Establishing and Running a Three-dimensional and Advanced Imaging Laboratory.

Multidetector CT technology has advanced during the past 2 decades from single-digit numbers of sections to 320-section CT scanners. The ability to perform three-dimensional (3D) postprocessing of acquired data has accompanied this technical progress. Multiple considerations are involved in developing and deploying a 3D and advanced imaging laboratory to provide postprocessing for both CT and MRI examinations. Establishing and running a 3D laboratory requires administrators to buy into the process and also requires regular input from radiologists and other stakeholders. Technologists with prior 3D experience are rare, and extensive immersive training is typically required. Laboratory space and equipment must be maintained and updated regularly to continue to meet stakeholders' needs. Postprocessing protocols must be established and reviewed periodically. Quality control is also necessary to ensure that postprocessing outputs adhere to the protocols. Laboratory technologists can also provide postprocessing support for research examinations and can lend their technical expertise to research projects. Regular review of laboratory productivity is essential to ensure longevity and availability of necessary resources (human, environmental, and technical). As the technologies for the imaging equipment, the picture archiving and communication system, and postprocessing continue to advance, the role of the 3D laboratory will also evolve to include other services. This evolution will affect ongoing training of technologists, as well as the requirements when new technologists are hired. The 3D laboratory is often positioned to take on novel technology and augment radiologists' workflow as new modalities and techniques enter the clinical workflow. ©RSNA, 2018.

Developing an automated database for monitoring ultrasound- and computed tomography-guided procedure complications and diagnostic yield.

Monitoring complications and diagnostic yield for image-guided procedures is an important component of maintaining high quality patient care promoted by professional societies in radiology and accreditation organizations such as the American College of Radiology (ACR) and Joint Commission. These outcome metrics can be used as part of a comprehensive quality assurance/quality improvement program to reduce variation in clinical practice, provide opportunities to engage in practice quality improvement, and contribute to developing national benchmarks and standards. The purpose of this article is to describe the development and successful implementation of an automated web-based software application to monitor procedural outcomes for US- and CT-guided procedures in an academic radiology department. The open source tools PHP: Hypertext Preprocessor (PHP) and MySQL were used to extract relevant procedural information from the Radiology Information System (RIS), auto-populate the procedure log database, and develop a user interface that generates real-time reports of complication rates and diagnostic yield by site and by operator. Utilizing structured radiology report templates resulted in significantly improved accuracy of information auto-populated from radiology reports, as well as greater compliance with manual data entry. An automated web-based procedure log database is an effective tool to reliably track complication rates and diagnostic yield for US- and CT-guided procedures performed in a radiology department.

An interactive RADIANCE toolkit for customizable CT dose monitoring and reporting.

The need for tools to monitor imaging-related radiation has grown dramatically in recent years. RADIANCE, a freely available open-source dose-monitoring tool, was developed in response to the need for an informatics solution in this realm. A number of open-source as well as commercial solutions have since been developed to enable radiology practices to monitor radiation dose parameters for modalities ranging from computed tomography to radiography to fluoroscopy. However, it is not sufficient to simply collect this data; it is equally important to be able to review it in the appropriate context. Most of the currently available dose-monitoring solutions have some type of reporting capability, such as a real-time dashboard or a static report. Previous versions of RADIANCE have included a real-time dashboard with pre-set screens that plot effective dose estimates according to different criteria, as well as monthly scorecards to summarize dose estimates for individuals within a radiology practice. In this work, we present the RADIANCE toolkit, a customizable reporting solution that allows users to generate reports of interest to them, summarizing a variety of metrics that can be grouped according to useful parameters. The output of the toolkit can be used for real-time dose monitoring or scheduled reporting, such as to a quality assurance committee. Making dose parameter data more accessible and more meaningful to the user promotes dose reduction efforts such as regular protocol review and optimization, and ultimately improves patient care by decreasing unnecessary radiation exposure.

Using the Microsoft Kinect for patient size estimation and radiation dose normalization: proof of concept and initial validation.

