Clinical Diagnostic Radiology Research Articles

Optimizing statistical evaluation of multiclass classification in diagnostic radiology: a study of the two-parameter multidimensional nominal response model.

This study aimed to enhance the multidimensional nominal response model (MDNRM) for multiclass classification in diagnostic radiology. This retrospective study involved the extension of the conventional nominal response model (NRM) to create the two-parameter MDNRM (2PL-MDNRM). Seven models of MDNRM, including the original MDNRM and subtypes of 2PL-MDNRM, were employed to estimate test-takers' abilities and test item complexity. These models were applied to a clinical diagnostic radiology dataset. Rhat values were calculated to evaluate model convergence. Additionally, values of the widely applicable information criterion (wAIC) and Pareto-smoothed importance sampling leave-one-out cross-validation (LOO) were calculated to evaluate the goodness of fit of the seven models. The best-performing model was selected based on the values of wAIC and LOO. Probability of direction (PD) was used to evaluate whether one estimated parameter significantly differed. All estimated parameters across the seven models demonstrated Rhat values below 1.10, indicating stable convergence. The best wAIC and LOO values (988 and 1,121, respectively) were achieved with 2PL-MDNRM r using the truncated normal distribution and 2PL-MDNRM a using the truncated normal distribution. Notably, one test-taker (radiologist) exhibited significantly superior ability compared to another based on PD results from the best models, while no significant difference was observed in nonoptimal models. 2PL-MDNRM successfully achieved parameter estimation convergence, and its superiority over the original MDNRM was demonstrated through wAIC and LOO values.

Contrast media- Guidelines for practical use

Contrast agents have become an indispensable part of everyday life in diagnostic radiology. In multidetector computed tomography (MDCT) and magnetic resonance imaging (MRI), they provide essential diagnostic information, especially for vascular, inflammatory or oncologic diseases, which otherwise could not be answered. The two most important groups are iodine- and gadolinium-containing contrast agents. Rare side effects include PC-AKI (post-contrast acute kidney injury); more common are allergic and chemotoxic reactions. Since the introduction of guidelines, nephrogenic fibrosis has not been reported anymore, whereas gadolinium deposition in the central nervous system (CNS) has become anew topic. Concerning contrast media use in patients with reduced renal function, at aeGFR threshold of <45 ml/min or <30 ml/min, hydration and areview of indication for enhanced MDCT, depending on the application, is recommended. Low kV and DE-scan protocols with MDCT can help to reduce the amount of iodinated contrast agents. In MRI examinations, only macrocyclic contrast agents should be used for enhanced MRI exams. There has to be acareful risk-benefit analysis with enhanced studies in pregnancy, during lactation and in the pediatric population. Patient information and legal aspects with nonapproved indications are indispensable parts of daily clinical routine. The continuous updating and broadening of knowledge regarding the appropriate use of the various contrast agents has to be an integral part of clinical diagnostic radiology.

ODIN: Adaptation of an HTML 5-Compatible DICOM Viewer for MIRC-TFS, Enhancing the Incorporation of Clinical Images into the Health Science Curriculum.

The use of diagnostic radiology images while teaching has been proven to enhance student learning. The incorporation of images in the Digital Imaging and Communications in Medicine (DICOM) format has been hindered by the lack of integrated, DICOM image-compatible, viewing software. A project was developed to facilitate the migration of clinical diagnostic radiology images, in their native DICOM format into the curriculum at the undergraduate level, including the Medical School, while anonymizing the images and presenting them in a way that simulates a clinical Picture Archive and Communication System (PACS). These images can be directly incorporated into Blackboard Learn or into other applications using embedded URLs. A teaching file server was deployed using the Medical Imaging Resource Community-Teaching File System (MIRC-TFS) platform. An HTML5, DICOM image viewing server was developed and deployed. Plugins were created to integrate the anonymized images in the MIRC-TFS server with the HTML5 DICOM viewer and Blackboard Learn. The solution was deployed at an undergraduate level. This project has resulted in the successful development and deployment of a variety of solutions, including an HTML 5 DICOM viewer, that has allowed for the incorporation of anonymized DICOM images from a clinical imaging repository into the undergraduate curriculum.

CT perfusion in solid-body tumours. Part I: technical issues

Functional imaging is becoming increasingly important in both research and clinical diagnostic radiology. Perfusion computed tomography (CTP) is a readily available and widely used tool that allows an objective measurement of tissue perfusion through the mathematical analysis of data obtained from repeated scans performed after administration of contrast agent. Recently, CTP has been increasingly used in the oncological field, being studied as a potential marker of neoplastic angiogenesis, which is one of the main targets of new tumour therapies. The aim of this paper was to provide the theoretical background and practical guidance for accurately performing CTP and interpreting results of examinations in solid-body tumours. CTP could be a valid tool for functional imaging of tumours if the acquisition technique is robust, if image and data analysis is accurate and if interpretation of results is adequately inserted within a clinical context.

