Artificial Intelligence (AI) carries the potential for unprecedented disruption in radiology, with possible positive and negative consequences. The integration of AI in radiology holds the potential ...to revolutionize healthcare practices by advancing diagnosis, quantification, and management of multiple medical conditions. Nevertheless, the ever-growing availability of AI tools in radiology highlights an increasing need to critically evaluate claims for its utility and to differentiate safe product offerings from potentially harmful, or fundamentally unhelpful ones.
This multi-society paper, presenting the views of Radiology Societies in the USA, Canada, Europe, Australia, and New Zealand, defines the potential practical problems and ethical issues surrounding the incorporation of AI into radiological practice. In addition to delineating the main points of concern that developers, regulators, and purchasers of AI tools should consider prior to their introduction into clinical practice, this statement also suggests methods to monitor their stability and safety in clinical use, and their suitability for possible autonomous function. This statement is intended to serve as a useful summary of the practical issues which should be considered by all parties involved in the development of radiology AI resources, and their implementation as clinical tools.
Key points
• The incorporation of artificial intelligence (AI) in radiological practice demands increased monitoring of its utility and safety.
• Cooperation between developers, clinicians, and regulators will allow all involved to address ethical issues and monitor AI performance.
• AI can fulfil its promise to advance patient well-being if all steps from development to integration in healthcare are rigorously evaluated.
To compare the diagnostic performance of dual-energy (DE) computed tomography (CT) with two simultaneously administered contrast agents (hereafter, dual contrast) with that of conventional CT in the ...evaluation of the presence and source of extravasation in penetrating abdominopelvic trauma.
Institutional animal care and use committee approval was obtained, and the study was performed in accordance with National Institutes of Health guidelines for the care and use of laboratory animals. Five rabbits with bowel trauma, vascular penetrating trauma, or both were imaged with simultaneous iodinated intravenous and bismuth subsalicylate enteric contrast material at DE CT. Four attending radiologists and six radiology residents without prior DE CT experience each evaluated 10 extraluminal collections to identify the vascular and/or enteric origin of extravasation and assess their level of diagnostic confidence, first with virtual monochromatic images simulating conventional CT and then with DE CT material decomposition attenuation maps.
Overall accuracy of identification of source of extravasation increased from 78% with conventional CT to 92% with DE CT (157 of 200 diagnoses vs 184 of 200 diagnoses, respectively; P < .001). Nine radiologists were more accurate with DE CT; one had no change. Mean confidence increased from 67% to 81% with DE CT (P < .001).
In a rabbit abdominopelvic trauma model, dual-contrast DE CT significantly increased accuracy and confidence in the diagnosis of vascular versus enteric extravasated contrast material.
The purpose of this article is to describe the imaging features of diseases that may closely simulate pancreatic adenocarcinoma, either radiologically or pathologically.
Neoplastic and inflammatory ...diseases that can closely simulate pancreatic adenocarcinoma include neuroendocrine tumor, metastasis to the pancreas, lymphoma, groove pancreatitis, autoimmune pancreatitis, and focal chronic pancreatitis. Atypical imaging findings that should suggest diagnoses other than adenocarcinoma include the absence of significant duct dilatation, incidental detection, hypervascularity, large size (> 5 cm), IV tumor thrombus, and intralesional ducts or cysts.
Workflow interruptions in the health care delivery environment are a major contributor to medical errors and have been extensively studied within numerous hospital settings, including the nursing ...environment and the operating room, along with their effects on physician workflow. Less understood, though, is the role of interruptions in other highly specialized clinical domains and subspecialty services, such as diagnostic radiology. The workflow of the on-call radiologist, in particular, is especially susceptible to disruption by telephone calls and other modes of physician-to-physician communication. Herein, the authors describe their initial efforts to quantify the degree of interruption experienced by on-call radiologists and examine its potential implications in patient safety and overall clinical care.
