Radiology Residents Research Articles

Factors for which radiology residents choose fellowship

PurposeThe objective of this study is to assess factors that influence radiology residents’ decision to pursue fellowship. Historically a large majority of Diagnostic Radiology (DR) residents have pursued fellowship, but with changes in the job market and the Covid-19 pandemic, this study analyzes the current trends associated with radiology fellowship choice. Materials and methodsAn anonymous 28-question survey was constructed based on literature review and pilot feedback from university radiology residents. The survey was distributed through APDR and to all program coordinators to be distributed to residents. Demographic information and questions related to fellowship choice were assessed. The survey was conducted through RedCap and consisted of multiple choice and sliding scale questions. Results214 radiology residents responded, representing 4.6 % of US DR residents across 199 programs. The top fellowship choices included neuroradiology (20.5 %), musculoskeletal imaging (17.3 %), body imaging (16.8 %), and breast imaging (16.4 %). Most influential factors for fellowship selection were strong personal interest, enjoyable rotation, work hours, job security, and compensation. Least influential factors were research opportunities and specific group practice. ConclusionThe decision to pursue fellowship remains almost unanimous among US DR residents. Strong personal interest, enjoyable rotations, and favorable work hours were rated as the most important factors in the decision to pursue fellowship. Neuroradiology, musculoskeletal, and body imaging remained the most popular specialties, with a notable increase in interest in breast radiology compared to literature. To attract prospective fellows, residency and fellowship programs should emphasize aforementioned factors and offer more early exposure to subspecialties during residency.

Tendinous Signal Alterations on MRI in the Asymptomatic Elbow: A Retrospective Cross-Sectional Study.

It is clinically relevant to prevent overtreatment of tendinopathy diagnosed solely on imaging. Therefore, the prevalence of presumable asymptomatic signal changes in the common flexor origin, biceps insertion, brachialis insertion, and triceps insertion were assessed. Two hundred and five magnetic resonance imaging (MRI) exams of the elbow with coronal and axial fat-saturated fluid-sensitive sequences between January 1, 2018 and July 31, 2022 were retrospectively identified in our center.Two radiology residents reviewed the exams independently. The elbow tendons were given a score from 0 to 4. Score 0: no signal abnormality; score 1: increased T2-weighted signal around the tendon; score 2: increased T2-weighted signal compared to muscle within the tendon; score 3: partial tear; and score 4: complete tear. The common flexor tendon showed signal alterations in 8% of patients; nine patients had an increased signal around the tendon, and eight patients had an increased signal within the tendon. Three patients (1.5%) had an altered signal intensity in the biceps tendon. All triceps tendons showed a linear hyperintense signal, suggesting that it is physiological. There were no partial or complete tears. No signal abnormalities were noted in the brachialis tendon among all patients. The prevalence of presumable asymptomatic signal alterations seen in the common flexor origin on MRI is not negligible; therefore, clinical correlation is advised to prevent overtreatment of tendinopathy in these cases. No partial or complete tears were seen.

Assessing the Capability of ChatGPT, Google Bard, and Microsoft Bing in Solving Radiology Case Vignettes.

Background The field of radiology relies on accurate interpretation of medical images for effective diagnosis and patient care. Recent advancements in artificial intelligence (AI) and natural language processing have sparked interest in exploring the potential of AI models in assisting radiologists. However, limited research has been conducted to assess the performance of AI models in radiology case interpretation, particularly in comparison to human experts. Objective This study aimed to evaluate the performance of ChatGPT, Google Bard, and Bing in solving radiology case vignettes (Fellowship of the Royal College of Radiologists 2A [FRCR2A] examination style questions) by comparing their responses to those provided by two radiology residents. Methods A total of 120 multiple-choice questions based on radiology case vignettes were formulated according to the pattern of FRCR2A examination. The questions were presented to ChatGPT, Google Bard, and Bing. Two residents wrote the examination with the same questions in 3 hours. The responses generated by the AI models were collected and compared to the answer keys and explanation of the answers was rated by the two radiologists. A cutoff of 60% was set as the passing score. Results The two residents (63.33 and 57.5%) outperformed the three AI models: Bard (44.17%), Bing (53.33%), and ChatGPT (45%), but only one resident passed the examination. The response patterns among the five respondents were significantly different ( p = 0.0117). In addition, the agreement among the generative AI models was significant (intraclass correlation coefficient [ICC] = 0.628), but there was no agreement between the residents (Kappa = -0.376). The explanation of generative AI models in support of answer was 44.72% accurate. Conclusion Humans exhibited superior accuracy compared to the AI models, showcasing a stronger comprehension of the subject matter. All three AI models included in the study could not achieve the minimum percentage needed to pass an FRCR2A examination. However, generative AI models showed significant agreement in their answers where the residents exhibited low agreement, highlighting a lack of consistency in their responses.

