The issues of big data in medical imaging informatics have special characters which need to be deal with in healthcare service and research. These characters lead to many technical challenges in medical imaging informatics. The data set sizes and hiding states of medical imaging are major factors which make big data of medical imaging be quite different from other kind of big data. The possible and perspective solutions of big data issues in medical imaging informatics are discussed in this presentation, and also some of our research projects related to five V features of big data in medical imaging and informatics have been briefed.

One of the challenges of today’s healthcare is that data from radiology is heterogeneous, stored and managed in silos created by PACS vendors. Also seen is a lack of coordinated use of harmonized reference information models and established healthcare standards.

Radiology in Region Västra Götaland has been entering the world of “Big Data” since 2006, 34 departments split into 4 private image center, 2 small-size hospital, 4 middle-sized hospital groups and one University hospital. Using the same information infrastructure as a means of collaborating and sharing information between. As an organization building for the future we must meet the values and requirements of the stakeholders and count the patient as the major actor.

Can ”Big Data” analytics be a valuable asset from a regional management perspective?

Our initial findings indicates that this is the case, based on three different perspectives – work practice changes, understanding data quality when sharing information and introducing new services in work practice. Going from local to enterprise workflow utilizing the power of “Big Data”, not only by volume but also by combining diverse sources and aggregate the information domains, visualize new trends as well as dependencies more effectively. Building trust by the use of Big Data in healthcare involves a long and winding journey, but the persevering infrastructure-oriented organization will give new ways of collaboration for the enterprise it serves. It also involves continuous negotiation with people concerning how and why they should collaborate with new actors within the region to achieve patient centric care. This will nurture a more open-minded, hopeful and life-affirming holistic approach involving all stakeholders, newcomers’ specialists and patients.

The cohort size required in epidemiological imaging genetics studies often mandates the pooling of data from multiple hospitals. Patient data, however, is subject to strict privacy protection regimes, and physical data storage may be legally restricted to a hospital network. To enable biomarker discovery, fast data access and interactive data exploration must be combined with high-performance computing resources, while respecting privacy regulations. We present a system using fast and inherently secure light-paths to access distributed data, thereby obviating the need for a central data repository. A secure private cloud computing framework facilitates interactive, computationally intensive exploration of this geographically distributed, privacy sensitive data. As a proof of concept, MRI brain imaging data hosted at two remote sites were processed in response to a user command at a third site. The system was able to automatically start virtual machines, run a selected processing pipeline and write results to a user accessible database, while keeping data locally stored in the hospitals. Individual tasks took approximately 50% longer compared to a locally hosted blade server but the cloud infrastructure reduced the total elapsed time by a factor of 40 using 70 virtual machines in the cloud. We demonstrated that the combination light-path and private cloud is a viable means of building an analysis infrastructure for secure data analysis. The system requires further work in the areas of error handling, load balancing and secure support of multiple users.

Our overall objective is the utilization of heterogeneous and distributed radio-oncological data in retrospective clinical trials. Previously, we have successfully introduced a central research database for collection of heterogeneous data from distributed systems. The next step is the integration of image analysis tools in the standard retrieval process. Hence, analyses for complex medical questions can be processed automatically and facilitated immensely. In radiation oncology recurrence analysis is a central approach for the evaluation of therapeutic concepts. However, various analysis steps have to be performed like image registration, dose transformation and dose statistics.

In this paper we show the integration of image analysis tools in the standard retrieval process by connecting them with our central research database using a service-oriented approach. A concrete problem from recurrence analysis has been selected to prove our concept exemplarily. We implemented service-oriented data collection and analysis tools to use them in a central analysis platform, which is based on a work flow management system. An analysis work flow has been designed that, at first, identifies patients in the research database fulfilling the inclusion criteria. Then the relevant imaging data is collected. Finally the imaging data is analyzed automatically. After the successful work flow execution, the results are available for further evaluation by a physician.

As a result, the central research database has been connected successfully with automatic data collection and image analysis tools and the feasibility of our service-oriented approach has been demonstrated. In conclusion, our approach will simplify retrospective clinical trials in our department in future.

