Artificial Intelligence/Machine Learning Education in Radiology: Multi-institutional Survey of Radiology Residents in the United States.

RATIONALE AND OBJECTIVES: To evaluate radiology residents' perspectives regarding inclusion of artificial intelligence/ machine learning (AI/ML) education in the residency curriculum. MATERIALS AND METHODS: An online anonymous survey was sent to 759 residents at 21 US radiology residency programs. Resident demographics, sub-specialty interests, educational background and research experiences, as well as the awareness, availability, and usefulness of various resources for AI/ML education were collected. RESULTS: The survey response rate was 27% (209/759). A total of 74% of respondents were male, 80% were training at large university programs, and only a minority (<20) had formal education or research experience in AI/ML. All four years of training were represented (range: 20%-38%). The majority of the residents agreed or strongly agreed (83%) that AI/ML education should be a part of the radiology residency curriculum and that such education should equip them with the knowledge to troubleshoot an AI tool in practice / determine whether a tool is working as intended (82%). Among the residency programs that offer AI/ML education, the most common resources were lecture series (43%), national informatics courses (28%), and in-house/institutional courses (26%). About 24% of the residents reported no AI/ML educational offerings in their residency curriculum. Hands on AI/ML laboratory (67%) and lecture series (61%) were reported as the most beneficial or effective. The majority of the residents preferred AI/ML education offered as a continuous course spanning the radiology residency (R1 to R4) (76%), followed by mini fellowship during R4 (32%) and as a course during PGY1 (21%). CONCLUSION: Residents largely favor the inclusion of formal AI/ML education in the radiology residency curriculum, prefer hands-on learning and lectures as learning tools, and prefer a continuous AI/ML course spanning R1-R4.

Imaging of Soft Tissue Infections.

Although superficial infections can often be diagnosed and managed clinically, physical examination may lack sensitivity and specificity, and imaging is often required to evaluate the depth of involvement and identify complications. Depending on the area of involvement, radiography, ultrasound, CT, MR imaging, or a combination of imaging modalities may be required. Soft tissue infections can be nonnecrotizing or necrotizing, with the later having a morbid and rapid course. Infectious tenosynovitis most commonly affects the flexor tendon sheaths of the hand, characterized by thickened and enhancing synovium with fluid-filled tendon sheaths.

Utilization of Chest and Abdominopelvic CT for Traumatic Injury From 2011 to 2018: Evaluation Using a National Commercial Database.

BACKGROUND. Increases in the use of CT to evaluate patients presenting with trauma have raised concern about inappropriate imaging. The evolving utilization of CT for trauma evaluation may be impacted by injury severity. OBJECTIVE. The purpose of this study was to explore patterns in utilization of chest and abdominopelvic CT among trauma-related emergency department (ED) visits across the United States. METHODS. This retrospective study was conducted with national commercial claims information extracted from the MarketScan Commercial Database. Trauma-related ED encounters were identified from the 2011-2018 MarketScan database files and classified by injury severity score (minor, intermediate, and major injuries) on the basis of International Classification of Diseases codes. ED encounters were also assessed for chest CT, abdominopelvic CT, and single-encounter chest and abdominopelvic CT examinations. Utilization per 1000 trauma-related ED encounters was determined. Multivariable Poisson regression models were used to determine incidence rate ratios (IRRs) as a measure of temporal changes in utilization. RESULTS. From 2011 to 2018, 8,369,092 trauma-related ED encounters were identified (5,685,295 for minor, 2,624,944 for intermediate, and 58,853 for major injuries). Utilization of chest CT per 1000 trauma-related ED encounters increased from 4.9 to 13.5 examinations (adjusted IRR, 1.15 per year; minor injuries, from 2.2 to 7.7 [adjusted IRR, 1.17]; intermediate injuries, from 8.5 to 21.5 [adjusted IRR, 1.16]; major injuries, from 117.8 to 200.1 [adjusted IRR, 1.08]). Utilization of abdominopelvic CT per 1000 trauma-related ED encounters increased from 7.5 to 16.4 (adjusted IRR, 1.12; minor injuries, 4.8 to 12.2 [adjusted IRR, 1.13]; intermediate injuries, 10.6 to 21.7 [adjusted IRR, 1.13]; major injuries, 134.8 to 192.6 [adjusted IRR, 1.07]). Utilization of single-encounter chest and abdominopelvic CT per 1000 trauma-related ED encounters increased from 3.4 to 8.9 [adjusted IRR, 1.16; minor injuries, 1.1 to 4.6 [adjusted IRR, 1.18]; intermediate injuries, 6.4 to 16.4 [adjusted IRR, 1.16]; major injuries, 99.6 to 179.9 [adjusted IRR, 1.08]). CONCLUSION. National utilization of chest and abdominopelvic CT for trauma-related ED encounters increased among commercially insured patients from 2011 to 2018, particularly for single-encounter chest and abdominopelvic CT examinations and for minor injuries. CLINICAL IMPACT. Given concerns about increased cost and detection of incidental findings, further investigation is warranted to explore the potential benefit of single-encounter chest and abdominopelvic CT examinations of patients with minor injuries and to develop strategies for optimizing appropriateness of imaging orders.