Monitoring patients' imaging-related radiation is currently a hot topic, but there are many obstacles to accurate, patient-specific dose estimation. While some, such as easier access to dose data and parameters, have been overcome, the challenge remains as to how accurately these dose estimates reflect the actual dose received by the patient. The main parameter that is often not considered is patient size. There are many surrogates-weight, body mass index, effective diameter-but none of these truly reflect the three-dimensional "size" of an individual. In this work, we present and evaluate a novel approach to estimating patient volume using the Microsoft Kinect™, a combination RGB camera-infrared depth sensor device. The goal of using this device is to generate a three-dimensional estimate of patient size, in order to more effectively model the dimensions of the anatomy of interest and not only enable better normalization of dose estimates but also promote more patient-specific protocoling of future CT examinations. Preliminary testing and validation of this system reveals good correlation when individuals are standing upright with their arms by their sides, but demonstrates some variation with arm position. Further evaluation and testing is necessary with multiple patient positions and in both adult and pediatric patients. Correlation with other patient size metrics will also be helpful, as the ideal measure of patient "size" may in fact be a combination of existing metrics and newly developed techniques.

A modern experience with saccular aortic aneurysms.

OBJECTIVE: Repair of saccular aortic aneurysms (SAAs) is frequently recommended based on a perceived predisposition to rupture, despite little evidence that these aneurysms have a more malignant natural history than fusiform aortic aneurysms. METHODS: The radiology database at a single university hospital was searched for the computed tomographic (CT) diagnosis of SAA between 2003 and 2011. Patient characteristics and clinical course, including the need for surgical intervention, were recorded. SAA evolution was assessed by follow-up CT, where available. Multivariate analysis was used to examine potential predictors of aneurysm growth rate. RESULTS: Three hundred twenty-two saccular aortic aneurysms were identified in 284 patients. There were 153 (53.7%) men and 131 women with a mean age of 73.5±10.0 years. SAAs were located in the ascending aorta in two (0.6%) cases, the aortic arch in 23 (7.1%), the descending thoracic aorta in 219 (68.1%), and the abdominal aorta in 78 (24.2%). One hundred thirteen (39.8%) patients underwent surgical repair of SAA. Sixty-two patients (54.9%) underwent thoracic endovascular aortic repair, 22 underwent endovascular aneurysm repair (19.5%), and 29 (25.6%) required open surgery. The average maximum diameter of SAA was 5.0±1.6 cm. In repaired aneurysms, the mean diameter was 5.4±1.4 cm; in unrepaired aneurysms, it was 4.4±1.1 cm (P<.001). Eleven patients (3.9%) had ruptured SAAs on initial scan. Of the initial 284 patients, 50 patients (with 54 SAA) had CT follow-up after at least 3 months (23.2±19.0 months). Fifteen patients (30.0%) ultimately underwent surgical intervention. Aneurysm growth rate was 2.8±2.9 mm/yr, and was only weakly related to initial aortic diameter (R2=.19 by linear regression, P=.09 by multivariate regression). Decreased calcium burden (P=.03) and increased patient age (P=.05) predicted increased aneurysm growth by multivariate analysis. CONCLUSIONS: While SAA were not found to have a higher growth rate than their fusiform counterparts, both clinical and radiologic follow-up is necessary, as a significant number ultimately require surgical intervention. Further clinical research is necessary to determine the optimal management of SAA.

An algorithm for intelligent sorting of CT-related dose parameters.

Imaging centers nationwide are seeking innovative means to record and monitor computed tomography (CT)-related radiation dose in light of multiple instances of patient overexposure to medical radiation. As a solution, we have developed RADIANCE, an automated pipeline for extraction, archival, and reporting of CT-related dose parameters. Estimation of whole-body effective dose from CT dose length product (DLP)--an indirect estimate of radiation dose--requires anatomy-specific conversion factors that cannot be applied to total DLP, but instead necessitate individual anatomy-based DLPs. A challenge exists because the total DLP reported on a dose sheet often includes multiple separate examinations (e.g., chest CT followed by abdominopelvic CT). Furthermore, the individual reported series DLPs may not be clearly or consistently labeled. For example, "arterial" could refer to the arterial phase of the triple liver CT or the arterial phase of a CT angiogram. To address this problem, we have designed an intelligent algorithm to parse dose sheets for multi-series CT examinations and correctly separate the total DLP into its anatomic components. The algorithm uses information from the departmental PACS to determine how many distinct CT examinations were concurrently performed. Then, it matches the number of distinct accession numbers to the series that were acquired and anatomically matches individual series DLPs to their appropriate CT examinations. This algorithm allows for more accurate dose analytics, but there remain instances where automatic sorting is not feasible. To ultimately improve radiology patient care, we must standardize series names and exam names to unequivocally sort exams by anatomy and correctly estimate whole-body effective dose.

Development of automated detection of radiology reports citing adrenal findings.