This paper presents progress on imaging the research field of Imaging Informatics, mapped as the clustering of its communities together with their main results by applying a process to produce a dynamical image of the interactions between their results and their common object(s) of research. The basic side draws from a fundamental research on the concept of dimensions and projective space spanning several streams of research about three-dimensional perceptivity and re-cognition and on their relation and reduction to spatial dimensionality. The application results in an N-dimensional mapping in Bio-Medical Imaging, with dimensions such as inflammatory activity, MRI acquisition sequencing, spatial resolution (voxel size), spatiotemporal dimension inferred, toxicity, depth penetration, sensitivity, temporal resolution, wave length, imaging duration, etc. Each field is represented through the projection of papers’ and projects’ ‘discriminating’ quantitative results onto the specific N-dimensional hypercube of relevant measurement axes, such as listed above and before reduction. Past published differentiating results are represented as red stars, achieved unpublished results as purple spots and projects at diverse progress advancement levels as blue pie slices. The goal of the mapping is to show the dynamics of the trajectories of the field in its own experimental frame and their direction, speed and other characteristics. We conclude with an invitation to participate and show a sample mapping of the dynamics of the community and a tentative predictive model from community contribution.

In the past, we have developed and displayed a multiple sclerosis eFolder system for patient data storage, image viewing, and automatic lesion quantification results stored in DICOM-SR format. The web-based system aims to be integrated in DICOM-compliant clinical and research environments to aid clinicians in patient treatments and disease tracking. This year, we have further developed the eFolder system to handle big data analysis and data mining in today’s medical imaging field. The database has been updated to allow data mining and data look-up from DICOM-SR lesion analysis contents. Longitudinal studies are tracked, and any changes in lesion volumes and brain parenchyma volumes are calculated and shown on the webbased user interface as graphical representations. Longitudinal lesion characteristic changes are compared with patients’ disease history, including treatments, symptom progressions, and any other changes in the disease profile. The image viewer is updated such that imaging studies can be viewed side-by-side to allow visual comparisons. We aim to use the web-based medical imaging informatics eFolder system to demonstrate big data analysis in medical imaging, and use the analysis results to predict MS disease trends and patterns in Hispanic and Caucasian populations in our pilot study. The discovery of disease patterns among the two ethnicities is a big data analysis result that will help lead to personalized patient care and treatment planning.

We demonstrate how a cloud-based PACS can exchange information with other medical systems, including other cloud-based PACS, to provide a comprehensive and integrated view of a patient's health record. Such a consolidated report will lead to improved patient care.

Medical imaging research depends basically on the availability of large image collections, image processing and analysis algorithms, hardware and a multidisciplinary research team. It has to be reproducible, free of errors, fast, accessible through a large variety of devices spread around research centers and conducted simultaneously by a multidisciplinary team. Therefore, we propose a collaborative research environment, named Adessowiki, where tools and datasets are integrated and readily available in the Internet through a web browser. Moreover, processing history and all intermediate results are stored and displayed in automatic generated web pages for each object in the research project or clinical study. It requires no installation or configuration from the client side and offers centralized tools and specialized hardware resources, since processing takes place in the cloud.

The increasing need of remote medical investigation services in the framework of collaborative multidisciplinary meetings (e.g. cancer follow-up) raises the challenge of on-line remote access of (large amount of) radiologic data in a limited period of time. This paper proposes a scalable compression framework of DICOM images providing low-latency display through low speed networks. The developed approach relies on useless information removal from images (i.e. not related with the patient body) and the exploitation of the JPEG2000 standard to achieve progressive quality encoding and access of the data. This mechanism also allows the efficient exploitation of any idle times (corresponding to on-line visual image analysis) to download the remaining data at lossless quality in a way transparent to the user, thus minimizing the perceived latency. The experiments performed in comparison with exchanging uncompressed or JPEGlossless compressed DICOM data, showed the benefit of the proposed approach for collaborative on-line remote diagnosis and follow-up services.