Musculoskeletal Radiology Education: A National Survey by the Italian College of Musculoskeletal Radiology.

Our aim was to understand how musculoskeletal training is structured in Italian residency programmes and the needs of young trainees. We sent out an online questionnaire (17 questions) to Italian Society of Radiology residents and board-certified radiologists aged up to 39 years. A total of 1144 out of 4210 (27.2%) members participated in the survey; 64.7% were residents and 35.3% were board-certified radiologists. Just 26.6% of participants had dedicated rotations for musculoskeletal training during their residency, although this percentage substantially increased in replies from northern Italy. One-fourth of residents had a scheduled period of musculoskeletal ultrasound. Most participants (76.3%) had <20 h per year of musculoskeletal lessons. The majority considered their musculoskeletal education poor (57.7%) or average (21.9%). According to 84.8% of replies, no dedicated training period about interventional musculoskeletal procedures was scheduled. Further, just 12.8% of residents took active part in such interventions. Nearly all participants believed that the musculoskeletal programme during residency needs to be improved, particularly concerning practices in ultrasound (92.8%), MRI cases interpretation/reporting (78.9%), and practice in ultrasound-guided interventional procedures (64.3%). Despite some differences in the structure of musculoskeletal education provided by different regions, there is a shared demand for improvement in musculoskeletal training.

Hands-on Fluoroscopy Safety Training for Radiology Residents: Assessment of Educational and Safety Perception Outcomes

Hands-on Fluoroscopy Safety Training for Radiology Residents: Assessment of Educational and Safety Perception Outcomes

Radiology residency oversight: A Qui Tam wake-up call

Like every physician practice, academic radiology practices must pay heed to all governmental regulations. The federal False Claims Act serves to protect US taxpayers and requires strict adherence. Violations, often brought forth by whistleblowers, can carry steep financial repercussions.

A Visual Aid Tool for Detection of Pancreatic Tumour-Vessel Contact on Staging CT: A Retrospective Cohort Study.

Purpose: In pancreatic adenocarcinoma, the difficult distinction between normal and affected pancreas on CT studies may lead to discordance between the pre-surgical assessment of vessel involvement and intraoperative findings. We hypothesize that a visual aid tool could improve the performance of radiology residents when detecting vascular invasion in pancreatic adenocarcinoma patients. Methods: This study consisted of 94 pancreatic adenocarcinoma patient CTs. The visual aid compared the estimated body fat density of each patient with the densities surrounding the superior mesenteric artery and mapped them onto the CT scan. Four radiology residents annotated the locations of perceived vascular invasion on each scan with the visual aid overlaid on alternating scans. Using 3 expert radiologists as the reference standard, we quantified the area under the receiver operating characteristic curve to determine the performance of the tool. We then used sensitivity, specificity, balanced accuracy ((sensitivity + specificity)/2), and spatial metrics to determine the performance of the residents with and without the tool. Results: The mean area under the curve was 0.80. Radiology residents' sensitivity/specificity/balanced accuracy for predicting vascular invasion were 50%/85%/68% without the tool and 81%/79%/80% with it compared to expert radiologists, and 58%/85%/72% without the tool and 78%/77%/77% with it compared to the surgical pathology. The tool was not found to impact the spatial metrics calculated on the resident annotations of vascular invasion. Conclusion: The improvements provided by the visual aid were predominantly reflected by increased sensitivity and accuracy, indicating the potential of this tool as a learning aid for trainees.