WE-B-L100F-01: The Current State of 3D and 4D Imaging in Diagnostic Radiology

The most common 3D visualization applications in clinical diagnostic radiology now are cardiac and colon, but the technology is often used in vascular, orthopedics, craniofacial deformities, thoracic, and many other areas. In the past, dedicated workstations and highly skilled operators were required to produce CT or MR angiograms, body region or organ surface views, and analysis products. Many centers have dedicated technologists who work in a “3D lab” to extract vessel trees, disarticulate limbs, synthesize transparencies and cine sequences. These services are now reimbursed by 3rd party payors, and almost all CT & MRI scanners today have integrated post-processing tools. Since generation of advanced visualizations is not a good use of scanner console time if the process cannot be accomplished unattended in a short time, new enterprise systems has been introduced that allow thin clients located anywhere to employ software tools and generate custom views interactively. In the surgeon's office, at the oncology clinic, or even in the operating room, it is now feasible to manipulate 3D image data sets. Very specialized tools for white matter tractography, MR spectroscopic imaging, functional MRI, and myocardial perfusion are examples of new software agents tailored to subspecialty requirements. Several vendors provide post-processing services for prosthesis or implant custom sizing and design, based on 3D image data sets, where the advanced visualization images are quantified and synthesized in their facilities. 4D imaging and higher dimensions are increasingly common, since almost all cardiac CT and MRI examinations require them. Whole organ and body region perfusion, in the brain and elsewhere employ 4D methods. Multimodality and multitemporal data sets, as well as multispectral (e.g., dual energy 3D data, for example) are acquired and analyzed with software tools that focus on specific clinical issues in neuroradiology and cardiothoracic practice. So, the most important current trends involve routinely collecting 3D and 4D data sets which are intended for more than simple subjective image review, but are destined to be post-processed and analyzed with enterprise software tools on thin clients, so immediate results can be obtained that are tailored to subspecialty needs. Educational Objectives: 1. To understand the sources of 3D and 4D images in clinical diagnostic radiology, including the data acquisition systems and protocols. 2. To learn how advanced visualization and image analysis are evolving from dedicated workstations to enterprise software applications. 3. To see how subspecialized software tools for highly specific imaging applications are used in clinical radiology. 4. To explain 4D datasets are used for cardiac and neuroimaging, where the image analysis results are often quantitative rather than morphologic.

Potential usefulness of digital imaging in clinical diagnostic radiology: computer-aided diagnosis.

Our results obtained with various CAD schemes are very encouraging. It seems now feasible to begin an initial clinical evaluation of some CAD schemes, such as that for detection of clustered microcalcifications in digitized mammograms, to gain more experience in computerized analysis of a very large number of clinical cases. In fact, we are currently implementing the first intelligent mammography workstation in the mammography section of our department.

Proposal for a Shared Resource in Clinical Diagnostic Radiology Research

Proposal for a Shared Resource in Clinical Diagnostic Radiology Research

100 Pearls in Clinical Diagnostic Radiology

HomeRadiologyVol. 154, No. 2 PreviousNext Reviews100 Pearls in Clinical Diagnostic RadiologyDaniel B. WoppererDaniel B. WoppererDaniel B. WoppererPublished Online:Feb 1 1985https://doi.org/10.1148/radiology.154.2.312MoreSectionsPDF ToolsImage ViewerAdd to favoritesCiteTrack CitationsPermissionsReprints ShareShare onFacebookTwitterLinked In Article HistoryPublished in print: Feb 1985 FiguresReferencesRelatedDetailsRecommended Articles RSNA Education Exhibits RSNA Case Collection Vol. 154, No. 2 Metrics Altmetric Score PDF download

The magnetically guided bronchial catheter of modified POD design. A new approach to selective bronchography.

The ability to selectively catheterize pulmonary bronchial segments and subsegments has many applications in clinical diagnostic radiology. Most of the reported technics for selective bronchial catheterization have evolved from the methods of selective arteriography (1, 4, 12, 14, 15), which utilize some means of mechanical guidance. The catheter is sometimes preshaped, or the guide wire tip may be directed from a remote controlling device. Multiple variations and combinations of these approaches are in use. The selective approach to bronchial catheterization has increased the accuracy of bronchography (12), allowed selective cytological and bacteriological sampling (1, 4, 9), made possible the introduction of the brush biopsy technic (3, 5, 9, 12), and suggested further application of the method of selective delivery of medication for the treatment of suitable localized pulmonary lesions. This concept was first successfully employed by Ramirez (11) in the endobronchial treatment of aspergillosis with pot...