Advanced MRI research and other imaging modalities may serve as biomarkers for the evaluation of traumatic brain injury (TBI) recovery. However, these advanced modalities typically require off-line processing which creates images that are incompatible with radiologist viewing software sold commercially. AGFA Impax is an example of such a picture archiving and communication system(PACS) that is used by many radiology departments in the United States Military Health System. By taking advantage of Impax’s use of the Digital Imaging and Communications in Medicine (DICOM) standard, we developed a system that allows for advanced medical imaging to be incorporated into clinical PACS. Radiology research can now be conducted using existing clinical imaging display platforms resources in combination with image processingtechniques that are only available outside of the clinical scanning environment. We extracted the spatial and identification elements of theDICOM standard that are necessary to allow research images to be incorporatedinto a clinical radiology system, and developed a tool that annotates research images with the proper tags. This allows for the evaluation of imaging representations of biological markers that may be useful in theevaluation of TBI and TBI recovery.

Previously, we presented an Interdisciplinary Comprehensive Arm Rehabilitation Evaluation (ICARE) imaging informatics system that supports a large-scale phase III stroke rehabilitation trial. The ePR system is capable of displaying anonymized patient imaging studies and reports, and the system is accessible to multiple clinical trial sites and users across the United States via the web. However, the prior multicenter stroke rehabilitation trials lack any significant neuroimaging analysis infrastructure. In stroke related clinical trials, identification of the stroke lesion characteristics can be meaningful as recent research shows that lesion characteristics are related to stroke scale and functional recovery after stroke. To facilitate the stroke clinical trials, we hope to gain insight into specific lesion characteristics, such as vascular territory, for patients enrolled into large stroke rehabilitation trials. To enhance the system’s capability for data analysis and data reporting, we have integrated new features with the system: a digital brain template display, a lesion quantification tool and a digital case report form. The digital brain templates are compiled from published vascular territory templates at each of 5 angles of incidence. These templates were updated to include territories in the brainstem using a vascular territory atlas and the Medical Image Processing, Analysis and Visualization (MIPAV) tool. The digital templates are displayed for side-by-side comparisons and transparent template overlay onto patients’ images in the image viewer. The lesion quantification tool quantifies planimetric lesion area from user-defined contour. The digital case report form stores user input into a database, then displays contents in the interface to allow for reviewing, editing, and new inputs. In sum, the newly integrated system features provide the user with readily-accessible web-based tools to identify the vascular territory involved, estimate lesion area, and store these results in a web-based digital format.

Recent advances in internet browser technologies makes it possible to incorporate advanced functionality of a traditional PACS for viewing DICOM medical images on standard web browsers without the need to pre-install any plug-ins, apps or software. We demonstrate some of the capabilities of standard web browsers setting the stage for a cloud-based PACS.

In this paper, we propose a web-based framework for traditional PACS that has the potential to revolutionize the management and display of medical images while simultaneously decreasing the cost of deployment. Specifically, we demonstrate that all the tools to replace traditional PACS are finally available to make the transition to the internet. In addition, web-based solutions over capabilities that go well beyond traditional PACS.

Full-Field Optical Coherence Tomography (FF-OCT) is a high resolution instrument in 3 dimensional (3D) space, including lateral and longitudinal direction. With FF-OCT, we can perform 3D scanning for excised biopsy or cell culture sample to obtain cellular information. In this work, we have set up a high resolution FF-OCT scanning instrument that can perform cellular resolution tomography scanning of skin tissue for histopathology study. In a scan range of 1cm(x), 1cm(y), 106μm(z), for example, digital data occupies 253 GB capacity. Copying these materials is time consuming, not to mention efficient browsing and analyzing of these data. To solve the problem of information delivery, we have established a network service to browse and analyze the huge volume data.

