The Certification for Imaging Informatics Professionals (CIIP) program is sponsored by the Society of Imaging Informatics in Medicine and the American Registry of Radiologic Technologists through the American Board of Imaging Informatics. In 2005, a survey was conducted of radiologists, technologists, information technology specialists, corporate information officers, and radiology administrators to identify the competencies and skill set that would define a successful PACS administrator. The CIIP examination was created in 2007 in response to the need for an objective way to test for such competencies, and there have been 767 professionals who have been certified through this program to date. The validity of the psychometric integrity of the examination has been previously established. In order to further understand the impact and future direction of the CIIP certification on diplomats, a survey was conducted in 2010. This paper will discuss the results of the survey.
PACS administration; Informatics training; Medical informatics applications
Digital Imaging and Communications in Medicine (DICOM) is the dominant standard for medical imaging data. DICOM-compliant devices and the data they produce are generally designed for clinical use and often do not match the needs of users in research or clinical trial settings. DicomBrowser is software designed to ease the transition between clinically oriented DICOM tools and the specialized workflows of research imaging. It supports interactive loading and viewing of DICOM images and metadata across multiple studies and provides a rich and flexible system for modifying DICOM metadata. Users can make ad hoc changes in a graphical user interface, write metadata modification scripts for batch operations, use partly automated methods that guide users to modify specific attributes, or combine any of these approaches. DicomBrowser can save modified objects as local files or send them to a DICOM storage service using the C-STORE network protocol. DicomBrowser is open-source software, available for download at http://nrg.wustl.edu/software/dicom-browser.
Digital imaging and communications in medicine (DICOM); Workflow; Image viewer; Imaging informatics
In this paper, a novel lesion segmentation within breast ultrasound (BUS) image based on the cellular automata principle is proposed. Its energy transition function is formulated based on global image information difference and local image information difference using different energy transfer strategies. First, an energy decrease strategy is used for modeling the spatial relation information of pixels. For modeling global image information difference, a seed information comparison function is developed using an energy preserve strategy. Then, a texture information comparison function is proposed for considering local image difference in different regions, which is helpful for handling blurry boundaries. Moreover, two neighborhood systems (von Neumann and Moore neighborhood systems) are integrated as the evolution environment, and a similarity-based criterion is used for suppressing noise and reducing computation complexity. The proposed method was applied to 205 clinical BUS images for studying its characteristic and functionality, and several overlapping area error metrics and statistical evaluation methods are utilized for evaluating its performance. The experimental results demonstrate that the proposed method can handle BUS images with blurry boundaries and low contrast well and can segment breast lesions accurately and effectively.
Breast neoplasm; Image segmentation; Ultrasound; Cellular automata
The digital imaging and communications in medicine (DICOM) 3.0 standard was first officially ratified by the national electrical manufacturers association in 1993. The success of the DICOM open standard cannot be overstated in its ability to enable an explosion of innovation in the best of breed picture archiving and communication systems (PACS) industry. At the heart of the success of allowing interoperability between disparate systems have been three fundamental DICOM operations: C-MOVE, C-FIND, and C-STORE. DICOM C-MOVE oversees the transfer of DICOM Objects between two systems using C-STORE. DICOM C-FIND negotiates the ability to discover DICOM objects on another node. This paper will discuss the efforts within the DICOM standard to adapt this core functionality to Internet standards. These newer DICOM standards look to address the next generation of PACS challenges including highly distributed mobile acquisition systems and viewing platforms.
Web technology; Wide area network (WAN); Systems integration; PACS integration; Image distribution; Integrating healthcare enterprise (IHE); Internet technology; Enterprise PACS; Digital imaging and communications in medicine (DICOM)
In part one of this series, best practices were described for acquiring and handling data at study sites and importing them into an image repository or database. Here, we present a similar treatment on data management practices for imaging-based studies.
Clinical trials; Research image database
Mammography is the most efficient technique for detecting and diagnosing breast cancer. Clusters of microcalcifications have been mainly targeted as a reliable early sign of breast cancer and their earliest detection is essential to reduce the probability of mortality rate. Since the size of microcalcifications is very tiny and may be overlooked by the observing radiologist, we have developed a Computer Aided Diagnosis system for automatic and accurate cluster detection. A three-phased novel approach is presented in this paper. Firstly, regions of interest that corresponds to microcalcifications are identified. This can be achieved by analyzing the bandpass coefficients of the mammogram image. The suspicious regions are passed to the second phase, in which the nodular structured microcalcifications are detected based on eigenvalues of second order partial derivatives of the image and microcalcification pixels are segmented out by exploiting the foveal segmentation in multiscale analysis. Finally, by combining the responses coming out from the second order partial derivatives and the foveal method, potential microcalcifications are detected. The detection performance of the proposed method has been evaluated by using 370 mammograms. The detection method has a TP ratio of 97.76 % with 0.68 false positives per image. We have examined the performance of our computerized scheme using free-response operating characteristics curve.
