A Review of the Clinical Applications of Artificial Intelligence in Abdominal Imaging.

Artificial intelligence (AI) has been a topic of substantial interest for radiologists in recent years. Although many of the first clinical applications were in the neuro, cardiothoracic, and breast imaging subspecialties, the number of investigated and real-world applications of body imaging has been increasing, with more than 30 FDA-approved algorithms now available for applications in the abdomen and pelvis. In this manuscript, we explore some of the fundamentals of artificial intelligence and machine learning, review major functions that AI algorithms may perform, introduce current and potential future applications of AI in abdominal imaging, provide a basic understanding of the pathways by which AI algorithms can receive FDA approval, and explore some of the challenges with the implementation of AI in clinical practice.

Pancreatic Transplantation: Surgical Anatomy and Complications.

Pancreatic transplantation is a complex surgical procedure performed for patients with chronic severe diabetes, often performed in combination with renal transplantation. Vascular and exocrine drainage anatomy varies depending on the surgical technique. Radiology plays a critical role in the diagnosis of postoperative complications, requiring an understanding of grayscale/Doppler ultrasound as well as computed tomography and MR imaging. In this review, we detail usual surgical methods and normal postoperative imaging appearances. We then review the most common complications following pancreatic transplants, emphasizing diagnostic features of vascular (arterial/venous), surgical, and diffuse parenchymal pathologic conditions on multiple imaging modalities.

Radiology Dictation Errors with COVID-19 Protective Equipment: Does Wearing a Surgical Mask Increase the Dictation Error Rate?

Our aim was to determine the effect of wearing a surgical mask on the number and type of dictation errors in unedited radiology reports. IRB review was waived for this prospective matched-pairs study in which no patient data was used. Model radiology reports (n = 40) simulated those typical for an academic medical center. Six randomized radiologists dictated using speech-recognition software with and without a surgical mask. Dictations were compared to model reports and errors were classified according to type and severity. A statistical model was used to demonstrate that error rates for all types of errors were greater when masks are worn compared to when they are not (unmasked: 21.7 ± 4.9 errors per 1000 words, masked: 27.1 ± 2.2 errors per 1000 words; adjusted p < 0.0001). A sensitivity analysis was performed, excluding a reader with a large number of errors. The sensitivity analysis found a similar difference in error rates for all types of errors, although significance was attenuated (unmasked: 16.9 ± 1.9 errors per 1000 words, masked: 20.1 ± 2.2 errors per 1000 words; adjusted p = 0.054). We conclude that wearing a mask results in a near-significant increase in the rate of dictation errors in unedited radiology reports created with speech-recognition, although this difference may be accentuated in some groups of radiologists. Additionally, we find that most errors are minor single incorrect words and are unlikely to result in a medically relevant misunderstanding.

Metastatic basal cell carcinoma with atypical pattern of spread.

Basal cell carcinoma is the most common malignancy in the United States. However, metastasis of basal cell carcinoma is exceedingly rare, with incidence estimates of 0.0028-0.055%. When it does metastasize, basal cell carcinoma most commonly spreads to regional lymph nodes and lungs, although other sites of disease can occur. This case report presents multi-modality imaging of a 54-year-old male who developed multifocal metastatic basal cell carcinoma approximately three years after initial presentation with an ulcerated groin lesion. Ultimately, metastases included many common and uncommon sites, including lymph nodes, lung, duodenum, spleen, and adrenal glands. This case provides an interesting example of an uncommon pattern of spread and associated symptoms of treatment-resistant metastatic basal cell carcinoma.

MRI of acute appendicitis.

Appendicitis is the most common cause of acute abdominal pain resulting in surgery. While historically ultrasound (US) and computed tomography (CT) have been used to evaluate for appendicitis and its related complications, magnetic resonance imaging (MRI) has become a highly accurate and increasingly utilized modality in the last two decades, particularly in the pediatric and pregnant patient populations in whom ionizing radiation is used reluctantly. This article discusses the advantages and disadvantages of MRI as a modality to evaluate for acute appendicitis, summarizes studies of the diagnostic performance relative to CT and US, provides a standard MR protocol, and describes MRI findings typical of acute appendicitis, common complications, and other differential diagnoses. Level of Evidence: 2 Technical Efficacy Stage: 2 J. Magn. Reson. Imaging 2019;50:1367-1376.

MRI in pregnancy: Indications and practical considerations.

There is a constantly evolving knowledgebase regarding the safety of MRI in pregnant patients, as well as the safety of gadolinium administration, given potential fetal risks. This review provides an overview of national and international recommendations for patient screening and safety by trimester, evaluates the most recent literature regarding administration of gadolinium in pregnant patients, and discusses technical requirements when imaging pregnant patients. A protocol for imaging pregnant patients is provided, and multiple common indications for MRI in pregnancy are discussed. Level of Evidence 5. Technical Efficacy Stage 5. J. Magn. Reson. Imaging 2019;49:621-631.