The Undermined ACGME Subcompetency: A Roadmap for Radiology Residency Programs to Foster Residents-as-Educators

Radiology Residency programs in the United States use a set of six core competencies as laid out by the Accreditation Council for Graduate Medical Education (ACGME) to evaluate the foundational skills of every resident. Despite the fact that educational skills are included under the heading of Practice-Based Learning and Improvement in the ACGME guidelines for radiology residents, it is often underappreciated and undervalued, when compared with medical knowledge or patient care. In this paper, the authors lay out the important role of residents-as-educators and how it can be inculcated as part of formal training during residency. They enunciate five pillars for academic programs to build and maintain the pedagogical skills of their radiology residents: Training, Practicing, Providing Feedback, Mentoring, and Changing the Culture. The authors believe that implementing this will holistically benefit radiology residents as well as radiology in building future educators. The authors also delineate the challenges that programs currently face in implementation and ways to overcome them.

Using AI to Improve Radiologist Performance in Detection of Abnormalities on Chest Radiographs.

Background Chest radiography remains the most common radiologic examination, and interpretation of its results can be difficult. Purpose To explore the potential benefit of artificial intelligence (AI) assistance in the detection of thoracic abnormalities on chest radiographs by evaluating the performance of radiologists with different levels of expertise, with and without AI assistance. Materials and Methods Patients who underwent both chest radiography and thoracic CT within 72 hours between January 2010 and December 2020 in a French public hospital were screened retrospectively. Radiographs were randomly included until reaching 500 radiographs, with about 50% of radiographs having abnormal findings. A senior thoracic radiologist annotated the radiographs for five abnormalities (pneumothorax, pleural effusion, consolidation, mediastinal and hilar mass, lung nodule) based on the corresponding CT results (ground truth). A total of 12 readers (four thoracic radiologists, four general radiologists, four radiology residents) read half the radiographs without AI and half the radiographs with AI (ChestView; Gleamer). Changes in sensitivity and specificity were measured using paired t tests. Results The study included 500 patients (mean age, 54 years ± 19 [SD]; 261 female, 239 male), with 522 abnormalities visible on 241 radiographs. On average, for all readers, AI use resulted in an absolute increase in sensitivity of 26% (95% CI: 20, 32), 14% (95% CI: 11, 17), 12% (95% CI: 10, 14), 8.5% (95% CI: 6, 11), and 5.9% (95% CI: 4, 8) for pneumothorax, consolidation, nodule, pleural effusion, and mediastinal and hilar mass, respectively (P < .001). Specificity increased with AI assistance (3.9% [95% CI: 3.2, 4.6], 3.7% [95% CI: 3, 4.4], 2.9% [95% CI: 2.3, 3.5], and 2.1% [95% CI: 1.6, 2.6] for pleural effusion, mediastinal and hilar mass, consolidation, and nodule, respectively), except in the diagnosis of pneumothorax (-0.2%; 95% CI: -0.36, -0.04; P = .01). The mean reading time was 81 seconds without AI versus 56 seconds with AI (31% decrease, P < .001). Conclusion AI-assisted chest radiography interpretation resulted in absolute increases in sensitivity for all radiologists of various levels of expertise and reduced the reading times; specificity increased with AI, except in the diagnosis of pneumothorax. © RSNA, 2023 Supplemental material is available for this article.

Reliability and clinical validity of the Kang MRI grading system for cervical central spinal stenosis

BackgroundThe Kang MRI grading system for central cervical spinal stenosis has been previously explored in literature. The aim of the current study was to re-visit the system before adopting it into our institute’s cervical spine MRI reports and to evaluate its reliability through measuring the degree of agreement between radiologists of different levels of experience. The secondary aim was to evaluate its clinical validity by correlating between the radiological grade and the clinical symptoms as well as the subsequent treatment received by the patients.ResultsA total of 82 patients (21 males and 61 females, mean age = 42.8 years) that underwent cervical spine MRI in our institute were included in the study. Two specialized musculoskeletal radiologists and two radiology residents at different levels of experience independently graded the central cervical spinal stenosis at the narrowest level using the Kang grading system. Substantial agreement was found between all the participating radiologists (K = 0.665–0.799, percentage of agreement = 78.04–86.58%). Correlation between radiological grade and the presence of neuropathic symptoms (upper extremity paresthesias and/or numbness), motor affection (upper extremity weakness and/or weak hand grip), and treatment received by the patients was moderate, with R ranging from 0.325–0.440, 0.420–0.506, and 0.452–0.592 respectively. Correlation between grade and the presence of cervical pain was weak, with R = 0.197–0.253. On subdividing the grade 2 patients into grade 2a (demonstrable spinal cord deformity at the anterior or posterior aspects of the cord only) and grade 2b (demonstrable spinal cord deformity at both the anterior and posterior aspects of the cord), a statistically significant difference was found regarding the prevalence of motor affection among the patients in both sub-groups (P < 0.001).ConclusionDespite being reliable and clinically valid, the Kang grading system needs revision of the use of sagittal T2 images only as well as the definition of grade 2. We showed that the current definition of grade 2 to be rather broad, putting a wide array of patients into a single category. Sub-dividing it might increase the accuracy of the grading system.

