The 2021 SIIM-FISABIO-RSNA Machine Learning COVID-19 Challenge: Annotation and Standard Exam Classification of COVID-19 Chest Radiographs.

We describe the curation, annotation methodology, and characteristics of the dataset used in an artificial intelligence challenge for detection and localization of COVID-19 on chest radiographs. The chest radiographs were annotated by an international group of radiologists into four mutually exclusive categories, including "typical," "indeterminate," and "atypical appearance" for COVID-19, or "negative for pneumonia," adapted from previously published guidelines, and bounding boxes were placed on airspace opacities. This dataset and respective annotations are available to researchers for academic and noncommercial use.

The RSNA Image Sharing Network.

In the era of health information exchanges, there are trade-offs to consider when sharing a patient's medical record among all providers that a patient might choose. Exchange among in-network partners on the same electronic medical records (EMR) and other integrated information systems is trivial. The patient identifier is common, as are the relevant departmental systems, to all providers. Difficulties arise when patient records including images (and reports) must be shared among different networks and even with the patients themselves. The National Institutes of Health (NIH) challenged Radiological Society of North America (RSNA) to develop a transport method that could supersede the need for physical media (for patients or other providers), replace point-to-point private networks among providers, and enable image exchange on an ad hoc basis between arbitrary health networks without long legal delays. In concert with the evolving US health care paradigm, patient engagement was to be fundamental. With Integrating Healthcare Enterprise's (IHE's) help, the challenge has been met with an operational system.

Aging and quality control of color LCDs for radiologic imaging.

Our practice has long been concerned with the effects of display quality, including color accuracy and matching among paired color displays. Three years of data have been collected on the historical behavior of color stability on our clinical displays. This has permitted an analysis of the color-aging behavior of those displays over that time. The results of that analysis show that all displays tend to yellow over time, but that they do so together. That is, neither the intra- nor inter-display color variances observed at initial deployment diverge over time as measured by a mean radial distance metric in color space (Commission Internationale d'Eclairage L', u', v' 1976). The consequence of this result is that color displays that are matched at deployment tend to remain matched over their lifetime even as they collectively yellow.

Image acquisition: ultrasound, computed tomography, and magnetic resonance imaging.

As the transition toward total digital image acquisition continues, radiology is transcending the current standard of two-dimensional (2-D) cross-sectional anatomic imaging to more complex models. Among these are three-dimensional (3-D) anatomic images, constructed either from a synthesis of traditional 2-D data sets, or directly from volumetrically acquired data. However, current trends are moving beyond mere anatomic imaging to include physiological data once mainly obtained via nuclear medicine. Recent magnetic resonance pulse sequences, in addition to Doppler and harmonic ultrasound methods, are providing insight into blood flow, oxygenation, and metabolite concentrations non-invasively. Through image registration techniques, these data (even from differing modalities) are being assembled into 2-D and 3-D "fusion" images that promise to revolutionize diagnosis. Furthermore, with improvements in miniaturization, reliability, speed, built-in intelligence, and ease of use, these new developments are finding their way into use by nonspecialists. For instance, a new hand-held ultrasound unit will likely become a common tool among emergency medical teams, military medical teams, and in NASA's manned space program. Portable computed tomography (CT) scanners are already being used in the operating room. The increasing sophistication of imaging instruments will bring about a complementary increase in ease of use for both scanning and data interpretation, bringing diagnostic imaging and therapeutic capabilities closer to the patient, rather than the converse.

Architecture of an image capable, Web-based, electronic medical record.

With each medical center department creating and maintaining its own patient care-related data, nursing and house staff may find it confusing to log into all the information systems necessary to achieve a global perspective of the patient's state. The Medical Information Network Database application provides a logically centralized Worldwide Web viewing application for the physically distributed data. In addition to coordinating data displays for histories, laboratories, pathology, radiology, and discharge summaries, the application can be configured to apply rule sets to the data and remind caregivers of follow-up tests or of possible reactions to treatment protocols. The viewing client runs on any HTML 2.0-compliant browser, although certain applet enhancements (notably for viewing radiological images) require a browser with Java abilities. With this "thin client" approach, the application can be configured to coexist with other applications (such as a PACS viewer), thus centralizing information and reducing the overall number of computers in the medical center.

Communicating with the family: the risks of medical radiation to conceptuses in victims of major blunt-force torso trauma.

BACKGROUND: Trauma surgeons must balance the risk and benefits of diagnostic radiographic procedures on potentially pregnant patients and should know the range and likelihood of effects that radiation might have on pregnancy. METHODS: We present guidelines for assessing such radiation risks. Knowledge of a patient's pregnancy status and an estimate of radiation dose to the conceptus (low, < 10 mGy [milligray]; intermediate, 10-250 mGy; high, > 250 mGy) allow provisional assessment of radiation-induced risks. RESULTS: Dose estimates may be estimated at 2 mGy per exposure (radiographs), 5 mGy per slice (computed tomography), and 10 mGy per minute of fluoroscopy, when the conceptus is within the x-ray field. A formal radiation exposure assessment is appropriate when provisional estimates exceed 10 mGy. CONCLUSION: A simple qualitative dose assessment can inform clinical decisions and guide appropriate triage to more formal quantitative assessment.

Cervical spine tomography with an angiographic C-arm.

Phantom studies were performed to develop a technique for linear tomography of the craniocervical junction with a digital fluoroscopic angiographic C-arm unit. Section thicknesses were similar to those used at conventional tomography, and the radiation dose was lower. C-arm tomography was possible with a 6-second exposure and a 40 degrees arc. C-arm tomography is a practical method for decreasing patient turnaround time.

Radiation shielding implications of computed tomography scatter exposure to the floor.

This article investigates whether CT scanners used primarily for either head or body work have scattered radiation levels that may impact shielding considerations. Several 35 x 43 cm film cassettes were used to cover the floor area around a General Electric Hi-Speed Advantage CT unit. Regions of maximum exposure were identified (visually with sensitometric confirmation). An ion chamber was used to obtain radiation exposure levels for typical head and body protocols. Greater scattered radiation was found behind the gantry for head scanning (20 vs. 16 microGy for a body scan) while greater scatter was observed in front of the scanner for body scans (46 vs. 31 microGy for the head scan). The resulting annual doses to the floor from typical workloads (assuming 32 cases a day) can be as great as 0.63 Gy for head protocols or 1 Gy from body work. Care may be needed to assure adequate shielding for floor areas near CT scanners specializing in either head or body work. Also, while increased throughput with helical scanners is currently x-ray tube heat limited, shielding plans should allow for enhanced heat capacity (and greater throughput) in future generation scanners.

Integration of DICOM images into an electronic medical record using thin viewing clients.

Over the past five years the University of Washington has created a clinical data repository. This repository combines in a distributed relational database information from multiple departmental databases (MIND). MINDscape provides a platform independent, web browser view of the MIND dataset that can easily be linked to other information resources on the network.