Breast cancer; Computer Aided Diagnosis; Hessian matrix; Foveal segmentation
Although mammography is the only clinically accepted imaging modality for screening the general population to detect breast cancer, interpreting mammograms is difficult with lower sensitivity and specificity. To provide radiologists “a visual aid” in interpreting mammograms, we developed and tested an interactive system for computer-aided detection and diagnosis (CAD) of mass-like cancers. Using this system, an observer can view CAD-cued mass regions depicted on one image and then query any suspicious regions (either cued or not cued by CAD). CAD scheme automatically segments the suspicious region or accepts manually defined region and computes a set of image features. Using content-based image retrieval (CBIR) algorithm, CAD searches for a set of reference images depicting “abnormalities” similar to the queried region. Based on image retrieval results and a decision algorithm, a classification score is assigned to the queried region. In this study, a reference database with 1,800 malignant mass regions and 1,800 benign and CAD-generated false-positive regions was used. A modified CBIR algorithm with a new function of stretching the attributes in the multi-dimensional space and decision scheme was optimized using a genetic algorithm. Using a leave-one-out testing method to classify suspicious mass regions, we compared the classification performance using two CBIR algorithms with either equally weighted or optimally stretched attributes. Using the modified CBIR algorithm, the area under receiver operating characteristic curve was significantly increased from 0.865 ± 0.006 to 0.897 ± 0.005 (p < 0.001). This study demonstrated the feasibility of developing an interactive CAD system with a large reference database and achieving improved performance.
Computer-aided detection and diagnosis (CAD); Content-based image retrieval (CBIR); Breast cancer; Mammograms
Breast ultrasound (BUS) image segmentation is a very difficult task due to poor image quality and speckle noise. In this paper, local features extracted from roughly segmented regions of interest (ROIs) are used to describe breast tumors. The roughly segmented ROI is viewed as a bag. And subregions of the ROI are considered as the instances of the bag. Multiple-instance learning (MIL) method is more suitable for classifying breast tumors using BUS images. However, due to the complexity of BUS images, traditional MIL method is not applicable. In this paper, a novel MIL method is proposed for solving such task. First, a self-organizing map is used to map the instance space to the concept space. Then, we use the distribution of the instances of each bag in the concept space to construct the bag feature vector. Finally, a support vector machine is employed for classifying the tumors. The experimental results show that the proposed method can achieve better performance: the accuracy is 0.9107 and the area under receiver operator characteristic curve is 0.96 (p < 0.005).
Multiple-instance learning (MIL); Breast ultrasound (BUS) image; SVM (support vector machine); Classification
Tablet computers such as the iPad, which have a large format, improved graphic display resolution and a touch screen interface, may have an advantage compared to existing mobile devices such as smartphones and laptops for viewing radiological images. We assessed their potential for emergency radiology teleconsultation by reviewing multi-image CT and MRI studies on iPad tablet computers compared to Picture Archival and Communication Systems (PACS) workstations. Annonymised DICOM images of 79 CT and nine MRI studies comprising a range of common on-call conditions, reported on full-featured diagnostic PACS workstation by one Reporting Radiologist, were transferred from PACS to three iPad tablet computers running OsiriX HD v 2.02 DICOM software and viewed independently by three reviewing radiologists. Structured documentation was made of major findings (primary diagnosis or other clinically important findings), minor findings (incidental findings), and user feedback. Two hundred and sixty four readings (88 studies read by three reviewing radiologists) were compared, with 3.4 % (nine of 264) major discrepancies and 5.6 % (15 of 264) minor discrepancies. All reviewing radiologists reported favorable user experience but noted issues with software stability and limitations of image manipulation tools. Our results suggest that emergency conditions commonly encountered on CT and MRI can be diagnosed using tablet computers with good agreement with dedicated PACS workstations. Shortcomings in software and application design should be addressed if the potential of tablet computers for mobile teleradiology is to be fully realized.
