Artificial Intelligence Chatbots' Understanding of the Risks and Benefits of Computed Tomography and Magnetic Resonance Imaging Scenarios.

PURPOSE: Patients may seek online information to better understand medical imaging procedures. The purpose of this study was to assess the accuracy of information provided by 2 popular artificial intelligence (AI) chatbots pertaining to common imaging scenarios' risks, benefits, and alternatives. METHODS: Fourteen imaging-related scenarios pertaining to computed tomography (CT) or magnetic resonance imaging (MRI) were used. Factors including the use of intravenous contrast, the presence of renal disease, and whether the patient was pregnant were included in the analysis. For each scenario, 3 prompts for outlining the (1) risks, (2) benefits, and (3) alternative imaging choices or potential implications of not using contrast were inputted into ChatGPT and Bard. A grading rubric and a 5-point Likert scale was used by 2 independent reviewers to grade responses. Prompt variability and chatbot context dependency were also assessed. RESULTS: ChatGPT's performance was superior to Bard's in accurately responding to prompts per Likert grading (4.36 ± 0.63 vs 3.25 ± 1.03 seconds, P < .0001). There was substantial agreement between independent reviewer grading for ChatGPT (κ = 0.621) and Bard (κ = 0.684). Response text length was not statistically different between ChatGPT and Bard (2087 ± 256 characters vs 2162 ± 369 characters, P = .24). Response time was longer for ChatGPT (34 ± 2 vs 8 ± 1 seconds, P < .0001). CONCLUSIONS: ChatGPT performed superior to Bard at outlining risks, benefits, and alternatives to common imaging scenarios. Generally, context dependency and prompt variability did not change chatbot response content. Due to the lack of detailed scientific reasoning and inability to provide patient-specific information, both AI chatbots have limitations as a patient information resource.

Pancreatic Cystic Lesions on MRI: What Is The Likelihood of a Present or Future Diagnosis of Pancreatic Carcinoma?

BACKGROUND: Pancreatic cystic lesions (PCLs) are followed for years due to older and likely biased works demonstrating a strong association with pancreatic carcinoma; more recent data are needed clarifying this relationship. PURPOSE: To determine the association between PCLs on MRI and a synchronous or future diagnosis of pancreatic carcinoma. STUDY TYPE: Single-center retrospective cohort. POPULATION: A total of 192 patients (111 female, 58%) with median age 66 years (range 26-87 years) with PCLs on abdominal MRI from 2011 to 2016. FIELD STRENGTH/SEQUENCES: 1.5 T and 3 T, including T2 WI, T1 WI, diffusion weighted imaging and contrast-enhanced T1 WI. ASSESSMENT: Each PCL was reviewed independently by 2 of 10 fellowship-trained abdominal radiologists. Fukuoka guideline worrisome features and high-risk stigmata were evaluated. Follow-up imaging and clinical notes were reviewed within a system that captures pancreatic carcinoma for the region, for a median follow-up of 67 months (interquartile range: 43-88 months). STATISTICAL TESTS: Pancreatic carcinoma prevalence and incidence rate for future carcinoma with 95% confidence intervals (95% CI). Fisher exact test, logistic regression with odds ratios (OR) and the Wilcoxon rank-sum test were used to assess PCL morphologic features with the Kolmogorov-Smirnov test used to assess for normality. P < 0.05 defined statistical significance. RESULTS: The prevalence of pancreatic carcinoma on initial MRI showing a PCL was 2.4% (95% CI: 0.9%, 5.2%). Thickened/enhancing cyst wall was associated with pancreatic carcinoma, OR 52 (95% CI: 4.5, 1203). Of 189 patients with a PCL but without pancreatic carcinoma at the time of initial MRI, one developed high-grade dysplasia and none developed invasive carcinoma for an incidence rate of 0.97 (95% CI: 0.02, 5.43) and 0 (95% CI: 0, 3.59) cases per 1000 person-years, respectively. DATA CONCLUSION: A low percentage of patients with a PCL on MRI had a pancreatic carcinoma at the time of initial evaluation and none developed carcinoma over a median 67 months of follow-up. EVIDENCE LEVEL: 3 TECHNICAL EFFICACY: 5.