Preprocedural Hypertension Is Not a Risk Factor for Postoperative Bleeding following Image-Guided Core Needle Breast Biopsy.

OBJECTIVE: To evaluate the association between preprocedural hypertension and the risk of prolonged bleeding following image-guided core needle breast biopsy in nonpregnant/nonlactating women. METHODS: A single institution-based, retrospective cohort study of 400 women who underwent image-guided core needle breast biopsy was conducted. Males and pregnant and lactating women were excluded. Preprocedural systolic or diastolic blood pressure greater than 140 or 90 mm of Hg, respectively, was defined as hypertension. Prolonged bleeding was defined >15 minutes of local, manual pressure required to achieve hemostasis following the biopsy. Severe bleeding complications defined as clinical significant hematoma formation, prolonged bleeding requiring an ER visit, hospitalization, or surgical intervention were also recorded. RESULTS: The difference in the mean time for which manual pressure was held after biopsy for patients with and without preprocedural hypertension was not statistically significant (13 ± 7 vs. 13 ± 8 minutes, respectively, P = 0.856). There was no difference in the number of patients requiring manual postoperative pressure >15 minutes between those with preprocedural hypertension and the normotensive patients (13% vs. 12%, respectively, P = 0.765). Bivariate analysis demonstrated statistically significant association between prolonged bleeding and current antithrombotic or antiplatelet medication use (P = 0.010), the use of stereotactic guidance (P = 0.019), and a tethered vacuum-assisted device (P = 0.045). The use of a tethered vacuum-assisted biopsy device was the only variable associated with prolonged bleeding in the multivariate model (P = 0.044). CONCLUSION: Preprocedural hypertension is not a risk factor for prolonged bleeding following image-guided core needle breast biopsies in nonpregnant/nonlactating women.

Factors Influential in the Selection of Radiology Residents in the Post-Step 1 World: A Discrete Choice Experiment.

OBJECTIVES: Reporting of United States Medical Licensing Examination Step 1 results will transition from a numerical score to a pass or fail result. We sought an objective analysis to determine changes in the relative importance of resident application attributes when numerical Step 1 results are replaced. METHODS: A discrete choice experiment was designed to model radiology resident selection and determine the relative weights of various application factors when paired with a numerical or pass or fail Step 1 result. Faculty involved in resident selection at 14 US radiology programs chose between hypothetical pairs of applicant profiles between August and November 2020. A conditional logistic regression model assessed the relative weights of the attributes, and odds ratios (ORs) were calculated. RESULTS: There were 212 participants. When a numerical Step 1 score was provided, the most influential attributes were medical school (OR: 2.35, 95% confidence interval [CI]: 2.07-2.67), Black or Hispanic race or ethnicity (OR: 2.04, 95% CI: 1.79-2.38), and Step 1 score (OR: 1.8, 95% CI: 1.69-1.95). When Step 1 was reported as pass, the applicant's medical school grew in influence (OR: 2.78, 95% CI: 2.42-3.18), and there was a significant increase in influence of Step 2 scores (OR: 1.31, 95% CI: 1.23-1.40 versus OR 1.57, 95% CI: 1.46-1.69). There was little change in the relative influence of race or ethnicity, gender, class rank, or clerkship honors. DISCUSSION: When Step 1 reporting transitions to pass or fail, medical school prestige gains outsized influence and Step 2 scores partly fill the gap left by Step 1 examination as a single metric of decisive importance in application decisions.