iPad; Tablet computer; CT; MRI; Emergency radiology; Teleradiology
While previous research has determined the contrast detection threshold in medical images, it has focused on uniform backgrounds, has not used calibrated monitors, or has involved a low number of readers. With complex clinical images, how the Grayscale Standard Display Function (GSDF) affects the detection threshold and whether the median background intensity shift has been minimized by GSDF remains unknown. We set out to determine if the median background affected the detection of a low-contrast object in a clustered lumpy background, which simulated a mammography image, and to define the contrast detection threshold for these complex images. Clustered lumpy background images were created of different median intensities and disks of varying contrasts were inserted. A reader study was performed with 17 readers of varying skill level who scored with a five-point confidence scale whether a disk was present. The results were analyzed using reader operating characteristic (ROC) methodology. Contingency tables were used to determine the contrast detection threshold. No statistically significant difference was seen in the area under the ROC curve across all of the backgrounds. Contrast detection fell below 50 % between +3 and +2 gray levels. Our work supports the conclusion that Digital Imaging and Communications in Medicine GSDF calibrated monitors do perceptually linearize detection performance across shifts in median background intensity. The contrast detection threshold was determined to be +3 gray levels above the background for an object of 1° visual angle.
Image perception; ROC-based analysis; Digital display; Contrast threshold; GSDF
The use of clinical imaging modalities within the pharmaceutical research space provides value and challenges. Typical clinical settings will utilize a Picture Archive and Communication System (PACS) to transmit and manage Digital Imaging and Communications in Medicine (DICOM) images generated by clinical imaging systems. However, a PACS is complex and provides many features that are not required within a research setting, making it difficult to generate a business case and determine the return on investment. We have developed a next-generation DICOM processing system using open-source software, commodity server hardware such as Apple Xserve®, high-performance network-attached storage (NAS), and in-house-developed preprocessing programs. DICOM-transmitted files are arranged in a flat file folder hierarchy easily accessible via our downstream analysis tools and a standard file browser. This next-generation system had a minimal construction cost due to the reuse of all the components from our first-generation system with the addition of a second server for a few thousand dollars. Performance metrics were gathered and the system was found to be highly scalable, performed significantly better than the first-generation system, is modular, has satisfactory image integrity, and is easier to maintain than the first-generation system. The resulting system is also portable across platforms and utilizes minimal hardware resources, allowing for easier upgrades and migration to smaller form factors at the hardware end-of-life. This system has been in production successfully for 8 months and services five clinical instruments and three pre-clinical instruments. This system has provided us with the necessary DICOM C-Store functionality, eliminating the need for a clinical PACS for day-to-day image processing.
PACS; DCMTK; DICOM; DICOM workflow; DICOM storage
To determine which Breast Imaging Reporting and Data System (BI-RADS) descriptors for ultrasound are predictors for breast cancer using logistic regression (LR) analysis in conjunction with interobserver variability between breast radiologists, and to compare the performance of artificial neural network (ANN) and LR models in differentiation of benign and malignant breast masses. Five breast radiologists retrospectively reviewed 140 breast masses and described each lesion using BI-RADS lexicon and categorized final assessments. Interobserver agreements between the observers were measured by kappa statistics. The radiologists’ responses for BI-RADS were pooled. The data were divided randomly into train (n = 70) and test sets (n = 70). Using train set, optimal independent variables were determined by using LR analysis with forward stepwise selection. The LR and ANN models were constructed with the optimal independent variables and the biopsy results as dependent variable. Performances of the models and radiologists were evaluated on the test set using receiver-operating characteristic (ROC) analysis. Among BI-RADS descriptors, margin and boundary were determined as the predictors according to stepwise LR showing moderate interobserver agreement. Area under the ROC curves (AUC) for both of LR and ANN were 0.87 (95% CI, 0.77–0.94). AUCs for the five radiologists ranged 0.79–0.91. There was no significant difference in AUC values among the LR, ANN, and radiologists (p > 0.05). Margin and boundary were found as statistically significant predictors with good interobserver agreement. Use of the LR and ANN showed similar performance to that of the radiologists for differentiation of benign and malignant breast masses.