Pancreatic ductal adenocarcinoma (PDAC) regional nodal disease at standard lymphadenectomy: is MRI accurate for identifying node-positive patients?

OBJECTIVE: To determine the accuracy of qualitative and quantitative MRI features for the diagnosis of pathologic regional lymph nodes at standard lymphadenectomy in patients with pancreatic ductal adenocarcinoma (PDAC). METHODS: All adult patients with pancreatic MRI performed from 2011 to 2021 within 3 months of a pancreaticoduodenectomy were eligible for inclusion in this single-center retrospective cohort study. Regional nodes at standard lymphadenectomy were independently reviewed by two fellowship-trained abdominal radiologists for the following qualitative features: heterogeneous T2 signal, round shape, indistinct margin, peri-nodal fat stranding, and restricted diffusion greater than the spleen. Quantitative characteristics including primary tumor size, largest node short- and long-axes length, number of regional nodes, absolute apparent diffusion coefficient (ADC) values, and ADC node-to-spleen signal index were assessed. Analysis was at the patient-level with surgical pathology as the reference standard. RESULTS: Of 75 patients, 85% (64/75) were positive for regional nodal disease on histopathology. None of the qualitative variables evaluated on MRI was associated with pathologic nodes. Median primary tumor maximum diameter was slightly larger for patients with pathologic nodes compared to those without (18 mm (10-42 mm) vs 16 mm (9-22 mm), p = 0.027). None of the other quantitative features was associated with pathologic nodes. Radiologist opinion was not associated with pathologic nodes (p = 0.520). Interobserver agreement was fair (kappa = 0.257). CONCLUSIONS: Lymph node morphologic features and radiologist opinion using MRI are of limited value for diagnosing PDAC regional nodal disease. Improved diagnostic techniques are needed given the prognostic implications of pathologic lymph nodes in these patients. KEY POINTS: • Multiple lymph node morphologic features routinely assessed on MRI for malignancies elsewhere in the body are likely not applicable when assessing for pancreatic ductal adenocarcinoma nodal disease. • Interobserver agreement for the presence or absence of pancreatic ductal adenocarcinoma lymph node morphologic features on MRI is fair (kappa = 0.257). • Many more lymph nodes are resected at PDAC standard lymphadenectomy than are detectable on MRI, median 25 vs 5 (p < 0.001), suggesting improved diagnostic techniques are needed to identify PDAC nodal disease.

Quality Improvement Report: Adherence to Follow-up Recommendations for Incidental Abdominal Aneurysms.

The use of national guidelines for the management of incidental radiologic findings remains low. Therefore, improving adherence to and consistency with follow-up recommendations for incidental findings was undertaken in a large academic practice. A gap analysis was performed, and incidental findings of abdominal aneurysms for which reporting management recommendations could be improved were identified. The Kotter change management framework was used, and institution-specific dictation macros were developed and implemented in February 2021 for the management of abdominal aortic aneurysms (AAAs), renal artery aneurysms (RAAs), and splenic artery aneurysms (SAAs). A retrospective medical record review was conducted for February through April in 2019, 2020, and 2021 to assess reporting adherence and imaging and clinical follow-up. Personal feedback was provided to radiologists in July 2021 with repeat data collection in September 2021. A significant increase in the number of correct follow-up recommendations was reported for incidental AAAs and SAAs after implementation of the macro (P < .001). However, there was no significant change for RAAs. Providing personal feedback to radiologists further improved adherence with standard recommendation macros for common findings and dramatically increased adherence for rare findings such as RAAs. New macros resulted in an increase in AAA and SAA imaging follow-up (P < .001). Institution-specific dictation macros were found to improve adherence to reporting recommendations for incidental abdominal aneurysms, with further improvement seen after feedback, which can have a significant effect on clinical follow-up. © RSNA, 2023.

Comparative Performance of ChatGPT and Bard in a Text-Based Radiology Knowledge Assessment.

