RPS 305 - Career, workforce issues and radiologist visibility

Author Block: P. Brennan¹, Y. Hughes-Roberts², J. Richenberg², I. Francis², M. Suleiman¹; ¹Sydney/AU, ²West Sussex/UK
Purpose: To develop a geographically-limitless REALM infrastructure that facilitates an optimised and reflective educational activity.
Methods or Background: Reviewing radiological discrepancies is a well-known activity that promotes diagnostic excellence, encourages reflection and minimises future errors. Formalisation of discrepancy reviews can be seen with the Royal College of Radiology REALM (Radiology Events and Learning Meetings) program where radiologists in the UK can engage both as authors and recipients. However, to promote a clinically-realistic REALM activity, comprehensive radiologic interactions with full resolution anonymised images should occur in a geographically limitless way. The infrastructures to support this, are elusive.

An existing technology DxCARES was modified for this activity. This incorporated: multi-modality viewing capabilities and 3D and multi-planar reconstructions (MPR): advanced AI-powered anonymization and masking engines to automatically remove sensitive data while maintaining diagnostic integrity; a user-friendly interface for users with different levels of expertise; a web-based architecture supporting high performance DICOM streaming and real time interaction with large data sets.
Results or Findings: We have built a new technology which allows clinicians to upload REALM cases from PACs systems and distribute across a geographically-limitless health enterprise to limitless recipients. Customisable authorship of each case is available so that each creator can demand from recipients the type of case-specific interaction or reflection that is required, regardless of image or pathology-type. The multifunctional cloud-based viewer allows the examination and manipulation of all cases as would occur with a primary diagnostic workstation. All REALM outputs and interactions are available for review and CPD accumulation. The new technology is currently being implemented across clinical centres.
Conclusion: This new technical innovation should promote widespread clinically-realistic REALM engagement by simplifying time/cost implications. The potential of radiological discrepancy review should be maximised.
Limitations: N/A
Funding for this study: N/A
Has your study been approved by an ethics committee? Not applicable
Ethics committee - additional information: N/A

Author Block: S. Dijk¹, C. Wollny², T. Kroencke², M. G. M. Hunink¹; ¹Rotterdam/NL, ²Augsburg/DE
Purpose: We analyzed sex-related disparities in inappropriate imaging requests using data from the Medical Imaging Decision And Support (MIDAS) trial.
Methods or Background: This study analyzed baseline data from the MIDAS trial, a multi-center cluster randomized trial conducted at three German academic hospitals. The study population encompassed all imaging requests submitted to the 26 participating departments via the computerized order entry system during a 15-month period. Imaging appropriateness was assessed using the ESR iGuide, a clinical decision support system (CDSS). Requests were categorized as inappropriate if imaging was deemed unlikely to be indicated or if the potential risks outweighed the benefits for the patient. Chi-square tests were employed to compare the proportion of inappropriate imaging requests between men and women, with a significance level of 0.01. Secondary analyses explored differences in inappropriate requests by age group and exam type, with a Bonferroni correction applied to account for multiple testing.
Results or Findings: Women had more inappropriate imaging requests (7.32%) than men (6.08%; χ² = 37.176, p < 0.001, OR 1.22 [95%CI 1.12-1.33]). This disparity was particularly evident in the 25-65 age group and for MRI examinations. Further research is needed to explore the underlying causes of this discrepancy, including potential differences in physician awareness of guidelines, adherence to guidelines, or limitations in the CDSSs ability to account for female-specific factors.
Conclusion: In our study clinicians were 22% more likely to request inappropriate imaging for women than for men across nearly all modalities and age-groups. While the absolute percentage-point differences were small (1.24%), the disparity warrants further investigation
Limitations: Our analysis judges each request independently, without considering the cumulative impact of requests per individual, underuse, or time between symptom onset and imaging. Our data only distinguished male/female sexes.
Funding for this study: The MIDAS study was funded by the German Innovation Fund (reference: Förderkennzeichen 01VSF18008).
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: Approval from the Medical Ethics Review Committee was obtained under protocol numbers 20-069 (Augsburg), B 238/21 (Kiel), 20-318 (Lübeck) and 2020-15125 (Mainz). The trial is registered in the ClinicalTrials.gov register under registration number NCT05490290.

