Can large language models pass official high-grade exams of the European Society of Neuroradiology courses? A direct comparison between OpenAI chatGPT 3.5, OpenAI GPT4 and Google Bard.

We compared different LLMs, notably chatGPT, GPT4, and Google Bard and we tested whether their performance differs in subspeciality domains, in executing examinations from four different courses of the European Society of Neuroradiology (ESNR) notably anatomy/embryology, neuro-oncology, head and neck and pediatrics. Written exams of ESNR were used as input data, related to anatomy/embryology (30 questions), neuro-oncology (50 questions), head and neck (50 questions), and pediatrics (50 questions). All exams together, and each exam separately were introduced to the three LLMs: chatGPT 3.5, GPT4, and Google Bard. Statistical analyses included a group-wise Friedman test followed by a pair-wise Wilcoxon test with multiple comparison corrections. Overall, there was a significant difference between the 3 LLMs (p < 0.0001), with GPT4 having the highest accuracy (70%), followed by chatGPT 3.5 (54%) and Google Bard (36%). The pair-wise comparison showed significant differences between chatGPT vs GPT 4 (p < 0.0001), chatGPT vs Bard (p < 0. 0023), and GPT4 vs Bard (p < 0.0001). Analyses per subspecialty showed the highest difference between the best LLM (GPT4, 70%) versus the worst LLM (Google Bard, 24%) in the head and neck exam, while the difference was least pronounced in neuro-oncology (GPT4, 62% vs Google Bard, 48%). We observed significant differences in the performance of the three different LLMs in the running of official exams organized by ESNR. Overall GPT 4 performed best, and Google Bard performed worst. This difference varied depending on subspeciality and was most pronounced in head and neck subspeciality.

Neuromuscular imaging in clinical practice: an ESNR survey of 30 centers.

PURPOSE: We assessed the current clinical imaging practice in the primary evaluation of neuromuscular disorders (NMD), with respect to standardized imaging, evaluation and reporting through a European and extra-European-wide survey. METHODS: An online questionnaire was emailed to all European Society of Neuroradiology (ESNR) members (n = 1662) who had expressed their interest in NMD. The questionnaire featured 40 individual items. Information was gathered on the context of the practices, available and preferred imaging modalities, applied imaging protocols and standards for interpretation, reporting and communication. RESULTS: A total of 30 unique entries from European and extra-European academic and non-academic institutions were received. Of these, 70% were neuroradiologists, 23% general radiologists and 7% musculoskeletal radiologists. Of the 30 responding institutes, 40% performed from 20 to 50 neuromuscular scans per year for suspected NMD. The principal modality used for a suspected myopathy was magnetic resonance imaging (MRI) (50%) or "mainly MRI" (47%). The primary imaging modality used for the evaluation of patients suspected of a neuropathy was MRI in 63% of all institutions and "mainly MRI" in 37%. For both muscle and nerve pathology, pelvic girdle and inferior limbs are the most scanned parts of the body (28%), followed by the thigh and leg (24%), whole body MR (24%), scapular girdle (16%), and the thigh in just 8% of institutions. Multiplanar acquisitions were performed in 50% of institutions. Convectional sequences used for muscle MRI included T2-STIR (88%), 2D T1weighted (w) (68%), T1 Dixon or equivalent (52%), T2 Dixon (40%), DWI (36%), 2D T2w (28%), T1 3D and T2 3D (20% respectively). For nerve MRI conventional sequences included T2-STIR (80%), DWI (56%), T2 3D (48%), 2D T2w (48%), T1 3D (44%), T1 Dixon or equivalent (44%), 2D T1 (36%), T2 Dixon (28%). Quantitative sequences were used regularly by 40% respondents. While only 28% of institutions utilized structured reports, a notable 88% of respondents expressed a desire for a standardized consensus structured report. Most of the respondents (93%) would be interested in a common MRI neuromuscular protocol and would like to be trained (87%) by the ESNR society with specific neuromuscular sessions in European annual meetings. CONCLUSIONS: Based on the survey findings, we can conclude that the current approach to neuromuscular imaging varies considerably among European and extra-European countries, both in terms of image acquisition and post-processing. Some of the challenges identified include the translation of research achievements (related to advanced imaging) into practical applications in a clinical setting, implementation of quantitative imaging post-processing techniques, adoption of structured reporting methods, and communication with referring physicians.

Adoption of AI in Oncological Imaging: Ethical, Regulatory, and Medical-Legal Challenges.

Artificial Intelligence (AI) algorithms have shown great promise in oncological imaging, outperforming or matching radiologists in retrospective studies, signifying their potential for advanced screening capabilities. These AI tools offer valuable support to radiologists, assisting them in critical tasks such as prioritizing reporting, early cancer detection, and precise measurements, thereby bolstering clinical decision-making. With the healthcare landscape witnessing a surge in imaging requests and a decline in available radiologists, the integration of AI has become increasingly appealing. By streamlining workflow efficiency and enhancing patient care, AI presents a transformative solution to the challenges faced by oncological imaging practices. Nevertheless, successful AI integration necessitates navigating various ethical, regulatory, and medical-legal challenges. This review endeavors to provide a comprehensive overview of these obstacles, aiming to foster a responsible and effective implementation of AI in oncological imaging.