Herlyn-Werner-Wunderlich syndrome: A rare cause of abdominal pain and dyspareunia.

Herlyn-Werner-Wunderlich (HWW) syndrome is a rare congenital disease characterized by Mullerian duct anomalies. The characteristic triad of this syndrome includes didelphys uterus, obstructed hemivagina, and ipsilateral renal agenesis (hence, also known as Obstructed Hemivagina and Ipsilateral Renal Agenesis or OHVIRA syndrome). The clinical presentation of HWW is nonspecific and patients may present with abdominal/pelvic pain, dyspareunia, or dysmenorrhea. However, the imaging findings of HWW are characteristic and discussed in this case report of a 16-year-old female who presented with abdominal pain and dyspareunia.

Salsalate improves glycaemia in overweight persons with diabetes risk factors of stable statin-treated cardiovascular disease: A 30-month randomized placebo-controlled trial.

OBJECTIVE: To assess long-term efficacy and safety of salsalate to improve glycemia in persons with diabetes risk, who are overweight with statin-treated, stable coronary heart disease. METHODS: Glycemic status was assessed in 192 persons without diabetes at baseline in a pre-specified secondary analysis from Targeting INflammation Using SALsalate in CardioVascular Disease (TINSAL-CVD), a multi-center, double-masked, randomized (1:1), placebo-controlled, parallel clinical trial. RESULTS: Participants were mostly Caucasian males, age 60±7 years, BMI 31.4±3.0 kg/m2 , fasting glucose 92.8±11.0 mg/dL, and HbA1c 5.8±0.3%. Reductions in mean fasting glucose -5.70 mg/dL (95%CI: -7.44 to -3.97 mg/dL, P<0.001), HbA1c -0.11% (95%CI: -0.210 to -0.002%, P=0.046) and glycated serum protein -81.8 µg/mL (95%CI: -93.7 to -69.9 µg/mL, P<0.001) were demonstrated in salsalate compared to placebo-assigned groups over 30 months. Reductions in fasting glucose and glycated serum protein were greater with salsalate compared to placebo in participants with prediabetes compared to a normoglycemic sub-group (Pinteraction =0.018). Salsalate lowered total white blood cell counts (mean difference -0.7x103 /µL, 95%CI: -1.0 to -0.4 x103 /µL, P<0.001) and increased adiponectin (mean difference 1.8 µg/mL, 95%CI: 0.9 to 2.6 µg/mL, P<0.001) and albuminurea (16.7 µg/mg, 95%CI: 6.4 to 27.1 µg/mg, P<0.001) compared to placebo, consistent with previous results for patients with type 2 diabetes taking salsalate for shorter times. CONCLUSIONS: Salsalate improves glycemia in obese persons at increased risk for diabetes, and hence may decrease risk of incident type 2 diabetes. Salsalate may inform new therapeutic approaches for diabetes prevention, but renal safety may limit clinical utility.

Effect of Targeting Inflammation With Salsalate: The TINSAL-CVD Randomized Clinical Trial on Progression of Coronary Plaque in Overweight and Obese Patients Using Statins.