Breakthrough Reactions to Gadobenate Dimeglumine.

PURPOSE: The aim of this study was to determine the severity of breakthrough reactions to gadobenate dimeglumine in patients premedicated with a 13-hour premedication regimen. METHODS: Institutional review board approval was obtained and informed consent waived for this Health Insurance Portability and Accountability Act-compliant retrospective cohort study. All acute allergic-like reactions to gadobenate dimeglumine from 11/1/2008 to 1/31/2016 were identified. Of these, 19 allergic-like reactions followed 13-hour premedication: 150 mg prednisone and 50 mg diphenhydramine (ie, "breakthrough reactions"). Reasons for premedication, risk factors, index reaction characteristics, and breakthrough reaction characteristics were catalogued. Reaction severities were assigned using American College of Radiology guidelines. Severities of breakthrough (n = 19) and nonbreakthrough reactions (n = 97) were compared with the Cochran-Armitage test for trend. RESULTS: Premedication was most commonly given (63% [12/19]) for a previous allergic-like reaction to gadolinium-based contrast material (GBCM); in 37% (7/19), it was given for a different risk factor. In those premedicated for a previous allergic-like reaction to GBCM of known severity (n = 9), the breakthrough reaction severity was the same as index reaction severity in 56% (5/9), less severe in 11% (1/9), and of greater severity in 33% (3/9). Two severe breakthrough reactions occurred; both were in subjects premedicated for risk factors other than a previous GBCM reaction. No subjects died. Five subjects were reexposed to GBCM a total of 9 times; no repeat breakthrough reactions occurred. Breakthrough reactions were more severe than nonbreakthrough reactions (P = 0.046), but the level of significance was borderline. CONCLUSION: Premedication does not eliminate severe reactions to gadobenate dimeglumine. Breakthrough reactions to gadobenate dimeglumine can be of greater severity than index reactions.

Intravenous Corticosteroid Premedication Administered 5 Hours before CT Compared with a Traditional 13-Hour Oral Regimen.

Purpose To determine if the allergic-like breakthrough reaction rate of intravenous corticosteroid prophylaxis administered 5 hours before contrast material-enhanced computed tomography (CT) is noninferior to that of a traditional 13-hour oral regimen. Materials and Methods Institutional review board approval was obtained and informed consent waived for this retrospective noninferiority cohort study. Subjects (n = 202) who completed an accelerated 5-hour intravenous corticosteroid premedication regimen before low-osmolality contrast-enhanced CT for a prior allergic-like or unknown-type reaction to iodine-based contrast material from June 1, 2008, to June 30, 2016, were identified. The breakthrough reaction rate was compared by using the Farrington and Manning noninferiority likelihood score to test subjects premedicated with a traditional 13-hour oral regimen (2.1% [13 of 626]). All subjects were premedicated for a prior allergic-like or unknown-type reaction to iodine-based contrast material. A noninferiority margin of 4.0% was selected to allow for no more than a clinically negligible 6.0% breakthrough reaction rate in the cohort that received 5-hour intravenous corticosteroid prophylaxis. Results The breakthrough reaction rate for 5-hour intravenous prophylaxis was 2.5% (five of 202 patients; 95% confidence interval: 0.8%, 5.7%), which was noninferior to the 2.1% (13 of 626 patients; 95% confidence interval: 1.1%, 3.5%) rate for the 13-hour regimen (P = .0181). The upper limits of the confidence interval for the difference between the two rates was 3.7% (0.4%; 95% confidence interval: -1.6%, 3.7%), which was within the 4.0% noninferiority margin. All breakthrough reactions were of equal or lesser severity to those of the index reactions (two severe, one moderate, and one mild reaction). Conclusion Accelerated intravenous premedication with corticosteroids beginning 5 hours before contrast-enhanced CT has a breakthrough reaction rate noninferior to that of a 13-hour oral premedication regimen. © RSNA, 2017.

What the Patient Wants: An Analysis of Radiology-Related Inquiries From a Web-Based Patient Portal.