Radiology resident competency in detecting basilar artery occlusion: a simulation-based assessment.

Basilar artery strokes are rare but can have characteristic imaging findings that can often be overlooked. This retrospective study aims to assess radiology residents' ability to identify CT imaging findings of basilar artery occlusion in a simulated call environment. The Wisdom in Diagnostic Imaging Emergent/Critical Care Radiology Simulation (WIDI SIM)-a tested and reliable computer-aided emergency imaging simulation-was employed to assess resident readiness for independent radiology call. The simulations include 65 cases of varying complexity, including normal studies, with one case specifically assessing basilar artery stroke. Residents were presented with a single, unique case of basilar artery occlusion in two separate years of testing and were only provided with non-contrast CT images. Residents' free text responses were manually scored by faculty members using a standardized grading rubric, with errors subsequently classified by type. A total of 454 radiology residents were tested in two separate years on the imaging findings of basilar artery occlusion using the Wisdom in Diagnostic Imaging simulation web-based testing platform. Basilar artery occlusion was consistently underdiagnosed by radiology residents being tested for call readiness irrespective of the numbers of years in training. On average, only 14% of radiology residents were able to correctly identify basilar artery occlusion on non-contrast CT. Our findings underscore a potential gap in radiology residency training related to the detection of basilar artery occlusion, highlighting the potential need for increased educational efforts in this area.

Iodine density mapping for the diagnosis of acute bowel ischemia using fast kV-switching dual-energy CT

PurposeTo evaluate the diagnostic capabilities of a supplementary color ramped iodine density map compared to virtual monoenergetic images (VMIs) at 74 keV in the diagnosis of acute bowel ischemia (ABI).MethodsData for this study were prospectively gathered and retrospectively evaluated. Patients referred to the Department of Diagnostic Radiology between October 2020 and August 2022 on the suspicion of ABI and underwent surgery < 12 h following fast kV-switching venous phase abdominal dual-energy CT (DECT) were consecutively included. Images were evaluated by two board-certified radiologists and two radiology residents. First round included only 74 keV VMIs resembling conventional 120 kVp images, and the second round included a supplementary iodine density map. Readers were asked to register presence of ABI as well as their confidence in their diagnosis based on a 5-point Likert scale. Sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were calculated for each observer with the surgical findings as the gold-standard. McNemar’s and Wilcoxon signed-rank test were used to compare registrations and diagnostic confidence across assessment rounds.ResultsA total of 29 patients resulting in 31 DECT scans were included. Fourteen cases of ischemic/necrotic bowel were reported following surgery. Sensitivity and NPV were decreased with the use of supplementary iodine map images compared to 120 kVp-like images without supplementary iodine map images for three of four observers (round 1 range: 71.4–92.9% and 78.0–94.8%; round 2 range: 57.1–78.6% and 70.1–83.3%, respectively), while specificity and PPV were increased for three of four observers (round 1 range: 64.7–94.1% and 67.4–93.1%; round 2 range: 88.2–94.1% and 73.8–91.1%, respectively). However, no significant difference in ABI diagnosis or diagnostic confidence was found (p-value range: 0.07–1.00 and 0.23–0.58, respectively).ConclusionNo significant difference for the diagnosis of ABI was found using supplementary iodine mapping. Our study may suggest a trend of increased specificity and decreased sensitivity, hence, the use of supplementary iodine mapping should be carefully considered.