Visualising volumetric medical images such as computed tomography and magnetic resonance imaging (MRI) on picture archiving and communication systems (PACS) clients is often achieved by image browsing in sagittal, coronal or axial views or three-dimensional (3D) rendering. This latter technique requires fine thresholding for MRI. On the other hand, computing virtual radiograph images, also referred to as digitally reconstructed radiographs (DRR), provides in a single two-dimensional (2D) image a complete overview of the 3D data. It appears therefore as a powerful alternative for MRI visualisation and preview in PACS. This study describes a method to compute DRR from T1-weighted MRI. After segmentation of the background, a histogram distribution analysis is performed and each foreground MRI voxel is labeled as one of three tissues: cortical bone, also known as principal absorber of the X-rays, muscle and fat. An intensity level is attributed to each voxel according to the Hounsfield scale, linearly related to the X-ray attenuation coefficient. Each DRR pixel is computed as the accumulation of the new intensities of the MRI dataset along the corresponding X-ray. The method has been tested on 16 T1-weighted MRI sets. Anterior-posterior and lateral DRR have been computed with reasonable qualities and avoiding any manual tissue segmentations. This proof-of-concept holds for research application for use in clinical PACS.

The evolution of cloud computing is driving the next generation of diagnostic imaging (DI) systems. Cloud-based DI systems are able to deliver better services to patients without constraining to their own physical facilities. However, privacy and security concerns have been consistently regarded as the major obstacle for adoption of cloud computing by healthcare domains. Furthermore, traditional computing models and interfaces employed by DI systems are not ready for accessing diagnostic images through mobile devices. RESTful is an ideal technology for provisioning both mobile services and cloud computing. OpenID Connect, combining OpenID and OAuth together, is an emerging REST-based federated identity solution. It is one of the most perspective open standards to potentially become the de-facto standard for securing cloud computing and mobile applications, which has ever been regarded as “Kerberos of Cloud”. We introduce OpenID Connect as an identity and authentication service in cloud-based DI systems and propose enhancements that allow for incorporating this technology within distributed enterprise environment. The objective of this study is to offer solutions for secure radiology image sharing among DI-r (Diagnostic Imaging Repository) and heterogeneous PACS (Picture Archiving and Communication Systems) as well as mobile clients in the cloud ecosystem. Through using OpenID Connect as an open-source identity and authentication service, deploying DI-r and PACS to private or community clouds should obtain equivalent security level to traditional computing model.

Objective: This research investigates the needs and opinions of radiologists on the use of enhanced information technologies and approaches to improve the functionality of Picture Archiving and Communication Systems (PACS). Method: Six interviews were conducted in the main governmental hospital in Kuwait (AL-Sabah Hospital) with radiologists, including two senior radiologists, two junior radiologists, and two trainee radiologists undertaking the Irish radiology board. Results: The radiologists identified a number of limitations that exist in current PACS and requirements to enhance usability and functionality. However, it was the case that some of the radiologists had little knowledge about the advanced trends in PACS. Four themes emerged from the thematic analysis of the data: (1) limitations of traditional PACS; (2) Features and requirements that can increase PACS functionality; (3) web based solutions of PACS; (4) PACS in mobile phones. Conclusion: It is widely recognized that PACS has limitations. This research has identified themes that, when incorporated, will enhance the functionality of PACS and provide better quality clinical practice. This research has determined the important future trends of PACS. Primarily web based solutions and use in mobile phones. The findings from this research can be used as recommendations to vendors, for product development and medical institutes to consider when undertaking implementation of PACS and training future radiologists.

DICOM Index Tracker (DIT) is an integrated platform to harvest rich information available from Digital Imaging and Communications in Medicine (DICOM) to improve quality assurance in radiology practices. It is designed to capture and maintain longitudinal patient-specific exam indices of interests for all diagnostic and procedural uses of imaging modalities. Thus, it effectively serves as a quality assurance and patient safety monitoring tool. The foundation of DIT is an intelligent database system which stores the information accepted and parsed via a DICOM receiver and parser. The database system enables the basic dosimetry analysis. The success of DIT implementation at Mayo Clinic Arizona calls for the DIT deployment at the enterprise level which requires significant improvements. First, for geographically distributed multi-site implementation, the first bottleneck is the communication (network) delay; the second is the scalability of the DICOM parser to handle the large volume of exams from different sites. To address this issue, DICOM receiver and parser are separated and decentralized by site. To facilitate the enterprise wide Quality Assurance (QA), a notable challenge is the great diversities of manufacturers, modalities and software versions, as the solution DIT Enterprise provides the standardization tool for device naming, protocol naming, physician naming across sites. Thirdly, advanced analytic engines are implemented online which support the proactive QA in DIT Enterprise.