The aim of this study was to determine the feasibility of automated detection of adrenal nodules, a common finding on CT, using a newly developed search engine that mines dictated radiology reports. To ensure Health Insurance Portability and Accountability Act compliance, we utilized a preexisting de-identified database of 32,974 CT reports from February 1, 2009 to February 28, 2010. Common adrenal descriptors from 29 staff radiologists were used to develop an automated rule-based algorithm targeting adrenal findings. Each sentence within the free text of reports was searched with an adapted NegEx negation algorithm. The algorithm was refined using a 2-week test period of reports and subsequently validated using a 6-week period. Manual review of the 3,693 CT reports in the validation period identified 222 positive reports while the algorithm detected 238 positive reports. The algorithm identified one true positive report missed on manual review for a total of 223 true positive reports. This resulted in a precision of 91% (217 of 238) and a recall of 97% (217 of 223). The sensitivity of the query was 97.3% (95% confidence interval (CI), 93.9-98.9%), and the specificity was 99.3% (95% CI, 99.1-99.6%). The positive predictive value of the algorithm was 91.0% (95% CI, 86.6-94.3%), and the negative predictive value was 99.8% (95% CI, 99.6-99.9%). The prevalence of true positive adrenal findings identified by the query (7.1%) was nearly identical to the true prevalence (7.2%). Automated detection of language describing common findings in imaging reports, such as adrenal nodules on CT, is feasible.

RADIANCE: An automated, enterprise-wide solution for archiving and reporting CT radiation dose estimates.

There is growing interest in the ability to monitor, track, and report exposure to radiation from medical imaging. Historically, however, dose information has been stored on an image-based dose sheet, an arrangement that precludes widespread indexing. Although scanner manufacturers are beginning to include dose-related parameters in the Digital Imaging and Communications in Medicine (DICOM) headers of imaging studies, there remains a vast repository of retrospective computed tomographic (CT) data with image-based dose sheets. Consequently, it is difficult for imaging centers to monitor their dose estimates or participate in the American College of Radiology (ACR) Dose Index Registry. An automated extraction software pipeline known as Radiation Dose Intelligent Analytics for CT Examinations (RADIANCE) has been designed that quickly and accurately parses CT dose sheets to extract and archive dose-related parameters. Optical character recognition of information in the dose sheet leads to creation of a text file, which along with the DICOM study header is parsed to extract dose-related data. The data are then stored in a relational database that can be queried for dose monitoring and report creation. RADIANCE allows efficient dose analysis of CT examinations and more effective education of technologists, radiologists, and referring physicians regarding patient exposure to radiation at CT. RADIANCE also allows compliance with the ACR's dose reporting guidelines and greater awareness of patient radiation dose, ultimately resulting in improved patient care and treatment.

Informatics in radiology: automated structured reporting of imaging findings using the AIM standard and XML.

Quantitative and descriptive imaging data are a vital component of the radiology report and are frequently of paramount importance to the ordering physician. Unfortunately, current methods of recording these data in the report are both inefficient and error prone. In addition, the free-text, unstructured format of a radiology report makes aggregate analysis of data from multiple reports difficult or even impossible without manual intervention. A structured reporting work flow has been developed that allows quantitative data created at an advanced imaging workstation to be seamlessly integrated into the radiology report with minimal radiologist intervention. As an intermediary step between the workstation and the reporting software, quantitative and descriptive data are converted into an extensible markup language (XML) file in a standardized format specified by the Annotation and Image Markup (AIM) project of the National Institutes of Health Cancer Biomedical Informatics Grid. The AIM standard was created to allow image annotation data to be stored in a uniform machine-readable format. These XML files containing imaging data can also be stored on a local database for data mining and analysis. This structured work flow solution has the potential to improve radiologist efficiency, reduce errors, and facilitate storage of quantitative and descriptive imaging data for research.

Use of a Dedicated Server to Perform Coronal and Sagittal Reformations in Trauma Examinations.

The purpose of this study was to evaluate the impact of implementing an automated process for generating coronal and sagittal reformatted images on radiologist workflow. When performing trauma-related CT examinations of the cervical, thoracic, and lumbar spine at our institution, technologists manually generate coronal and sagittal reconstructions at the scanner console and send these images to a picture archiving and communication system (PACS) for interpretation by radiologists and clinical viewing. Although certain PACS, thin-client three-dimensional systems, and CT scanners are capable of automatically generating reconstructed or reformatted images, the systems at our institution do not support this functionality. We have recently integrated a dedicated server that is capable of automatically generating multiplanar reformatted (MPR) images from source thin-section axial images and sending these images to PACS without requiring technologist input. This dedicated server was used to generate coronal and sagittal MPRs for trauma-related spine studies in parallel with technologist-generated coronal and sagittal reconstructions. When comparing the two methods, using the dedicated server to automatically generate reformations resulted in substantial time savings for the radiologist compared to technologist-generated reconstructions. Additionally, a survey of interpreting radiologists indicated that a significant majority preferred to view the automatically generated MPRs on PACS compared to the thin-client system, considered the image quality to be good or excellent, and believed that viewing MPRs increased diagnostic accuracy and confidence. It is expected that this automated process will significantly improve radiologist workflow with respect to image interpretation time and report turnaround time.