Wellness Among Interventional Radiologists: Results From a Multidimensional Survey

To evaluate wellness among interventional radiologists using a multidimensional survey. An anonymous 53-item survey, including 36 Perceived Wellness Survey (PWS) prompts, was created in Qualtrics (an online survey tool) to assess wellness among interventional radiology attendings and residents. The survey was open from June toSeptember, 2022, 2 years into the COVID-19 Pandemic. The survey was distributed via Society of Interventional Radiology Forums, social media (Twitter, LinkedIn, and Facebook), and personal correspondence. PWS scores were categorized into Wellness Composite and subscores (physical, emotional, intellectual, psychological, social, and spiritual). 367 surveys were completed. 300 (81.7%) respondents were male and 67 (18.3%) were female. Respondents included attending physicians (297; 81.0%) and residents (70; 19.0%). Practice settings included academic (174; 47.4%), private (114; 31.0%), private-academic hybrid (62; 16.9%), and others (17; 4.7%). Mean Wellness Composite at academic centers (14.74±3.16; range: 5.42-23.50) was significantly lower than at private (15.22±3.37; range: 6.13-23.51) and hybrid (16.14±2.47; range: 10.20-23.50) practices (p=0.01). Respondents <40 years old had significantly lower emotional wellness subscores compared to those ≥40 years old (4.34±0.86 vs. 4.60±0.87; p=0.006). There were no significant differences between geographic regions in the United States, however, International respondents had significantly lower social and intellectual wellness (INT) subscores. Physical wellness subscore (4.00±0.9) was significantly lower than the other subscores (4.59±0.81) (p<.001). Overall mean Wellness Composite was 15.11±3.13 (range: 5.42-23.51). Overall self-reported wellness was lower among interventional radiologists practicing at academic centers. Interventional Radiologists <40 years old and residents had lower emotional wellness, while international respondents had lower social and INT. Overall wellness scores were lower than prior PWS studies.

Interactive Web-based Emergency Neuroradiology Course with Self-Assessment for Radiology Residents: A Pilot Project.

To assess the effectiveness of an online, case-based interactive course in emergency neuroradiology to prepare radiology residents for night call in neuroradiology. A total of 15 residents participated in a pretest assessment of preparedness for neuroradiology call. After completing a 6-week interactive course incorporating case review, didactic lectures, quiz feedback, and references for further review, the same residents were quizzed on cognitive questions and feelings of readiness to enter the on-call pool for neuroradiology. Knowledge and confidence both significantly increased due to the course. Knowledge-wise, the scores for precourse quiz to postcourse quiz went from 18.4%-72.2% (53.8% increase) in Brain Imaging and went from 22.3%-77.1% (54.8% increase) in Head, Neck, and Spine (P < 0.001 for both). Confidence-wise, residents demonstrated statistically significant increases in all 6 confidence measures. Prior to the course, 29% were not confident, 71% were fairly confident, and 0% were confident/very confident. After the course, 0% were not confident, 43% were fairly confident, and 57% were confident/very confident. Belief in the statement "I can provide high quality Neuroimaging services in the emergency care setting" increased from a confidence score of 1.29-2.57 after training (P=0.004). Nearly all residents completing their first emergency call reported that they felt more confident reading neuroradiology studies during their call as a direct result of the course. Completing the multi-pronged interactive, case-based online emergency neuroradiology course led to improved funds of knowledge and feelings of confidence and impacted imaging approach in residents taking neuroradiology call.

The Influence of Extracurricular Activities on Radiology Resident Selection Decisions