The deployment of regional and national Electronic Health Record solutions has been a focus of many countries throughout the past decade. A major challenge for these deployments has been support for ubiquitous image viewing. More specifically, these deployments require an imaging solution that can work over the Internet, leverage any point of service device: desktop, tablet, phone; and access imaging data from any source seamlessly. Whereas standards exist to enable ubiquitous image viewing, few if any solutions exist that leverage these standards and meet the challenge. Rather, most of the currently available web based DI viewing solutions are either proprietary solutions or require special plugins. We developed a true zero foot print browser based DI viewing solution based on the Web Access DICOM Objects (WADO) and Cross-enterprise Document Sharing for Imaging (XDS-I.b) standards to a) demonstrate that a truly ubiquitous image viewer can be deployed; b) identify the gaps in the current standards and the design challenges for developing such a solution. The objective was to develop a viewer, which works on all modern browsers on both desktop and mobile devices. The implementation allows basic viewing functionalities of scroll, zoom, pan and window leveling (limited). The major gaps identified in the current DICOM WADO standards are a lack of ability to allow any kind of 3D reconstruction or MPR views. Other design challenges explored include considerations related to optimization of the solution for response time and low memory foot print.

The deployment of regional and national Electronic Health Record solutions has been a focus of many countries throughout the past decade. Most of these deployments have taken the approach of “sharing” imaging exams via portals and web-based viewers. The motivation of portal/web-based access is driven by a) the perception that review of imaging exams via portal methods is satisfactory to all users and b) the perceived complexity of ingesting foreign exams into local systems. This research project set out to objectively evaluate who really needs foreign exams within their local systems, what those systems might be and how often this is required. Working on the belief that Foreign Exam Management (FEM) is required to support clinical workflow, the project implemented a FEM capability within an XDSI. b domain to identify the design challenges and nuances associated with FEM.

The digital imaging and communications in medicine (DICOM) protocol is nowadays the leading standard for capture, exchange and storage of image data in medical applications. A broad range of commercial, free, and open source software tools supporting a variety of DICOM functionality exists. However, different from patient’s care in hospital, DICOM has not yet arrived in electronic data capture systems (EDCS) for clinical trials. Due to missing integration, even just the visualization of patient’s image data in electronic case report forms (eCRFs) is impossible. Four increasing levels for integration of DICOM components into EDCS are conceivable, raising functionality but also demands on interfaces with each level. Hence, in this paper, a comprehensive evaluation of 27 DICOM viewer software projects is performed, investigating viewing functionality as well as interfaces for integration. Concerning general, integration, and viewing requirements the survey involves the criteria (i) license, (ii) support, (iii) platform, (iv) interfaces, (v) two-dimensional (2D) and (vi) three-dimensional (3D) image viewing functionality. Optimal viewers are suggested for applications in clinical trials for 3D imaging, hospital communication, and workflow. Focusing on open source solutions, the viewers ImageJ and MicroView are superior for 3D visualization, whereas GingkoCADx is advantageous for hospital integration. Concerning workflow optimization in multi-centered clinical trials, we suggest the open source viewer Weasis. Covering most use cases, an EDCS and PACS interconnection with Weasis is suggested.

Modality filtering is an important feature in biomedical image searching systems and may significantly improve the retrieval performance of the system. This paper presents a new method for extracting endoscopic image figures from photograph images in biomedical literature, which are found to have highly diverse content and large variability in appearance. Our proposed method consists of three main stages: tissue image extraction, endoscopic image candidate extraction, and ophthalmic image filtering. For tissue image extraction we use image patch level clustering and MRF relabeling to detect images containing skin/tissue regions. Next, we find candidate endoscopic images by exploiting the round shape characteristics that commonly appear in these images. However, this step needs to compensate for images where endoscopic regions are not entirely round. In the third step we filter out the ophthalmic images which have shape characteristics very similar to the endoscopic images. We do this by using text information, specifically, anatomy terms, extracted from the figure caption. We tested and evaluated our method on a dataset of 115,370 photograph figures, and achieved promising precision and recall rates of 87% and 84%, respectively.