IMPORTANCE: Inflammation may contribute to pathological associations among obesity, diabetes mellitus, and cardiovascular disease. OBJECTIVE: To determine whether targeting inflammation using salsalate compared with placebo reduces progression of noncalcified coronary artery plaque. DESIGN, SETTING, AND PARTICIPANTS: In the Targeting Inflammation Using Salsalate in Cardiovascular Disease (TINSAL-CVD) trial participants were randomly assigned between September 23, 2008, and July 5, 2012, to 30 months of salsalate or placebo in addition to standard, guideline-based therapies. Randomization was computerized and centrally allocated, with patients, health care professionals, and researchers masked to treatment assignment. Participants were overweight and obese statin-using patients with established, stable coronary heart disease. INTERVENTIONS: Salsalate (3.5 g/d) or placebo orally over 30 months. MAIN OUTCOMES AND MEASURES: The primary outcome was progression of noncalcified coronary artery plaque assessed by multidetector computed tomographic angiography. Secondary outcomes were other measures of safety and efficacy. RESULTS: Two hundred fifty-seven participants were randomized to salsalate (n = 129) or placebo (n = 128). Their mean (SD) age was 60.8 (7.0) years, and 94.0% (236 of 251) were male. One hundred ninety participants (89 in the salsalate group and 101 in the placebo group) completed the study. Compared with baseline, there was no increase in noncalcified plaque volume in the placebo-treated patients and no difference in change between the salsalate and placebo groups (mean difference, -1 mm3; 95% CI, -11 to 9 mm3; P = .87). Salsalate treatment decreased total white blood cell, lymphocyte, monocyte, and neutrophil counts and increased adiponectin levels without change in C-reactive protein levels. Fasting glucose, triglycerides, uric acid, and bilirubin levels were decreased in the salsalate group compared with the placebo group, while hemoglobin levels were increased. Urinary albumin levels increased, with tinnitus and atrial arrhythmias more common, in the salsalate group compared with the placebo group. CONCLUSIONS AND RELEVANCE: Salsalate when added to current therapies that include a statin does not reduce progression of noncalcified coronary plaque volume assessed by multidetector computed tomographic angiography in statin-using patients with established, stable coronary heart disease. The absence of progression of noncalcified plaque volume in the placebo group may limit interpretation of the trial results. TRIAL REGISTRATION: clinicaltrials.gov Identifier: NCT00624923.

Incidental Findings on Bedside Ultrasonography: Detection Rate and Accuracy of Resident-Performed Examinations in the Acute Setting.

BACKGROUND: Bedside ultrasonography has become a valuable tool in the emergent care setting for triage and rapid evaluation of patients who are acutely ill. Given the cross-sectional nature of ultrasound technology, incidental findings are frequently encountered during imaging. These can impact clinical management and can pose a diagnostic dilemma for emergency medicine (EM) physicians and EM residents. PURPOSE: Our retrospective study was designed to evaluate the prevalence and detection rate of incidental findings on bedside ultrasound examinations performed by EM residents. We also sought to identify types of incidental findings encountered and the diagnostic accuracy of those findings. MATERIALS AND METHODS: Board-certified radiologists retrospectively reviewed bedside ultrasonography examinations performed and interpreted by EM residents at a large urban academic hospital. Our sample included patients who presented with traumatic and nontraumatic symptoms in the acute setting. Findings were defined as incidental only if they were previously unknown and not related to a patients presenting symptoms. The results were corroborated with electronic medical records and additional pertinent imaging when available. RESULTS: Of 196 examinations analysed, EM residents identified incidental findings on 26% of the studies, which mostly involved the renal and biliary system. Radiologist review detected incidental findings in 20.9% but was more accurate when supplemental imaging was available. EM residents detected incidental findings at rates similar to that published previously and had moderate interobserver agreement with radiologist review. Worrisome and indeterminate findings were confirmed by additional work-up and further imaging. CONCLUSION: Incidental findings are frequently encountered on bedside ultrasonography and have the potential to alter clinical management. Expertise in detection and knowledge of the presence and spectrum of these incidental findings is essential for appropriate triage, patient management, and follow-up.

Association of lumbar fractures, abdominal aortic calcification, and osteopenia.

PURPOSE: The purpose was to assess if abdominal aortic calcification (AAC) and low bone mineral density (BMD) are associated with fractures on lumbar spine radiographs in trauma patients. METHODS: Retrospectively, 303 consecutive lumbar radiographs were independently reviewed by two radiologists for AAC, low BMD, and traumatic findings. RESULTS: Thirty-one percent of patients had low BMD, 34% had AAC, and 24% had both. Eleven percent of radiographs showed traumatic findings. Seventy-six percent of positive cases had low BMD (P<.001), and 64% had AAC (P<.001). CONCLUSION: A higher index of suspicion for fractures is warranted when AAC and low BMD are present.

Factors affecting patient compliance in the acute setting: an analysis of 20,000 imaging reports.