PURPOSE: With the development of patient portals, the opportunity exists to identify gaps in practice by analyzing priorities patients place on the receipt and comprehension of radiology reports. Our purpose was to describe the nature of radiology-specific patient information requests by analysis of patient-initiated messages submitted through a web-based electronic patient portal. METHODS: Institutional review board approval was obtained and informed consent waived for this HIPAA-compliant retrospective cross-sectional study. All patient-initiated messages submitted to the web-based patient portal at a large academic medical center between October 1, 2014 and December 11, 2014 were analyzed. Messages containing radiology-specific key terms including "x-ray," "xray," "xr," "ct," "cat," "mri," "scan," "ultrasound," "image," and "radiology" were identified and messages categorized by content. The demographics of message writers were also analyzed. Diagnostic imaging studies performed during this period were tabulated by modality. Proportions were compared with χ2 tests. RESULTS: During the time period studied, there were 1,597 messages from 1,489 patients inquiring about 1,609 examinations. Messages containing ≥1 radiology-specific keyword were significantly more likely to originate from women than from men (64% [946/1,489] versus 36% [543/1,489], P < .0001), with 53% of studies (52,322/98,897) performed on female patients and 47% (46,575/98,897) on male patients. The relative percentages of modality-specific patient inquiries were significantly discrepant (P < .001) from actual scan volume for some modalities (MRI: 38% [607/1,609] versus 11% [11,152/98,897], CT: 25% [400/1,609] versus 19% [19,032/98,897], plain radiography: 23% [368/1,609] versus 55% [54,497/98,897]). The most common inquiry was for imaging results (33% [521/1,597], P < .001); these were submitted a median of 5 days (range: 0-368 days) after imaging. The radiology turnaround time (between exam completion in the Radiology Information System and signoff on report) was 5 hours, versus 70 hours for referring provider review. Inquiries about radiation dose or radiation risk represented 0.1% (2/1,597) of all inquiries. CONCLUSION: Patients submitting radiology-specific messages through an electronic patient portal are most concerned about imaging results, particularly those pertaining to advanced (CT and MRI) imaging studies.

Diagnostic Accuracy of Ultrasound, Contrast-enhanced CT, and Conventional MRI for Differentiating Leiomyoma From Leiomyosarcoma.

RATIONALE AND OBJECTIVES: This study aimed to determine whether uterine leiomyoma can be distinguished from uterine leiomyosarcoma on ultrasound (US), computed tomography (CT), and/or magnetic resonance imaging (MRI) without diffusion-weighted imaging. MATERIALS AND METHODS: Institutional review board approval was obtained and informed consent was waived for this Health Insurance Portability and Accountability Act-compliant retrospective case-control diagnostic accuracy study. All subjects with resected uterine leiomyosarcoma diagnosed over a 17-year period (1998-2014) at a single institution for whom pre-resection US (n = 10), CT (n = 11), or MRI (n = 7) was available were matched by tumor size and imaging modality with 28 subjects with resected uterine leiomyoma. Six blinded radiologists (three attendings, three residents) assigned 5-point Likert scores for the following features: (1) margins, (2) necrosis, (3) hemorrhage, (4) vascularity, (5) calcifications, (6) heterogeneity, and (7) likelihood of malignancy (primary end point). Mean suspicion scores were calculated and receiver operating characteristic curves were generated. The ability of individual morphologic features to predict malignancy was assessed with logistic regression. RESULTS: Mean suspicion scores were 2.5 ± 1.2 (attendings) and 2.4 ± 1.3 (residents) for leiomyoma, and 2.7 ± 1.3 (attendings) and 2.7 ± 1.4 (residents) for leiomyosarcoma. The areas under the receiver operating characteristic curves (range: 0.330-0.685) were not significantly different from chance, either overall (P = .36-.88) or by any modality (P = .28-.96), for any reader. Reader experience had no effect on diagnostic accuracy. No morphologic parameter was significantly predictive of malignancy (P = .10-.97). CONCLUSIONS: Uterine leiomyoma cannot be differentiated accurately from leiomyosarcoma on US, CT, or MRI without diffusion-weighted imaging.

Indirect Cost and Harm Attributable to Oral 13-Hour Inpatient Corticosteroid Prophylaxis before Contrast-enhanced CT.

PURPOSE: To estimate the effect of an oral 13-hour inpatient corticosteroid premedication regimen on length of stay, hospital cost, and hospital-acquired infections (HAIs) by using a combination of real and hypothetical study populations. MATERIALS AND METHODS: Institutional review board approval was obtained and informed consent waived for this HIPAA-compliant retrospective study. Inpatients who received an oral 13-hour corticosteroid premedication regimen before contrast material-enhanced CT (n = 1424) from 2008 to 2013 were matched by age, sex, and year when CT was performed to a control cohort (n = 1425) of patients who underwent contrast-enhanced CT without premedication and who had similar rates of 13 comorbid diseases. Length of stay in the hospital and time from admission to CT were compared by using the Mann-Whitney U test. Rates of prospectively reported HAIs were compared by using χ(2) tests. The indirect cost and risk of HAI with premedication were estimated by using published data. RESULTS: Premedicated inpatients had a significantly longer median length of stay (+25 hours; 158 vs 133 hours, P < .001), a significantly longer median time to CT (+25 hours, 42 vs 17 hours, respectively; P < .001), and a significantly greater risk of HAI (5.1% [72 of 1424] vs 3.1% [44 of 1424], respectively; P = .008) compared with nonpremedicated control subjects. On the basis of these data and existing references, the prolonged length of stay was estimated to result in 0.04 HAI-related deaths and a cost of $159 131 (in U.S. dollars) for each prevented reaction of any severity and 32 HAI-related deaths and a cost of $131 211 400 for each prevented reaction-related death. CONCLUSION: Oral 13-hour inpatient corticosteroid prophylaxis is associated with substantial cost relative to its modest benefit, and may cause more indirect harm than the direct harm that it prevents.

