted two level modeling method, making clinical information divided into lower level, the information model, and upper level, archetypes and templates. But one critical challenge posed to the development of openEHR is the information sharing problem, especially in imaging information sharing. For example, some important imaging information cannot be displayed in an openEHR file. In this paper, to enhance the readability of a DICOM file and semantic interoperability of an openEHR file, we developed a method of mapping a DICOM file to an openEHR file by adopting the form of archetype defined in openEHR. Because an archetype has a tree structure, after mapping a DICOM file to an openEHR file, the converted information is structuralized in conformance with openEHR format. This method enables the integration of DICOM and openEHR and data exchange without losing imaging information between two standards.

Digital Imaging and Communications in Medicine (DICOM) is a standard for handling, storing, printing, and transmitting information in medical imaging. XML (Extensible Markup Language) is a set of rules for encoding documents in machine-readable form which has become more and more popular. The combination of these two is very necessary and promising. Using XML tags instead of numeric labels in DICOM files will effectively increase the readability and enhance the clear hierarchical structure of DICOM files. However, due to the fact that the XML tags rely heavily on the orders of the tags, the strong data dependency has a lot of influence on the flexibility of inserting and exchanging data. In order to improve the extensibility and sharing of DICOM files, this paper introduces XML Path-Tag to DICOM. When a DICOM file is converted to XML format, adding simple Path-Tag into the DICOM file in place of complex tags will keep the flexibility of a DICOM file while inserting data elements and give full play to the advantages of the structure and readability of an XML file. Our method can solve the weak readability problem of DICOM files and the tedious work of inserting data into an XML file. In addition, we set up a conversion engine that can transform among traditional DICOM files, XML-DCM and XML-DCM files involving XML Path-Tag efficiently.

Multiple Sclerosis (MS) is a disease which is caused by damaged myelin around axons of the brain and spinal cord. Currently, MR Imaging is used for diagnosis, but it is very highly variable and time-consuming since the lesion detection and estimation of lesion volume are performed manually. For this reason, we developed a CAD (Computer Aided Diagnosis) system which would assist segmentation of MS to facilitate physician's diagnosis. The MS CAD system utilizes K-NN (k-nearest neighbor) algorithm to detect and segment the lesion volume in an area based on the voxel. The prototype MS CAD system was developed under the MATLAB environment. Currently, the MS CAD system consumes a huge amount of time to process data. In this paper we will present the development of a second version of MS CAD system which has been converted into C/C++ in order to take advantage of the GPU (Graphical Processing Unit) which will provide parallel computation. With the realization of C/C++ and utilizing the GPU, we expect to cut running time drastically. The paper investigates the conversion from MATLAB to C/C++ and the utilization of a high-end GPU for parallel computing of data to improve algorithm performance of MS CAD.

In this paper, we proposed a novel architecture integrated with RIS/PACS system that combined image annotation, CBIR techniques and high-dimensional index to retrieve similar medical images with one or more relevant focus in large scale medical image database. In our designed system, regions of interest (ROIs) were labeled by symptom descriptions found in relevant radiology reports as semantic navigation. The annotations were saved as xml file with image makeup language (IML). Then low level features such as texture and statistic features were extracted from the ROIs of lesions and inserted into a database. Recursive feature elimination algorithm was applied to find a high performance feature subset for each symptom. These subsets were used to build high dimensional index with semantic labels guiding the searching path as the navigation. As there might be more than one focus in one image, weight values specified by the user were introduced to calculate the final similarities. The searching results of medical images with multi-focal diseases are likely to have the same pathologies and visual effects with example image and are valuable for imaging diagnosis. The system was implemented for lung CT images, but it could be easily extended to other organs.