Tuberculosis (TB) is a major public health problem worldwide, and highly prevalent in developing countries. According to the World Health Organization (WHO), over 95% of TB deaths occur in low- and middle- income countries that often have under-resourced health care systems. In an effort to aid population screening in such resource challenged settings, the U.S. National Library of Medicine has developed a chest X-ray (CXR) screening system that provides a pre-decision on pulmonary abnormalities. When the system is presented with a digital CXR image from the Picture Archive and Communication Systems (PACS) or an imaging source, it automatically identifies the lung regions in the image, extracts image features, and classifies the image as normal or abnormal using trained machine-learning algorithms. The system has been trained on adult CXR images, and this article presents enhancements toward including pediatric CXR images. Our adult lung boundary detection algorithm is model-based. We note the lung shape differences during pediatric developmental stages, and adulthood, and propose building new lung models suitable for pediatric developmental stages. In this study, we quantify changes in lung shape from infancy to adulthood toward enhancing our lung segmentation algorithm. Our initial findings suggest pediatric age groupings of 0 - 23 months, 2 - 10 years, and 11 - 18 years. We present justification for our groupings. We report on the quality of boundary detection algorithm with the pediatric lung models.

One approach to image perception studies in mammography using virtual clinical trials involves the insertion of simulated lesions into normal mammograms. To facilitate this, a method has been developed that allows for sampling of lesion positions across the cranio-caudal and medio-lateral radiographic projections in accordance with measured distributions of real lesion locations.

6825 mammograms from our mammography image database were segmented to find the breast outline. The outlines were averaged and smoothed to produce an average outline for each laterality and radiographic projection.

Lesions in 3304 mammograms with malignant findings were mapped on to a standardised breast image corresponding to the average breast outline using piecewise affine transforms. A four dimensional probability distribution function was found from the lesion locations in the cranio-caudal and medio-lateral radiographic projections for calcification and noncalcification lesions. Lesion locations sampled from this probability distribution function were mapped on to individual mammograms using a piecewise affine transform which transforms the average outline to the outline of the breast in the mammogram.

The four dimensional probability distribution function was validated by comparing it to the two dimensional distributions found by considering each radiographic projection and laterality independently. The correlation of the location of the lesions sampled from the four dimensional probability distribution function across radiographic projections was shown to match the correlation of the locations of the original mapped lesion locations.

The current system has been implemented as a web-service on a server using the Python Django framework. The server performs the sampling, performs the mapping and returns the results in a javascript object notation format.

Advances in medical knowledge give clinicians more objective information for a diagnosis. Therefore, there is an increasing need for bibliographic search engines that can provide services helping to facilitate faster information search.

The ImageCLEFmed benchmark proposes a medical case-based retrieval task. This task aims at retrieving articles from the biomedical literature that are relevant for differential diagnosis of query cases including a textual description and several images. In the context of this campaign many approaches have been investigated showing that the fusion of visual and text information can improve the precision of the retrieval. However, fusion does not always lead to better results.

In this paper, a new query-adaptive fusion criterion to decide when to use multi-modal (text and visual) or only text approaches is presented. The proposed method integrates text information contained in MeSH (Medical Subject Headings) terms extracted and visual features of the images to find synonym relations between them. Given a text query, the query-adaptive fusion criterion decides when it is suitable to also use visual information for the retrieval.

Results show that this approach can decide if a text or multi{modal approach should be used with 77.15% of accuracy.

We had designed a semantic searching engine (SSE) for radiological imaging to search both reports and images in RIS-integrated PACS environment. In this presentation, we present evaluation results of this SSE about how it impacting the radiologists’ behaviors in reporting for different kinds of examinations, and how it improving the performance of retrieval and usage of historical images in RIS-integrated PACS.