Rates of Breakthrough Reactions in Inpatients at High Risk Receiving Premedication Before Contrast-Enhanced CT.

OBJECTIVE: The purpose of this study was to determine the rate of allergiclike breakthrough reactions among inpatients at high risk receiving premedication before undergoing CT with IV iodinated low-osmolality contrast material (LOCM). MATERIALS AND METHODS: Inpatients (n = 1051) completing a 13-hour corticosteroid and diphenhydramine premedication regimen before LOCM-enhanced CT from January 1, 2010, through December 31, 2013, were included in the study. Breakthrough reaction rates were compared with the ordinary allergiclike reaction rate in the general population (0.6% [545/84,928]) by use of chi-square tests. Multivariate logistic regression was performed. Number needed to treat (NNT) was calculated for patients premedicated for a previous contrast reaction. RESULTS: Sixty percent (626/1051) of premedicated patients had had a previous reaction to iodinated contrast material, and 40% (425/1051) were premedicated for other reasons. The overall breakthrough reaction rates were 1.2% (13/1051) (p < 0.0001 vs the general population), 2.1% (13/626) for those with a previous iodinated contrast reaction (p < 0.0001), and 0% (0/425) for those premedicated for other reasons (p = 0.18). There were no severe breakthrough reactions. Younger age (p = 0.046; odds ratio, 1.03 per year; 95% CI, 1.001-1.07) and multiple indications for premedication (p < 0.0001; odds ratio, 2.7 per indication; 95% CI, 1.5-4.8) significantly increased the likelihood of a breakthrough reaction. The estimated NNTs were 69 (95% CI, 39-304) to prevent a reaction of any severity and 569 (95% CI, 389-1083) to prevent a severe reaction. CONCLUSION: Patients premedicated for a previous reaction to iodinated contrast material have a breakthrough reaction rate 3-4 times the ordinary reaction rate in the general population. Patients receiving premedication for other reasons have a breakthrough reaction rate near 0%. Many patients must receive premedication to prevent one reaction.

Paralabral cysts of the hip: sonographic evaluation with magnetic resonance arthrographic correlation.

This retrospective study demonstrates the sonographic appearance paralabral cysts of the hip with magnetic resonance (MR) arthrography as the reference standard. Consensus review by 2 musculoskeletal radiologists was used to assess the paralabral cysts and determine their characteristics. The 3 paralabral cysts seen on sonography and confirmed with MR arthrography in this study were anechoic or hypoechoic, lobulated, and filled with contrast on MR imaging. A labral tear was also noted in all cases on both sonography and MR arthrography. Sonographic assessment for an anterior hip paralabral cyst and labral tear may be of diagnostic benefit.

A low mortality model of chronic pulmonary hypertension in sheep.

BACKGROUND: Pulmonary hypertension and right ventricular failure are major contributors to morbidity and mortality in chronic lung disease. Therefore, large animal models of pulmonary hypertension and right ventricular hypertrophy are needed to study underlying disease mechanisms and test new treatment modalities. The objective of this study was to create a low-mortality model of chronic pulmonary hypertension and right ventricular hypertrophy in sheep. METHODS: The vena cavae of nine sheep weighing 62 ± 2 (SEM) kg were injected with 0.375 g of dextran beads (sephadex) every day for 60 d. Pulmonary hemodynamics were assessed via pulmonary artery catheterization prior to the first injection and again on d 14, 28, 35, 42, 49, and 56. At the end of the experiment, the heart was removed, dissected, and weighed to determine the ratio of right ventricular mass to left ventricle plus septal mass (RV:LV+S). RESULTS: All sheep survived to 60 d. The average pulmonary artery pressure rose from 17 ± 1 mmHg at baseline to 35 ± 3 mmHg on d 56 with no significant change in cardiac output (8.7 ± 0.7 to 9.8 ± 0.7 L/min, P = 0.89). The RV:LV+S was significantly higher (0.42 ± 0.01, P < 0.001) than a historic group of untreated normal animals (0.35 ± 0.01, n = 13). CONCLUSION: This study provides a low-mortality large animal model of moderate chronic pulmonary hypertension and right ventricular hypertrophy.