Generalized Born (GB) models provide a computationally efficient means of representing the electrostatic effects of solvent and are widely used, especially in molecular dynamics (MD). Accurate and ...facile computation of the effective Born radii is a key for the performance of GB models. Here, we examine a simple integral prescription, R6, based on the exact solution of the Poisson-Boltzmann (PB) equation for a perfect sphere. Numerical tests on 22 molecules representing a variety of structural classes show that R6 may be more accurate than the more complex integral-based approaches such as GBMV2. At the same time, R6 is computationally less demanding. Fundamental limitations of current integration-based methods for calculating effective radii, including R6, are explored and the deviations from the numerical PB results are correlated with specific topological and geometrical features of the molecular surface. A small systematic bias observed in the R6-based radii can be removed with a single, transferable constant offset; when the resulting effective radii are used in the "classical" (Still et al.'s) GB formula to compute the electrostatic solvation free energy, the average deviation from the PB reference is no greater than when the "perfect" (PB-based) effective radii are used. This deviation is also appreciably smaller than the uncertainty of the PB reference itself, as estimated by comparison to explicit solvent.
Abstract Background Brain metastases from colorectal cancer (CRC) are uncommon. There has been relatively little published on the host and tumor factors that might lead to this clinical scenario. We ...reviewed all cases of brain metastases from CRC at Dartmouth-Hitchcock Medical Center over a more than 20-year period to establish incidence and to identify patient and cancer characteristics which were associated with their development. Patients and Methods We present a retrospective review of 39 confirmed cases of brain metastases from CRC diagnosed between 1984 and 2006. Immunohistochemical staining for CXCR4 was performed on all available brain metastasis biopsy specimens. Results The incidence of brain metastases from CRC was 2.3%. Left-sided primary colon tumors predominated. The majority of patients had pulmonary metastases at the time brain metastases were identified, and those with preexisting pulmonary metastases had progression of that disease. All patients were symptomatic from brain metastases, and the cerebellum was the most common area of brain involvement. Immunohistochemical analysis confirmed strong expression of CXCR4 in all brain metastases sampled. Conclusion The incidence of brain metastases from CRC is low. Primary tumor in the left colon, long-standing pulmonary metastases, especially those with recent progression, and CXCR4 expression by tumor cells are all associated with increased risk of brain metastases. Increased survival among patients with metastatic CRC will likely result in an increased incidence of brain metastases. Further characterization of the role of tumor and host factors might yield better insight into the development, and potentially the prevention, of this devastating situation.
Brain metastases from colorectal cancer (CRC) are uncommon. There has been relatively little published on the host and tumor factors that might lead to this clinical scenario. We reviewed all cases ...of brain metastases from CRC at Dartmouth-Hitchcock Medical Center over a more than 20-year period to establish incidence and to identify patient and cancer characteristics which were associated with their development.
We present a retrospective review of 39 confirmed cases of brain metastases from CRC diagnosed between 1984 and 2006. Immunohistochemical staining for CXCR4 was performed on all available brain metastasis biopsy specimens.
The incidence of brain metastases from CRC was 2.3%. Left-sided primary colon tumors predominated. The majority of patients had pulmonary metastases at the time brain metastases were identified, and those with preexisting pulmonary metastases had progression of that disease. All patients were symptomatic from brain metastases, and the cerebellum was the most common area of brain involvement. Immunohistochemical analysis confirmed strong expression of CXCR4 in all brain metastases sampled.
The incidence of brain metastases from CRC is low. Primary tumor in the left colon, long-standing pulmonary metastases, especially those with recent progression, and CXCR4 expression by tumor cells are all associated with increased risk of brain metastases. Increased survival among patients with metastatic CRC will likely result in an increased incidence of brain metastases. Further characterization of the role of tumor and host factors might yield better insight into the development, and potentially the prevention, of this devastating situation.
The purpose of this study was to objectively quantify the impact of implementing picture archiving and communication system-electronic medical record (PACS-EMR) integration on the time required to ...access data in the EMR and the frequency with which data are accessed by radiologists. Time to access a clinic note in the EMR was measured before and after integration with a stopwatch and compared by
t
test. An IRB-approved, HIPAA-compliant retrospective review of EMR access data from security audit logs was conducted for a 14-month period spanning the integration. Correlation of these data with report signatures identified the studies in which the radiologist accessed the EMR to obtain additional clinical data. Proportions of studies with EMR access were plotted and compared before and after integration using a chi-square test. Time to access the EMR decreased from 52 to 6 s (
p
< 0.001). Proportion of studies with EMR access increased from 36.7% (10,175/27,773) to 44.9% (10,843/24,153) after integration (
p
< 0.001). Integrating PACS and the EMR substantially decreases the time to access the EMR and is associated with a significant increase in the proportion of studies for which radiologists obtain additional clinical data.