X-Ray Diagnosis for Clinical Students and Practitioners

In the development of a teaching program in diagnostic radiology, it is necessary for the medical student to have available systematic instruction in radiographic anatomy in his preclinical training, and this should extend into the training period in clinical medicine. On such a background provided by both radiologists and anatomists, clinical diagnostic radiology can be more effectively taught and its place as an aid to clinical diagnosis firmly established. The author has written a book primarily for the medical student during his period of clinical training and one that would have applicability to house officers and physicians in their early days of practice. The author does not intend to cover all aspects of radiology but he contributes toward the problem of incorporating the teaching of radiology into the general clinical curriculum. The author's experience was primarily the source for material in the volume, which deals with the indications, value, and

The normal lateral pneumotomographic appearance of the pancreas.

If we review the major unsolved problems in clinical diagnostic radiology, our attention is immediately attracted to the parenchymatous organs of the upper abdomen. The need to examine these organs radiographically is challenging, since diagnostic problems involving the epigastrium are frequent. Because of the deep location of the pancreas and its relatively small dimensions, significant changes in size and shape of the gland may not be noted by palpation or by routine radiographic methods. Indirect radiographic signs, such as displacement of adjacent organs, are present only if the pancreatic enlargement is considerable and may be absent or questionable even in massive lesions. Frequently, therefore, with disease of the pancreas or biliary tree, a conclusive evaluation is obtained only after operation. Since many of these conditions are surgical, a definitive clinical diagnosis is often not very important, and a radiological diagnosis would be desirable only if the procedure were safe, conclusive, relatively simple, and with well defined limitations. In a minority of cases, however an evaluation of the pancreas without resort to laparotomy is desirable. There are two practical methods for visualizing a parenchymatous organ possessing an excretory system. One method strives for a delineation of the gross appearance of the organ. A second method utilizes opacification of the excretory system to demonstrate the internal architecture. This latter approach has been successfully accomplished following insertion of a T-tube at the time of operation by Doubilet and his group (1). A more useful method for demonstration of the secretory system of an organ employs parenchymal saturation with a contrast substance, as in nephrography or nephrotomography. As yet, no such substance which fulfills the pharmacological requirements has been found for the pancreas. There is little doubt that this would be the ideal method! Because of the peculiar vascular distribution and the small circulation of the pancreas, angiographic methods do not seem to be promising. There remains the introduction of gas as a contrast agent. The simplest procedure of this type is evaluation of the width of the retrogastric space by inflation of the stomach according to Engel and Lysholm (2). Kuhlmann (3), Lüdin (4), Bonomini (5), Macarini and Oliva (6, 9), Sansone et at. (7), and Lura (8), have used retroperitoneal insufflation combined with gastric insufflation or pneumoperitoneum, with tomography in different planes. Technic Our technic is based on the anatomical knowledge that the pancreas is bounded posteriorly by the retroperitoneal fat tissues, anteriorly by the omental bursa, and below by the inframesocolic peritoneal cavity and its intestinal contents (Fig. lA). Inflation of the retroperitoneal space as well as the peritoneal cavity, especially the omental bursa, will provide the contrast necessary to delineate the anterior and posterior walls of the pancreas.

Sixty years of radiology; radiology.

An account of the first sixty years of radiology would require no less than twenty or thirty thousand words, and then the account would be a brief one. To accomplish this in a few hundred words is extremely difficult. At this last moment I feel tempted to say that I am not equal to the task. A mere summing up of the highlights of radiology in 60 years would tell nothing new. One has only to refer to the 60 volumes of the Archives of Skiagraphy, later the Archives of the Röntgen Ray, and finally The British Journal of Radiology, to learn the full history of radiology. It is true that the early volumes are not generally available, and so for guidance it is perhaps worthwhile to summarise briefly the evolution of modern radiology. In the beginning it was the physicist who really began the application of the X-rays to medicine, but very shortly the physicians took over, as seemed fit in that early phase of development of radiology, but within a few years the work of the radiation physicist again assumed an indispensable place in radiologic practice. In fact, the field of radiology, even as restricted to medicine and biologic research, has extended so greatly that one has to choose whether he will devote his attention to clinical diagnostic radiology, to clinical therapeutic radiology, to research radiology, or to pure radiologic and biologic investigations; and yet in each of these fields at least some knowledge of the other phases of radiology is necessary.