PURPOSE: Bard by Google, a direct competitor to ChatGPT, was recently released. Understanding the relative performance of these different chatbots can provide important insight into their strengths and weaknesses as well as which roles they are most suited to fill. In this project, we aimed to compare the most recent version of ChatGPT, ChatGPT-4, and Bard by Google, in their ability to accurately respond to radiology board examination practice questions. METHODS: Text-based questions were collected from the 2017-2021 American College of Radiology's Diagnostic Radiology In-Training (DXIT) examinations. ChatGPT-4 and Bard were queried, and their comparative accuracies, response lengths, and response times were documented. Subspecialty-specific performance was analyzed as well. RESULTS: 318 questions were included in our analysis. ChatGPT answered significantly more accurately than Bard (87.11% vs 70.44%, P < .0001). ChatGPT's response length was significantly shorter than Bard's (935.28 ± 440.88 characters vs 1437.52 ± 415.91 characters, P < .0001). ChatGPT's response time was significantly longer than Bard's (26.79 ± 3.27 seconds vs 7.55 ± 1.88 seconds, P < .0001). ChatGPT performed superiorly to Bard in neuroradiology, (100.00% vs 86.21%, P = .03), general & physics (85.39% vs 68.54%, P < .001), nuclear medicine (80.00% vs 56.67%, P < .01), pediatric radiology (93.75% vs 68.75%, P = .03), and ultrasound (100.00% vs 63.64%, P < .001). In the remaining subspecialties, there were no significant differences between ChatGPT and Bard's performance. CONCLUSION: ChatGPT displayed superior radiology knowledge compared to Bard. While both chatbots display reasonable radiology knowledge, they should be used with conscious knowledge of their limitations and fallibility. Both chatbots provided incorrect or illogical answer explanations and did not always address the educational content of the question.

Chemotherapy-Induced Toxicities: An Imaging Primer[Formula: see text].

The Coronavirus Disease of 2019 (COVID-19) pandemic has caused significant delays in the delivery of cancer treatments in Canada. As cancer treatment and imaging volumes return to normal, radiologists will encounter more cases of chemotherapy-induced toxicities. These toxicities have varied appearances on imaging, and can affect multiple organ systems. The purpose of this review is to offer a unified resource for general radiologists regarding the imaging appearances of chemotherapy-induced toxicities.

The feasibility of an interactive voice response system (IVRS) for monitoring patient safety after discharge from the ED.

BACKGROUND: Return ED visits are frequent and may be due to adverse events: adverse outcomes related to healthcare received. An interactive voice response system (IVRS) is a technology that translates human telephone input into digital data. Use of IVRS has been explored in many healthcare settings but to a limited extent in the ED. We determined the feasibility of using an IVRS to assess for adverse events after ED discharge. METHODS: This before and after study assessed detection of adverse events among consecutive high-acuity patients discharged from a tertiary care ED pre-IVRS and post-IVRS over two 2-week periods. The IVRS asked if the patient was having a health problem and if they wanted to speak to a nurse. Patients responding yes received a telephone interview. We searched health records for deaths, admissions to hospital and return ED visits. Three trained emergency physicians independently determined adverse event occurrence. We analysed the data using descriptive statistics. RESULTS: Of 968 patients studied, patients' age, sex, acuity and presenting complaint were comparable pre-IVRS and post-IVRS. Postimplementation, 393 (81.7%) of 481 patients had successful IVRS contact. Of these, 89 (22.6%) wanted to speak to a nurse. A total of 37 adverse events were detected over the two periods: 10 patients with 10 (6.5%) adverse events pre-IVRS and 16 patients with 27 (16.9%) adverse events post-IVRS. In the postimplementation period, the adverse events of seven patients were detected by the IVRS and five patients spontaneously requested assistance navigating post-ED care. CONCLUSIONS: This was a successful proof-of-concept study for applying IVRS technology to assess patient safety issues for discharged high-acuity ED patients.

Medical school radiology lectures: what are determinants of lecture satisfaction?