Breast; Ultrasonography; Artificial neural network; Breast neoplasm; Logistic regression
The Digital Imaging and Communications in Medicine (DICOM) is the standard for encoding and communicating medical imaging information. It is used in radiology as well as in many other imaging domains such as ophthalmology, dentistry, and pathology. DICOM information objects are used to encode medical images or information about the images. Their usage outside of the imaging department is increasing, especially with the sharing of medical images within Electronic Health Record systems. However, learning DICOM is long and difficult because it defines and uses many specific abstract concepts that relate to each other. In this paper, we present an approach, based on problem solving, for teaching DICOM as part of a graduate course on healthcare information. The proposed approach allows students with diversified background and no software development experience to grasp a large breadth of knowledge in a very short time.
Digital Imaging and Communications in Medicine; DICOM; Medical imaging; Healthcare information; Teaching; Problem solving
The purpose of this study was to demonstrate the robustness of our prior computerized texture analysis method for breast cancer risk assessment, which was developed initially on a limited dataset of screen-film mammograms. This current study investigated the robustness by (1) evaluating on a large clinical dataset, (2) using full-field digital mammograms (FFDM) as opposed to screen-film mammography, and (3) incorporating analyses over two types of high-risk patient sets, as well as patients at low risk for breast cancer. The evaluation included the analyses on the parenchymal patterns of women at high risk of developing of breast cancer, including both BRCA1/2 gene mutation carriers and unilateral cancer patients, and of women at low risk of developing breast cancer. A total of 456 cases, including 53 women with BRCA1/2 gene mutations, 75 women with unilateral cancer, and 328 low-risk women, were retrospectively collected under an institutional review board approved protocol. Regions-of-interest (ROIs), were manually selected from the central breast region immediately behind the nipple. These ROIs were subsequently used in computerized feature extraction to characterize the mammographic parenchymal patterns in the images. Receiver operating characteristic analysis was used to assess the performance of the computerized texture features in the task of distinguishing between high-risk and low-risk subjects. In a round robin evaluation on the FFDM dataset with Bayesian artificial neural network analysis, AUC values of 0.82 (95% confidence interval [0.75, 0.88]) and 0.73 (95% confidence interval [0.67, 0.78]) were obtained between BRCA1/2 gene mutation carriers and low-risk women, and between unilateral cancer and low-risk women, respectively. These results from computerized texture analysis on digital mammograms demonstrated that high-risk and low-risk women have different mammographic parenchymal patterns. On this large clinical dataset, we validated our methods for quantitative analyses of mammographic patterns on FFDM, statistically demonstrating again that women at high risk tend to have dense breasts with coarse and low-contrast texture patterns.
Computerized texture analysis; Breast cancer risk assessment; Mammographic parenchymal patterns; Full-field digital mammograms; Quantitative imaging analysis
Preclinical medical education; Digital teaching files; Web technology; MIRC; Radiology teaching files; Learning management systems; BlackBoard Learn
Advances in handheld computing now allow review of DICOM datasets from remote locations. As the diagnostic ability of this tool is unproven, we evaluated the ability to diagnose acute appendicitis on abdominal CT using a mobile DICOM viewer. This HIPAA compliant study was IRB-approved. Twenty-five abdominal CT studies from patients with RLQ pain were interpreted on a handheld device (iPhone) using a DICOM viewer (OsiriX mobile) by five radiologists. All patients had surgical confirmation of acute appendicitis or follow-up confirming no acute appendicitis. Studies were evaluated for the ability to find the appendix, maximum appendiceal diameter, presence of an appendicolith, periappendiceal stranding and fluid, abscess, and an assessment of the diagnosis of acute appendicitis. Results were compared to PACS workstation. Fifteen cases of acute appendicitis were correctly identified on 98% of interpretations, with no false-positives. Eight appendicoliths were correctly identified on 88% of interpretations. Three abscesses were correctly identified by all readers. Handheld device measurement of appendiceal diameter had a mean 8.6% larger than PACS measurements (p = 0.035). Evaluation for acute appendicitis on abdominal CT studies using a portable device DICOM viewer can be performed with good concordance to reads performed on PACS workstations.