The dramatic increase of diagnostic imaging capabilities over the past decade has contributed to increased radiation exposure to patient populations. Several factors have contributed to the increase in imaging procedures: wider availability of imaging modalities, increase in technical capabilities, rise in demand by patients and clinicians, favorable reimbursement, and lack of guidelines to control utilization. The primary focus of this research is to provide in depth information about radiation doses that patients receive as a result of CT exams, with the initial investigation involving abdominal CT exams. Current dose measurement methods (i.e. CTDI_vol Computed Tomography Dose Index) do not provide direct information about a patient's organ dose. We have developed a method to determine CTDI_vol normalized organ doses using a set of organ specific exponential regression equations. These exponential equations along with measured CTDI_vol are used to calculate organ dose estimates from abdominal CT scans for eight different patient models. For each patient, organ dose and CTDI_vol were estimated for an abdominal CT scan. We then modified the DICOM Radiation Dose Structured Report (RDSR) to store the pertinent patient information on radiation dose to their abdominal organs.

With the widely use of healthcare information technology in hospitals, the patients' medical records are more and more complex. To transform the text- or image-based medical information into easily understandable and acceptable form for human, we designed and developed an innovation indexing method which can be used to assign an anatomical 3D structure object to every patient visually to store indexes of the patients' basic information, historical examined image information and RIS report information. When a doctor wants to review patient historical records, he or she can first load the anatomical structure object and the view the 3D index of this object using a digital human model tool kit. This prototype system helps doctors to easily and visually obtain the complete historical healthcare status of patients, including large amounts of medical data, and quickly locate detailed information, including both reports and images, from medical information systems. In this way, doctors can save time that may be better used to understand information, obtain a more comprehensive understanding of their patients' situations, and provide better healthcare services to patients.

Learning radiology requires systematic and comprehensive study of a large knowledge base of medical images. In this work is presented the development of a digital radiology teaching file system. The proposed system has been created in order to offer a set of customized services regarding to users' contexts and their informational needs. This has been done by means of an electronic infrastructure that provides easy and integrated access to all relevant patient data at the time of image interpretation, so that radiologists and researchers can examine all available data to reach well-informed conclusions, while protecting patient data privacy and security. The system is presented such as an environment which implements a distributed clinical database, including medical images, authoring tools, repository for multimedia documents, and also a peer-reviewed model which assures dataset quality. The current implementation has shown that creating clinical data repositories on networked computer environments points to be a good solution in terms of providing means to review information management practices in electronic environments and to create customized and contextbased tools for users connected to the system throughout electronic interfaces.

Advances in biology, computer technology and imaging technology have given rise to a scientific specialty referred to as molecular imaging, which is the in vivo imaging of cellular and molecular pathways using contrast-enhancing targeting agents. Increasing amounts of molecular imaging research are being performed at pre-clinical stages, generating diverse datasets that are unstructured and thereby lacking in archiving and distribution solutions. Since PACS in radiology is a mature clinical archiving solution, a method is proposed to convert current imaging files from preclinical molecular imaging studies into DICOM formats for archival and retrieval from PACS systems. A web-based DICOM gateway is presented with an emphasis on metadata mapping in the DICOM header, system connectivity, and overall user workflow. This effort to conform preclinical imaging data to the DICOM standard is necessary to utilize current PACS solutions for preclinical imaging data content archiving and distribution.

We have developed the teleradiology network system with a new information security solution that provided with web medical image conference system. In the teleradiology network system, the security of information network is very important subjects. We are studying the secret sharing scheme as a method safely to store or to transmit the confidential medical information used with the teleradiology network system. The confidential medical information is exposed to the risk of the damage and intercept. Secret sharing scheme is a method of dividing the confidential medical information into two or more tallies. Individual medical information cannot be decoded by using one tally at all. Our method has the function of RAID. With RAID technology, if there is a failure in a single tally, there is redundant data already copied to other tally. Confidential information is preserved at an individual Data Center connected through internet because individual medical information cannot be decoded by using one tally at all. Therefore, even if one of the Data Centers is struck and information is damaged, the confidential medical information can be decoded by using the tallies preserved at the data center to which it escapes damage. We can safely share the screen of workstation to which the medical image of Data Center is displayed from two or more web conference terminals at the same time. Moreover, Real time biometric face authentication system is connected with Data Center. Real time biometric face authentication system analyzes the feature of the face image of which it takes a picture in 20 seconds with the camera and defends the safety of the medical information. We propose a new information transmission method and a new information storage method with a new information security solution.