This paper presents a novel approach to biomedical image retrieval by mapping image regions to local concepts and represent images in a weighted entropy-based concept feature space. The term concept refers to perceptually distinguishable visual patches that are identified locally in image regions and can be mapped to a glossary of imaging terms. Further, the visual significance (e.g., visualness) of concepts is measured as Shannon entropy of pixel values in image patches and is used to refine the feature vector. Moreover, the system can assist user in interactively select a Region-Of-Interest (ROI) and search for similar image ROIs. Further, a spatial verification step is used as a post-processing step to improve retrieval results based on location information. The hypothesis that such approaches would improve biomedical image retrieval, is validated through experiments on a data set of 450 lung CT images extracted from journal articles from four different collections.

Retained surgical items (RSIs) in patients is a major operating room (OR) patient safety concern. An RSI is any surgical tool, sponge, needle or other item inadvertently left in a patients body during the course of surgery. If left undetected, RSIs may lead to serious negative health consequences such as sepsis, internal bleeding, and even death. To help physicians efficiently and effectively detect RSIs, we are developing computer-aided detection (CADe) software for X-ray (XR) image analysis, utilizing large amounts of currently available image data to produce a clinically effective RSI detection system. Physician analysis of XRs for the purpose of RSI detection is a relatively lengthy process that may take up to 45 minutes to complete. It is also error prone due to the relatively low acuity of the human eye for RSIs in XR images. The system we are developing is based on computer vision and machine learning algorithms. We address the problem of low incidence by proposing synthesis algorithms. The CADe software we are developing may be integrated into a picture archiving and communication system (PACS), be implemented as a stand-alone software application, or be integrated into portable XR machine software through application programming interfaces. Preliminary experimental results on actual XR images demonstrate the effectiveness of the proposed approach.

We have developed a comprehensive DICOM RT specific database of retrospective treatment planning data for radiation therapy of head and neck cancer. Further, we have designed and built an imaging informatics module that utilizes this database to perform data mining. The end-goal of this data mining system is to provide radiation therapy decision support for incoming head and neck cancer patients, by identifying best practices from previous patients who had the most similar tumor geometries. Since the performance of such systems often depends on the size and quality of the retrospective database, we have also placed an emphasis on developing infrastructure and strategies to encourage data sharing and participation from multiple institutions. The infrastructure and decision support algorithm have both been tested and evaluated with 51 sets of retrospective treatment planning data of head and neck cancer patients. We will present the overall design and architecture of our system, an overview of our decision support mechanism as well as the results of our evaluation.

Super resolution optical microscopy allows the capture of images with a higher resolution than the diffraction limit. However, due to the lack of a standard format, the processing, visualization, transfer, and exchange of Super Resolution Optical Microscope (SROM) images are inconvenient. In this work, we present an approach to standardize the SROM images based on the Digital Imaging and Communication in Medicine (DICOM) standard. The SROM images and associated information are encapsulated and converted to DICOM images based on the Visible Light Microscopic Image Information Object Definition of DICOM. The new generated SROM images in DICOM format can be displayed, processed, transferred, and exchanged by using most medical image processing tools.

In this presentation, we present a web-based 3D medical image visualization solution which enables interactive large medical image data processing and visualization over the web platform. To improve the efficiency of our solution, we adopt GPU accelerated techniques to process images on the server side while rapidly transferring images to the HTML5 supported web browser on the client side. Compared to traditional local visualization solution, our solution doesn’t require the users to install extra software or download the whole volume dataset from PACS server. By designing this web-based solution, it is feasible for users to access the 3D medical image visualization service wherever the internet is available.