The aim of this study was to identify the risk factors associated with noncompliance of recommendations made by emergency radiologists. Between March 2012 and August 2012, our team retrospectively reviewed 20,000 consecutive emergency department (ED) patients receiving imaging to assess how often emergency radiologists made recommendations, how often they were followed, and what factors were associated with noncompliance. Faculty (Radiology or Emergency Medicine) were considered senior if they had been practicing for >5 years post residency/fellowship training. Faculty practicing <5 years were considered junior physicians. The following data was extracted from the electronic medical records and the hospital information system: recommendation in imaging report, age, gender, race (Caucasian, African-American, and others), insurance status, primary care contact, distance from the hospital to patient residence, and primary language. Recommendations were categorized as follows: (1) immediate follow-up, (2) follow-up 1-4 weeks, (3) follow-up 1-3 months, (4) follow-up 4-6 months, (5) follow-up 7-12 months, and (6) clinical/laboratory follow-up recommendations irrespective of time. We identified 1,650 recommendations (1,650/20,000 = 8.25 %). Using a one-to-one logistic regression analysis, the following factors were significant (P < 0.05) when evaluating noncompliance: increasing age, no primary care physician, lack of insurance, primary language other than English, increased distance from hospital, and extended follow-up interval. Noncompliance with recommended additional imaging (RAI) is multifactorial. Primary and/or referring physicians should take notice of the aforementioned compliance trends and mitigating factors, adopt systematic safety measures and create interdepartmental dialogue with radiology to ensure compliance, and counsel and educate patients about the importance of imaging recommendations.

Incidental findings detected on abdomino-pelvic multidetector computed tomography performed in the acute setting.

OBJECTIVE: The objective of the study is to determine the prevalence and significance of incidental findings in patients with a chief complaint of abdominal pain presenting to the emergency department (ED) who received abdomino-pelvic multidetector computed tomography. MATERIALS AND METHODS: We conducted a retrospective review of data collected for 290 patients over a period of 5 months (April to September 2012) from 3 different university-affiliated EDs. Two board-certified radiologists reviewed the original images independently and recorded the incidental findings. These findings were classified as benign, indeterminate, and worrisome. Only those findings present in the original report were included in the study. If an indeterminate or worrisome incidental finding was identified, the patient's medical records were reviewed to determine if the incidental finding was previously known, whether recommendation was made for further evaluation, and whether this recommendation led to any change in management. RESULTS: We identified 283 incidental findings­144 benign (51%), 114 indeterminate (40%), and 25 worrisome (9%) findings. A statistically significant difference was observed in the percentage of patients who experienced a change in management among those who received recommendations as compared with those who did not, in both previously known (87% vs 22%, P=.001) and previously unknown (70% vs 2%, P=.001) indeterminate findings. CONCLUSION: Unlike benign incidental findings, indeterminate and worrisome findings frequently alter the course of management. Recommendation from radiologists appears to significantly contribute to the management of indeterminate incidental findings.

Ventriculoperitoneal shunt malfunction: cumulative effect of cost, radiation, and turnaround time on the patient and the health care system.

OBJECTIVE: The purpose of this study was to reiterate the predominance of CT in evaluating ventriculoperitoneal shunt malfunction in terms of cost-effectiveness, reduction of radiation exposure, and turnaround time as the measurement parameters. MATERIALS AND METHODS: This retrospective study included patients 18 years and older with a history of ventriculoperitoneal shunt insertion who presented to the emergency department with symptoms of shunt malfunction and underwent shunt series radiography and head CT within 12 hours. Shunt revision occurring contemporaneously with imaging was defined as revision within 48 hours of the original imaging report. The effective radiation dose was calculated by multiplying dose-length product from the scanner with the standard conversion coefficient k (k = 0.0021 mSv/mGy × cm). The turnaround time for patients who underwent both head CT and shunt series radiography was calculated from time of the first study to the time of completion of the last study. RESULTS: There were 16 shunt revisions in 239 patients. The sensitivity of CT was 87.5%; specificity, 91.4%; positive predictive value, 42.4%; and negative predictive value, 99%. The sensitivity of shunt series radiography was 18.7%; specificity, 90.9%; positive predictive value, 13%; and negative predictive value, 93.9%. There were 223 observations of CT radiation dose per patient (mean, 1.87 ± 0.45). There also were 223 observations of shunt radiography radiation dose per patient (mean, 1.57 ± 0.60). The median turnaround time among patients undergoing CT and shunt radiography was 109 ± 84 minutes. CONCLUSION: Shunt series radiography is a low-yield diagnostic imaging modality for identifying shunt malfunction and prolongs turnaround time, increases medical cost, and exposes patients to unnecessary radiation.