OBJECTIVE: Slideshow presentations are a popular teaching method in undergraduate medical education; however, there are scant data on determinants of lecture satisfaction. The purpose of this study is to determine which features of undergraduate medical school radiology lectures are associated with better student evaluations. MATERIALS AND METHODS: All undergraduate medical school radiology presentations and student evaluations at the University of Ottawa from January to December 2013 were compiled. A standardized data extraction sheet was applied by two independent reviewers, including a 10% overlap for audit. Student evaluations were reported on a 5-point Likert scale from which an overall mean score per lecture was calculated. Correlation coefficients were calculated for continuous variables in relation to mean evaluation score. Student t tests and univariate ANOVAs were performed for categoric data. Quantitative content analysis of student comments was also undertaken. RESULTS: Sixty-four slideshows by 33 lecturers were analyzed. The overall mean (SD) evaluation score was 4.38 ± 0.30. The strongest positive correlation with mean evaluation score was for type size (r = 0.32; p = 0.01), whereas the strongest negative association was for number of clinical cases presented (r = -0.32; p = 0.01). No association with percentage of text slides (r = 0.19; p = 0.14) or mean number of images on an image slide (r = -0.22; p = 0.08) was identified. Content analysis revealed a moderate positive correlation between percentage of total slides containing text only and the percentage of positive comments (r = 0.31) and a weak correlation between the mean number of images per image slide and the percentage of negative comments (r = 0.24). CONCLUSION: Larger type size and a higher proportion of text slides were more favored by medical students.

Using a book chat to improve attitudes and perceptions of long-term care staff about dementia.

This study sought to evaluate a book chat intervention based on Lisa Genova's novel, Still Alice, to influence long-term care (LTC) staff perceptions and attitudes when caring for individuals with dementia. A qualitative descriptive design was used. Eleven participants partook in a 2.5-hour book chat at a southern Ontario LTC facility. Following the book chat, participants answered two open-ended questions to assess how the book chat influenced their views on dementia. Thematic content analysis was used to analyze the qualitative questionnaire. Content analysis of the participants' responses revealed that the book chat positively influenced their attitudes and perceptions toward dementia, particularly by providing more insight into the individual's personal struggle with the disease. Furthermore, participants found that the book chat influenced their care practices. By creating innovative learning opportunities, attitudes and perceptions about dementia care can be transcended and greatly benefit staff, family, and residents.

Revamping Undergraduate Radiology Education in a Problem-Based Learning Driven Institution: Successes, Challenges, and Lessons Learned.

Radiology teaching for undergraduate medical students is often limited in amount and lacking a longitudinal approach spanning multiple years. At our institution, we had a goal to develop a formal and sustainable radiology curriculum. There were sequential changes made to the curriculum over several years and here, the successes, challenges, and lessons learned are outlined. Alongside these elements, student radiology quiz results and survey data over several years are also summarized. Curricular development included integrated anatomy and radiology sessions, revamped lectures focusing on interactivity and eliminating unnecessary redundancies across the multi-year curriculum, radiology media files to supplement problem-based learning, clinical skills videos, and a pre-clerkship radiology bootcamp elective.

Lung, Pleural, and Mediastinal Biopsies: From Preprocedural Assessment to Technique and Management of Complications.

Biopsies of the lung, pleura, and mediastinum play a crucial role in the workup of thoracic lesions. Percutaneous image-guided biopsy of thoracic lesions is a relatively safe and noninvasive way to obtain a pathologic diagnosis which is required to direct patient management. This article reviews how to safely perform image-guided biopsies of the lung, pleura, and mediastinum, from the preprocedural assessment to reviewing intraprocedural techniques, and how to avoid and manage complications.

The Impact of the COVID-19 Pandemic on Radiology Resident Education: Where Do We Go From Here?

The Coronavirus Disease of 2019 (COVID-19) pandemic caused a dramatic shift in radiology resident education. Primarily, physical distancing prompted a general transition to virtual learning. Common changes made by radiology residency programs included virtual rounds and readouts, the use of simulation technology, and case-based learning which utilized pedagogical approaches such as the flipped classroom for teaching residents. Virtual learning appears to be a suitable alternative to traditional, in-person learning, and may have a place post-pandemic as part of a blended curriculum with in-person and virtual components. The extent of disruption to radiology resident education varied based on the local impact of COVID-19 and the prevalence of redeployment, as did residents' mental health and wellbeing. Accessibility of mental health resources for residents was highlighted as an issue that programs need to address during these difficult times. Moreover, the pandemic resulted in unavoidable reductions in procedural exposure which programs mitigated through the use of simulation technologies and virtual learning resources. Professional development activities such as mentorship and career planning were also dramatically impacted by the pandemic and remains a challenge that programs need to consider moving forward post-pandemic. The purpose of this review is to outline the changes made to radiology resident education as a result of the COVID-19 pandemic and suggest what changes may be worthwhile to continue.