Author Block: P. Dankerl, J. Lang, A. Glaser, H. P. Beyer, M. Forsting; Dortmund/DE
Purpose: The increasing complexity and volume of radiological examinations have led to growing concerns about radiologists' workload, diagnostic efficiency and accuracy. The aim of this study was to evaluate and compare radiological workload in a multicenter and multimodality analysis while creating a benchmarking metric.
Methods or Background: Over 100 days radiologists’ reading times for all examinations and modalities across 34 centers have been collected utilizing RIS-export, while grouping these into 67 different body regions. In order to sort, retrieve and evaluate the various output formats from the RIS-data, a uniform nomenclature was introduced and all examinations received these additional unique identifying labels. For benchmarking, reporting times were translated into relative value units – as defined by us as the mean reporting time of all X-ray exams and termed RADPoints.
Results or Findings: We examined 290.748 examinations and found significant variations in average reporting times across modalities and body region, e.g. abdominal MRI 14.83min and CT 13.59min. Reporting times varied considerably, with the highest average times observed in complex examinations such as cardiac MRI (26.04min) and CT (16.59min). Conversely, regions like the fingers showed much lower averages, e.g. 10.95min for MR and 9.83min for CT. We found our relative value unit time of one RADPoint to equivalent to 2.23min which further served as common devisor in order to allocate specific RADPoints to all body region specific examinations.
Conclusion: The findings underscore the need for targeted workload management strategies in radiology departments, particularly when high-complexity cases are in the mix. Benchmarking reporting times across modalities and body regions by the means of presented RADPoints provides a critical reference for optimizing radiologist workload, potentially leading to enhanced diagnostic accuracy and efficiency.
Limitations: None
Funding for this study: None
Has your study been approved by an ethics committee? Not applicable
Ethics committee - additional information: Retrospective evaluation and blinding of patinet as well as doctor identyfiers makes this not applicable

Author Block: J. Gregory¹, M. L. Kofoed-Ottesen², B. Lindlbauer², C. Loewe², V. Vilgrain¹; ¹Clichy/FR, ²Vienna/AT
Purpose: This study evaluates the perceived impact of European School of Radiology (ESOR) training programs on radiologists' professional development.
Methods or Background: A cross-sectional survey targeted alumni who participated in ESOR fellowships from 2011 to 2023. The survey included questions on demographics, professional background, ESOR program details, and career impact. Data were collected via a web-based questionnaire and analyzed using descriptive statistics and thematic analysis.
Results or Findings: A total of 190 participants responded, with a median age of 31 years (range 29-33), and 54% were female. Most worked in public healthcare (62%) and were involved in academic activities (24%). Fellowship types included Visiting Scholarship Program (44%), Bracco Fellowship (32%), and Exchange Program for Fellowships (25%). The majority (59%) reported the fellowship helped them reach their current position, and 35% noted it upgraded their CV. Significant application of learned skills was reported by 69%. Ongoing cooperation with former tutors was maintained by 54%. Financial support was crucial, with 41% stating they could not have completed the training without it, 33% considering it very important, and 13% deeming it important. Participants rated the impact on clinical skills with a median score of 9 out of 10. Other areas of impact included research skills (median score 7), subspecialization (median score 9), exposure to diverse practices (median score 9), networking opportunities (median score 10), and personal and professional growth (median score 10).
Conclusion: ESOR training programs significantly enhance radiologists' professional development through comprehensive support, high-quality training, and substantial financial aid, ensuring participants are well-equipped for career advancement.
Limitations: This study has limitations, including reliance on self-reported data, potential recall-bias, and a 20% response rate. The survey may not fully represent all ESOR alumni, and program heterogeneity could influence the generalizability of results.
Funding for this study: None
Has your study been approved by an ethics committee? Not applicable
Ethics committee - additional information: Given the nature of the survey involving professional feedback without sensitive personal data, ethical approval was not required. However, all participants were informed about the purpose of the survey and the anonymous handling of their data.

Author Block: S. Faghani¹, C. Tiegs-Heiden², M. Moassefi², G. Powell², M. Ringler², B. J. Erickson², N. Rhodes²; ¹Minneapolis, MN/US, ²Rochester, MN/US
Purpose: To use artificial intelligence (AI) as a teaching tool to identify new imaging findings and improve the radiologist’s ability to recognize subtle imaging findings without additional AI assistance.
Methods or Background: We studied the learning of a new task by humans using a deep learning (DL) model that can identify sex differences from frontal knee radiographs with high accuracy. We then ascertained imaging features via occlusion interpretation maps (“heat maps”) to help human readers improve their ability to perform this task. Three human readers were tasked to classify 50 frontal knee radiographs into male and female sex. They were then informed of the patient’s sex and were given the associated AI-derived “heat maps” for subsequent study. After two weeks, the group was retested with a new set of 50 radiographs.
Results or Findings: The DL model categorized sex with 0.96 accuracy. The average accuracy of the 3 human readers was initially 0.59 (range: 0.56-0.66). After study of AI-derived “heat maps” and associated radiographs, the average accuracy of the human readers increased to 0.80 (range: 0.76-0.84), a statistically significant improvement (p=0.0270).
Conclusion: We believe this improvement serves as a proof of concept for the use of AI as a tool for discovery science to advance human learning, an idea that we have not seen advanced in the radiology literature.
Limitations: 1. This is an education item that does not fall well into an AI or MSK section.
2. Understanding how AI algorithms work is limited.
3. Our study is small. Large enough to prove an improvement after training, but too small to dissect more.
Funding for this study: None.
Has your study been approved by an ethics committee? Not applicable
Ethics committee - additional information: N/A