Automated extraction of radiation dose information for CT examinations.

Exposure to radiation as a result of medical imaging is currently in the spotlight, receiving attention from Congress as well as the lay press. Although scanner manufacturers are moving toward including effective dose information in the Digital Imaging and Communications in Medicine headers of imaging studies, there is a vast repository of retrospective CT data at every imaging center that stores dose information in an image-based dose sheet. As such, it is difficult for imaging centers to participate in the ACR's Dose Index Registry. The authors have designed an automated extraction system to query their PACS archive and parse CT examinations to extract the dose information stored in each dose sheet. First, an open-source optical character recognition program processes each dose sheet and converts the information to American Standard Code for Information Interchange (ASCII) text. Each text file is parsed, and radiation dose information is extracted and stored in a database which can be queried using an existing pathology and radiology enterprise search tool. Using this automated extraction pipeline, it is possible to perform dose analysis on the >800,000 CT examinations in the PACS archive and generate dose reports for all of these patients. It is also possible to more effectively educate technologists, radiologists, and referring physicians about exposure to radiation from CT by generating report cards for interpreted and performed studies. The automated extraction pipeline enables compliance with the ACR's reporting guidelines and greater awareness of radiation dose to patients, thus resulting in improved patient care and management.

Use of high-frequency jet ventilation for respiratory immobilization during coronary artery CT angiography.

Multidetector ECG-gated CT angiography permits imaging of structures such as the coronary arteries and pulmonary veins with peripheral administration of contrast media. Respiratory motion artifact limits the applicability of this technique in critically ill patients due to an inability to cooperate with prolonged breath holds necessary for quality images. A case in which high-frequency jet ventilation via an uncuffed tracheostomy tube in an unmedicated patient permitted respiratory immobilization sufficient to acquire diagnostic images, is presented.

Radiologist use of and perceived need for patient data access.

Given the increasing volume of radiological exams, the decreasing frequency of direct communication with the referring provider, and the distribution of patient data over many clinical systems, radiologists often do not have adequate clinical information at the time of interpretation. We have performed a survey of radiologists to determine the need and actual utilization of patient data at the time of image interpretation. Our findings demonstrate that most radiologists want more clinical information when interpreting images and that this information would impact their report, but they are discouraged by the time it takes to access this information. In addition, current mechanisms for monitoring necessary patient follow-up are inadequate.

Informatics in radiology (infoRAD): survey of personal digital assistant use in radiology.

There has been widespread adoption of personal digital assistants (PDAs) within medicine in recent years. However, information on the prevalence and usage of these devices among radiologists is limited. A survey was designed and mailed to randomly selected members of the Radiological Society of North America to determine the percentage of PDA users, their use patterns, and the types of applications that they would like to see in the future. The use patterns of attending radiologists were compared with those of trainees (residents and fellows). Overall usage was also compared with the relevant findings in two surveys of internal medicine users. It was found that slightly less than one-half of respondents used PDAs on a daily basis, a finding that was comparable to that in the internal medicine surveys. However, less than one-quarter of PDA users had radiology-specific applications installed on their devices, whereas a greater percentage of internal medicine users had software such as drug databases and clinical references on their PDAs. Radiology trainees had a higher rate of both PDA ownership and radiology application usage than did attending radiologists. It is likely that, as PDA hardware becomes more powerful, with higher display resolution, better wireless networking capabilities, and greater memory capacity, PDA ownership as well as radiology application usage will increase.

MR microscopy of normal human brain.

MR microscopy at 9.4T depicts the architecture of the brain in exquisite detail, including the individual laminae of the cortex, the individual nuclei of the basal ganglia, the thalami, subthalami and metathalami, and the orientations and relationship among the dominant nuclei and white matter tracts of the brain. The authors believe that these anatomic relations will ultimately be displayed in vivo as clinical MR scanners begin to operate at field strengths of 4.7T, 7T, and 8T. Then, those familiar with this anatomy will be able to interpret patient images with far greater sophistication.