Resident Experiences With Virtual Radiology Learning During the COVID-19 Pandemic.

RATIONALE AND OBJECTIVES: COVID-19 has disrupted radiology education and forced a transition from traditional in-person learning to a virtual platform. As a result of hospital and state mandates, our radiology residency program quickly transitioned to a virtual learning platform to continue dissemination of knowledge, maintain resident engagement, and ensure professional development. The goal of this study is to assess the strengths and weaknesses of the virtual learning platform at our institution using resident ratings. MATERIALS AND METHODS: This institutional IRB-exempt study involved a survey of 17 questions which was electronically distributed to 45 radiology residents using SurveyMonkey. Questions encompassed resident satisfaction with teaching and professional development, scheduling changes, and engagement with the virtual platform. Answers to most questions were submitted on a Likert scale. RESULTS: A total of 31 of 45 respondents completed the survey (response rate = 69%). Most residents were satisfied with the virtual platform with teaching activities identified as a strength and the incorporation of professional development as a weakness. The most frequent barriers to attending the virtual curriculum were technical difficulties (43%) and childcare (36%). Residents who reported experiencing barriers were less likely to adhere to the virtual curriculum (p = 0.004). Most respondents (81%) reported a desire to maintain elements of the virtual learning practice postpandemic. CONCLUSION: The majority of residents reported high satisfaction with virtual learning during the COVID-19 pandemic. Teaching activities are a curricular strength. Weaknesses identified include the incorporation of professional development and extrinsic barriers, such as technical difficulties and family obligations, which require further support for trainees.

Differentiation of pancreatic neuroendocrine tumors from pancreas renal cell carcinoma metastases on CT using qualitative and quantitative features.

PURPOSE: To assess qualitative and quantitative imaging features on enhanced CT that may differentiate pancreatic neuroendocrine tumors (PNETs) from pancreatic renal cell carcinoma (RCC) metastases. METHODS: This IRB-approved multi-center retrospective case-control study compared 43 resected PNETs and 28 resected RCC metastases with pre-operative enhanced CT identified consecutively between 2003 and 2017. Two blinded radiologists (R1/R2) independently assessed tumor location, attenuation (relative to pancreas), composition (solid/cystic/mixed), homogeneity (homogeneous/heterogeneous), calcification, multiplicity, and for main pancreatic duct (MPD) dilation. Tumors were segmented for quantitative texture analysis. Data were analyzed with Chi square, logistic regression, and receiver operating characteristic (ROC). Inter-observer agreement was assessed (Cohen's kappa). RESULTS: There was no difference in age, gender, location, attenuation, or composition (P > 0.05) between groups. PNETs were larger than RCC metastases (37 ± 23 mm vs. 26 ± 21 mm, P = 0.038), more frequently solitary (P < 0.001), subjectively more heterogeneous (P = 0.033/0.144, R1/R2), and associated with calcification (P = 0.002/0.004) and MPD dilation (P = 0.025/0.006). Agreement for subjective features was moderate-to-almost perfect (K = 0.4879-0.9481). Quantitative texture analysis showed higher entropy in PNETs (6.32 ± 0.49 versus 5.96 ± 0.53; P = 0.004) with no difference in other features studied (P > 0.05). Entropy had ROC area under the curve for diagnosis of PNET of 0.77 ± 0.06, with optimal sensitivity/specificity of 71.4/79.1%. CONCLUSIONS: Compared to pancreatic RCC metastases, PNETs are larger, more frequently solitary, contain calcification, show MPD dilation, and are subjectively and quantitatively more heterogeneous tumors.