PurposeExtracurricular activities (EAs) listed on radiology residency applications can signal traits and characteristics desired in holistic reviews. The authors conducted an objective analysis to determine the influence of EAs on resident selection decisions. MethodsA discrete-choice experiment was designed to model radiology resident selection and determine the relative weights of EAs among academic and demographic application factors. Faculty members involved in resident selection at 30 US radiology programs chose between hypothetical pairs of applicant profiles between October 2021 and February 2022. Each applicant profile included one of 22 EAs chosen for study. A conditional logistic regression model assessed the relative weights of the attributes and odds ratios (ORs) were calculated. ResultsTwo hundred forty-four participants completed the exercise. Community-service EAs were ranked most highly by participants. LGBTQ Pride Alliance (OR, 1.56; 95% confidence interval [CI], 1.14-2.15; P = .006) and Young Republicans (OR, 0.60; 95% CI, 0.43-0.82; P = .001) significantly influenced decisions. The highest ranked EAs were significantly preferred over the lowest ranked EAs (OR, 1.916; 95% CI, 1.671-2.197; P < .001). Participants preferred EAs that reflected active over passive engagement (OR, 1.154; 95% CI, 1.022-1.304; P = .021) and progressive over conservative ideology (OR, 1.280; 95% CI, 1.133-1.447; P < .001). Participants who ranked progressive EAs more highly preferred applicants with progressive EAs (P < .05 for all). ConclusionsThe influence of EAs on resident selection decisions is significant and likely to gain importance in resident selection as medical student performance metrics are further eliminated. Applicants and selection committees should consider this influence and the bias that EAs can bring to resident selection decisions.

Interventional Neuroradiology: Why Don't Brazilian Female Physicians Like It?

Abstract Objective Much has been discussed about gender diversity in all professional fields, particularly in medicine. Recent studies showing that there are fewer than ten female professionals in interventional neuroradiology (INR) in a continental country like Brazil, and less than thirty in all of Latin America, have prompted an investigation into the causes of the low numbers of women choosing to follow this medical specialty. This project intended to reveal the reasons through an anonymous form applied to women in the final year of the three medical residencies considered prerequisites for INR in Brazil: neurology, radiology and neurosurgery. Materials and Methods The questionnaire addressed to each of the three fields contemplated the professionals' preference to follow their respective subspecialties, in addition to common considerations that could be deemed barriers to INR. Results Surprisingly, the weak link in the chain was demonstrated to be radiology, which is the exclusive prerequisite residency for training in INR in many countries. However, in Brazil, most doctors who graduate in INR come from Neurosurgery and, secondly, from neurology. These two specialties together account for less than half of the female residents compared to radiology alone. Conclusion All of the following items were already expected: difficulty in reconciling a double shift; employment opportunity; long training time; the lack of female leaders inspiring new generations of doctors; issues involving radiation and motherhood; and still the barriers of misogyny and machismo. But specific studies need to be conducted to find out why very few radiology residents follow INR training in Brazil.

Hospital payment systems and physician reimbursement: A primer for radiology residents

Prior to practicing independently, radiology trainees spend considerable time and energy to learn and understand the practice of radiology. However, upon graduation trainees are often deficient in understanding the business of hospitals and the structure of reimbursement. Specifically, the workflow of relative value units (RVUs) and its impact on practice of radiology after completing training. In this manuscript, we provide a resource for trainees to understand the workflow of physician reimbursement. This article includes information on the mixed model healthcare structure of the United States and two government programs that influence reimbursement: Diagnosis-Related Groups (DRG) and Hospital Value-Based Purchasing (HVBP) programs. Furthermore, we explain the method by which the Center of Medicare and Medicate Service's (CMS) reimburses physicians via the Medicare Physician Fee Schedule (MPFS) using the Resource Based Relative Value Scale. Understanding the structure of these payments along with the challenges and current landscape of radiology reimbursement will help new radiologists prior to seeking employment where reimbursements are integral to contract expectations.

Radiology Residents' Perceptions of Artificial Intelligence: Nationwide Cross-Sectional Survey Study.