In previous years we demonstrated an imaging informatics system designed to support multi-institutional research focused on the utilization of proton radiation for treating spinal cord injury (SCI)-related pain. This year we will demonstrate an update on the system with new modules added to perform image processing on evaluation data using immunhistochemistry methods to observe effects of proton therapy. The overarching goal of the research is to determine the effectiveness of using the proton beam for treating SCI-related neuropathic pain as an alternative to invasive surgical lesioning. The research is a joint collaboration between three major institutes, University of Southern California (data collection/integration and image analysis), Spinal Cord Institute VA Healthcare System, Long Beach (patient subject recruitment), and Loma Linda University and Medical Center (human and preclinical animal studies). The system that we are presenting is one of its kind which is capable of integrating a large range of data types, including text data, imaging data, DICOM objects from proton therapy treatment and pathological data. For multi-institutional studies, keeping data secure and integrated is very crucial. Different kinds of data within the study workflow are generated at different stages and different groups of people who process and analyze them in order to see hidden patterns within healthcare data from a broader perspective. The uniqueness of our system relies on the fact that it is platform independent and web-based which makes it very useful in such a large-scale study.

Radiological imaging is fundamental within the healthcare industry and has become routinely adopted for diagnosis, disease monitoring and treatment planning. With the advent of digital imaging modalities and the rapid growth in both diagnostic and therapeutic imaging, the ability to be able to harness this large influx of data is of paramount importance. The Oncology Medical Image Database (OMI-DB) was created to provide a centralized, fully annotated dataset for research. The database contains both processed and unprocessed images, associated data, and annotations and where applicable expert determined ground truths describing features of interest. Medical imaging provides the ability to detect and localize many changes that are important to determine whether a disease is present or a therapy is effective by depicting alterations in anatomic, physiologic, biochemical or molecular processes. Quantitative imaging features are sensitive, specific, accurate and reproducible imaging measures of these changes.

Here, we describe an extension to the OMI-DB whereby a range of imaging features and descriptors are pre-calculated using a high throughput approach. The ability to calculate multiple imaging features and data from the acquired images would be valuable and facilitate further research applications investigating detection, prognosis, and classification. The resultant data store contains more than 10 million quantitative features as well as features derived from CAD predictions. Theses data can be used to build predictive models to aid image classification, treatment response assessment as well as to identify prognostic imaging biomarkers.

Our goal was to investigate the feasibility of an open-source, PACS-integrated, DICOM header-based tool that automatically provides granular data for monitoring of CT radiation exposure. To do so, we constructed a radiation exposure extraction engine (RE3) that is seamlessly connected to the PACS using the digital imaging and communications in medicine (DICOM) toolkit (DCMTK) and runs on the fly within the workflow. We evaluated RE3’s ability to determine the number of acquisitions and calculate the exposure metric dose length product (DLP) by comparing its output to the vendor dose pages. RE3 output closely correlated to the dose pages for both contiguously acquired exams (R² =0.9987) and non-contiguously acquired exams (R² =0.9994). RE3 is an open-source, automated radiation monitoring program to provide study-, series-, and slice-level radiation data.

In order to enable multiple disciplines of medical researchers, clinical physicians and biomedical engineers working together in a secured, efficient, and transparent cooperative environment, we had designed an e-Science platform for biomedical imaging research and application cross multiple academic institutions and hospitals in Shanghai and presented this work in SPIE Medical Imaging conference held in San Diego in 2012. In past the two-years, we implemented a biomedical image chain including communication, storage, cooperation and computing based on this e-Science platform. In this presentation, we presented the operating status of this system in supporting biomedical imaging research, analyzed and discussed results of this system in supporting multi-disciplines collaboration cross-multiple institutions.

Diffusion weighted magnetic resonance images(DW-MRI) such as diffusion tensor imaging(DTI) and diffusion kurtosis imaging(DKI) are widely used in understanding the complex cellular microstructures non-invasively. With the increased usage of DTI and DKI in the recent years, the need for software packages for processing magnetic resonance(MR) diffusion data has also gained much importance. We have developed a new graphical toolkit named 'DTI-DKI Fitting' which is an interactive software for processing diffusion MR images is presented for the first time. The features included in this toolbox are processing of 4D diffusion weighted data in formats such as Dicom and NIFTI, estimation of diffusion tensor and diffusion kurtosis parametric maps and visualization of those parametric maps. The toolbox is developed in Matlab as a stand alone application and main advantage being simple with minimal functionalities and user friendly. The functionalities of the toolbox are tested with multiple DW MR data acquired from normal subjects.