Appendicitis; Computed tomography; Gastrointestinal; Mobile; Teleradiology
Increasing radiology studies has led to the emergence of new requirements for management medical information, mainly affecting the storage of digital images. Today, it is a necessary interaction between workflow management and legal rules that govern it, to allow an efficient control of medical technology and associated costs. Another important topic that is growing in importance within the healthcare sector is compliance, which includes the retention of studies, information security, and patient privacy. Previously, we conducted a series of extensive analysis and measurements of pre-existing operating conditions. These studies and projects have been described in other papers. The first phase: hardware and software installation and initial tests were completed in March 2006. The storage phase was built step by step until the PACS-INR was totally completed. Two important aspects were considered in the integration of components: (1) the reliability and performance of the system to transfer and display DICOM images, and (2) the availability of data backups for disaster recovery and downtime scenarios. This paper describes the high-availability model for a large-scale PACS to support the storage and retrieve of data using CAS and DAS technologies to provide an open storage platform. This solution offers a simple framework that integrates and automates the information at low cost and minimum risk. Likewise, the model allows an optimized use of the information infrastructure in the clinical environment. The tests of the model include massive data migration, openness, scalability, and standard compatibility to avoid locking data into a proprietary technology.
Archive; Information storage and retrieval; Database management systems; Digital image management; Enterprise PACS
This paper presents novel multiple keywords annotation for medical images, keyword-based medical image retrieval, and relevance feedback method for image retrieval for enhancing image retrieval performance. For semantic keyword annotation, this study proposes a novel medical image classification method combining local wavelet-based center symmetric–local binary patterns with random forests. For keyword-based image retrieval, our retrieval system use the confidence score that is assigned to each annotated keyword by combining probabilities of random forests with predefined body relation graph. To overcome the limitation of keyword-based image retrieval, we combine our image retrieval system with relevance feedback mechanism based on visual feature and pattern classifier. Compared with other annotation and relevance feedback algorithms, the proposed method shows both improved annotation performance and accurate retrieval results.
Image annotation; Random forests; Confidence score; Body relation graph; Relevance feedback
The measurement of angles between anatomical structures is common in radiological and orthopedic practice. Frequently used measurements include scapholunate angle for assessment of wrist instability and Cobb’s angle used for assessment of scoliosis. Measurements of these angles are easily performed on plain X-ray radiographs. However, the situation is more complicated when these measurements are to be performed on cross-sectional (CT or MRI) examinations. On some of the diagnostic workstations, it is not possible to perform angle measurements between the structures if they are not identified on the same image and are located on different images of the same projection or plane. We present a simple solution to measure angles between structures on different images that can be used both in CT and MR.
Computed tomography; Image viewer; Magnetic resonance imaging
In many medical imaging applications, it is desirable and important to localize and remove the patient table from CT images. However, existing methods often require user interactions to define the table and sometimes make inaccurate assumptions about the table shape. Due to different patient table designs, shapes, and characteristics, these methods are not robust in identifying and removing the patient table. This paper proposes a new automatic approach which first identifies and locates the patient table in the sagittal planes and then removes it from the axial planes. The method has been tested successfully against different tables in different products from multiple vendors, showing it is both a versatile and robust technique for patient table removal.
Computed tomography; Patient table; Hough transform
In the filmless imaging department, an integrated imaging and reporting system is only as strong as its weakest link. An outage or downtime of a key segment, such as the Picture Archive Communications System (PACS), is a significant threat to efficient workflow, quality of image interpretation, ordering clinician’s review, and ultimately patient care. A multidisciplinary team (including physicists, technologists, radiologists, operations, and IT) developed a backup system to provide business continuity (i.e., quality control, interpretation, reporting, and clinician access) during an extended outage of the main departmental PACS.
Computer hardware; Computer networks; Computers in medicine
Current speech recognition software allows exam-specific standard reports to be prepopulated into the dictation field based on the radiology information system procedure code. While it is thought that prepopulating reports can decrease the time required to dictate a study and the overall number of errors in the final report, this hypothesis has not been studied in a clinical setting. A prospective study was performed. During the first week, radiologists dictated all studies using prepopulated standard reports. During the second week, all studies were dictated after prepopulated reports had been disabled. Final radiology reports were evaluated for 11 different types of errors. Each error within a report was classified individually. The median time required to dictate an exam was compared between the 2 weeks. There were 12,387 reports dictated during the study, of which, 1,173 randomly distributed reports were analyzed for errors. There was no difference in the number of errors per report between the 2 weeks; however, radiologists overwhelmingly preferred using a standard report both weeks. Grammatical errors were by far the most common error type, followed by missense errors and errors of omission. There was no significant difference in the median dictation time when comparing studies performed each week. The use of prepopulated reports does not alone affect the error rate or dictation time of radiology reports. While it is a useful feature for radiologists, it must be coupled with other strategies in order to decrease errors.
Standardized report; Structured report; Prepopulated reports; Speech recognition; Turnaround time