Author Block: S. Ramirez; Bogota/CO
Purpose: The preparation of a radiological report involves integrating knowledge, skills, and training. While many articles offer recommendations for quality reports, there is a lack of literature on teaching strategies for efficiently constructing these reports. Despite this, some residents and radiologists have developed this competency over time, revealing a gap in understanding how residents acquire the critical thinking skills needed. This work aims to determine how radiology residents learn to create their reports.
Methods or Background: A qualitative multi-case study was conducted involving three cases represented by residents of a radiology residency program, the program's faculty, and a group of specialists in high-quality report production. The study included document reviews of the syllabus, semi-structured interviews, and a focus group to capture experiences, perspectives, and learning processes, as well as the factors and strategies that influenced the development of the knowledge and skills needed for creating radiological reports.
Results or Findings: Information was collected on concepts, perceptions, beliefs, emotions, interactions, and experiences to construct coherent meanings about the topic. Common patterns and challenges in learning to create radiological reports were identified. Participants described the hidden curriculum as as the main component of the educational process through which residents learn to write reports. Preparing preliminary reports was regarded as the most valuable pedagogical activity for developing critical thinking skills. All residents agreed on the need to design educational activities focused on learning to prepare radiological reports.
Conclusion: This study represents, to our knowledge, the first research on how the critical thinking skills necessary for creating a radiological report are acquired. By understanding how a person effectively acquires a skill, teaching strategies can be designed and adapted to maximize meaningful learning.
Limitations: None
Funding for this study: No
Has your study been approved by an ethics committee? Yes
Ethics committee - additional information: Ethics committee Fundación Universitaria Sanitas

Author Block: M. Alvarez García, M. E. Pueyo, P. Boldó, C. Urtasun Iriarte, A. Ezponda Casajus, P. Chico, J. Pueyo Villoslada; Pamplona/ES
Purpose: This study aimed to explore the student satisfaction with Instagram as an educational tool in radiology education for fourth-year medical students, compared to a traditional learning platform.
Methods or Background: The study focused on gender differences and overall preferences in platform usage among fourth-year medical students (64% female). Both Instagram and the traditional platform were used to provide multiple-choice test questions with images based on radiology class content. Students chose their preferred platform for completing quizzes, and their experiences were assessed via anonymous satisfaction surveys. Statistical analysis explored gender-based differences in platform usage, preferences, and perceived learning benefits. The Mann-Whitney U test was applied to analyze the results.
Results or Findings: A total of 65.8% of participants chose Instagram as their preferred platform for answering clinical cases. No significant gender differences were observed in overall platform choice. The majority (71%) of students expressed moderate to high satisfaction with Instagram as a complementary learning tool. A significant gender difference (p=0.019) was found in the reasons for preferring Instagram: women favored its user-friendly interface, while men prioritized its speed. No significant differences were found between genders regarding Instagram's usefulness for personal study or whether it caused distractions (p=0.31). Additionally, 97.1% of students preferred Instagram over other social networks.
Conclusion: Instagram was generally preferred over traditional learning platforms and other social media platforms for radiology education, with a moderate to high level of student satisfaction.
Limitations: Certain students do not use Instagram. There might be missing data.
Instagram tool “Stories” only allows users (students) to see the content during a period of 24h.
Funding for this study: None
Has your study been approved by an ethics committee? Not applicable
Ethics committee - additional information: This study has no ethical implications

Author Block: L. Hartog¹, J. M. L. Bosmans², J. Bouziotis², P. M. Parizel³, A. Snoeckx²; ¹Antwerp/BE, ²Edegem/BE, ³Perth/AU
Purpose: Value-based radiology requires thorough understanding by patients of the role of radiologists in healthcare. News media frequently enlighten content on radiology with seemingly outdated images from stock photo databases. We conducted a critical evaluation of the ability of stock photos in three major databases to reliably represent contemporary radiology.
Methods or Background: We collected two hundred consecutive photos from three major commercial databases, using the keyword ‘radiologist’. Each image was evaluated for overall quality, image description, descriptive elements, displayed body regions, diagnostic or interventional context, depicted modality, inclusion of medical professionals, and accurate display on the viewing device.
Results or Findings: Just 6% of the stock photos received a high overall quality score. Radiographs were the most frequently depicted modality (52%), followed by MRI (21%) and CT (19%), with a focus primarily on diagnostic imaging (99%). On 4% of the stock photos, images were displayed on a PACS workstation, while 66% showed a radiologist holding up films ‘in the air’, and 10% on an X-ray viewbox. In 48% of the photos, the lead person was shown wearing a stethoscope. In all, only 9% of the stock photos presented a realistic view of a radiologist’s job.
Conclusion: The vast majority of stock photos fail to accurately portray the contemporary role of radiologists, contributing to gross misinterpretation of our profession by the general public.
Limitations: The evaluation was done by a radiology resident but reviewed by a consultant radiologist. To our knowledge, this kind of evaluation of an essential source of information for the public has never been undertaken before.
Funding for this study: No funding was received for this study.
Has your study been approved by an ethics committee? Not applicable
Ethics committee - additional information: No patient data was used for this study.