Artificial intelligence (AI) is transforming various fields, with health care, especially diagnostic specialties such as radiology, being a key but controversial battleground. However, there is limited research systematically examining the response of "human intelligence" to AI. This study aims to comprehend radiologists' perceptions regarding AI, including their views on its potential to replace them, its usefulness, and their willingness to accept it. We examine the influence of various factors, encompassing demographic characteristics, working status, psychosocial aspects, personal experience, and contextual factors. Between December 1, 2020, and April 30, 2021, a cross-sectional survey was completed by 3666 radiology residents in China. We used multivariable logistic regression models to examine factors and associations, reporting odds ratios (ORs) and 95% CIs. In summary, radiology residents generally hold a positive attitude toward AI, with 29.90% (1096/3666) agreeing that AI may reduce the demand for radiologists, 72.80% (2669/3666) believing AI improves disease diagnosis, and 78.18% (2866/3666) feeling that radiologists should embrace AI. Several associated factors, including age, gender, education, region, eye strain, working hours, time spent on medical images, resilience, burnout, AI experience, and perceptions of residency support and stress, significantly influence AI attitudes. For instance, burnout symptoms were associated with greater concerns about AI replacement (OR 1.89; P<.001), less favorable views on AI usefulness (OR 0.77; P=.005), and reduced willingness to use AI (OR 0.71; P<.001). Moreover, after adjusting for all other factors, perceived AI replacement (OR 0.81; P<.001) and AI usefulness (OR 5.97; P<.001) were shown to significantly impact the intention to use AI. This study profiles radiology residents who are accepting of AI. Our comprehensive findings provide insights for a multidimensional approach to help physicians adapt to AI. Targeted policies, such as digital health care initiatives and medical education, can be developed accordingly.

Why Medical Students Pursue Radiology: A Current Longitudinal Survey on Motivations and Controversial Issues in Radiology

Radiology is an increasingly competitive specialty. Various current factors influence medical students' decision to pursue a radiology career, including artificial intelligence (AI), remote reading, and COVID-19. This study seeks to determine the decision-making factors of all alumni from our medical school who matched into a radiology residency, and to gather opinions on emerging radiology topics. A survey querying decision-making factors and opinions on current radiology topics was distributed to all alumni from our medical school (first graduating class in 2011) who previously matched into a diagnostic or interventional radiology residency program (n=57). Wilcoxon Rank-Sum and Fisher's Exact tests were used to determine statistical significance. Forty-three of fifty-seven responses were received (75% response rate). The most influential factor that sparked respondents' interest in radiology was a radiology elective (25/43, 58%). Students who will finish radiology training in 2023 or later were more likely to be influenced by a mentor (15/23, 65%) than those who finished radiology training before 2023 (5/20, 25%) (p=0.04). Respondents reported a 1.6/5 concern about AI negatively impacting their future career in radiology. There was 1.7/5 concern about performing radiology procedures on patients during the COVID-19 pandemic. Respondents predicted that remote reading would have a 3.2/5 positive impact on helping them achieve their preferred lifestyle. Job satisfaction among attending radiologists is rated at 4.3/5. Radiology electives had the greatest influence in piquing students' interest in radiology, while mentorship is assuming increasing influence. AI is perceived as a relatively minimal threat to negatively impact radiologists' jobs. Respondents had little concern about performing radiology procedures during the COVID-19 pandemic. Remote reading is viewed as having a moderately positive impact on lifestyle. Responding radiologists enjoy notably high job satisfaction.

Impact of deep learning on radiologists and radiology residents in detecting breast cancer on CT: a cross-vendor test study

To investigate the effect of deep learning on the diagnostic performance of radiologists and radiology residents in detecting breast cancers on computed tomography (CT). In this retrospective study, patients undergoing contrast-enhanced chest CT between January 2010 and December 2020 using equipment from two vendors were included. Patients with confirmed breast cancer were categorised as the training (n=201) and validation (n=26) group and the testing group (n=30) using processed CT images from either vendor. The trained deep-learning model was applied to test group patients with (30 females; mean age=59.2±15.8 years) and without (19 males, 21 females; mean age=64±15.9 years) breast cancer. Image-based diagnostic performance of the deep-learning model was evaluated with the area under the receiver operating characteristic curve (AUC). Two radiologists and three radiology residents were asked to detect malignant lesions by recording a four-point diagnostic confidence score before and after referring to the result from the deep-learning model, and their diagnostic performance was evaluated using jackknife alternative free-response receiver operating characteristic analysis by calculating the figure of merit (FOM). The AUCs of the trained deep-learning model on the validation and test data were 0.976 and 0.967, respectively. After referencing with the result of the deep learning model, the FOMs of readers significantly improved (reader 1/2/3/4/5: from 0.933/0.962/0.883/0.944/0.867 to 0.958/0.968/0.917/0.947/0.900; p=0.038). Deep learning can help radiologists and radiology residents detect breast cancer on CT.