Transcending Language Barriers: Can ChatGPT Be the Key to Enhancing Multilingual Accessibility in Healthcare?

OBJECTIVE: To explore the capabilities of ChatGPT for the purpose of simplifying and translating radiology reports into Spanish, Hindi, and Russian languages, with comparisons to its performance in simplifying to the English language. METHODS: 50 deidentified abdomen-pelvis CT reports were fed to ChatGPT (4.0), instructing it to simplify and translate the report. The processed reports were rated on factual correctness (I), potential harmful errors (II), completeness (III), and explanation of medical terms (IV). The translated versions were also rated on the quality of translation (V). The scores in each category were compared between the translated versions and each translated version was compared with the English version in the first four categories. The original reports and the simplified English reports were rated on the Flesch Reading Ease Score (FRES) and the Flesch Kincaid Grade Level (FKRL). RESULTS: The Spanish translation outperformed the Hindi and Russian version significantly in categories I and III (p<0.05). All translated versions performed significantly worse compared to the English version in category IV (p<0.001). Notably, the Hindi translated version performed significantly worse in all 4 categories (p<0.05). The Russian translated version was also significantly worse in category III (p<0.05). In the first three categories, the Spanish translation, and the Russian translation in the first two categories demonstrated no statistically significant difference from the English version. No statistically significant difference was observed in the FRES and FKRL of the simplified English reports. Typographical errors in the original reports negatively affected the translation. CONCLUSION: ChatGPT demonstrates potential ability in translating reports and communicating pertinent clinical information with limited errors. More training and tailoring are required for languages that are not as commonly used in medical literature. LLMs can be used for translating and simplifying radiology reports, potentially improving access to healthcare, and helping reduce healthcare costs.

Gallstones Detection on Dual-Energy Computerized Tomography-Is It Ready for Real-World Use? A Retrospective Observational Study.

AIMS: The aims of the study are to evaluate the performance of dual-energy computed tomography (DECT) imaging in the detection of noncalcified gallstones (GSs) and to assess its performance relative to transabdominal ultrasound (US) in identifying cholelithiasis. METHOD: This study is a retrospective review of radiology records and images to find all patients who had both US and DECT scans within a 6-month period and were found to have GSs. Patients who did not have GSs on US served as the control group. The CT scans were reviewed by 4 radiologists who did not have access to the US results when assessing the presence or absence of GSs on the DECT scans. In case of any discrepancies among the radiologists, the majority opinion was considered. If there was a split opinion, a fifth reviewer was consulted. The data were analyzed to calculate sensitivity, specificity, positive and negative predictive values, as well as overall accuracy and to evaluate interreader variability. The absolute Hounsfield unit (HU) differences of the GSs and bile were compared between polychromatic (PC), virtual noncontrast (VNC), and virtual monochromatic (VMC) images. RESULTS: Considering at least 3-reader agreement, sensitivity, specificity, positive predictive value, negative predictive value, and accuracy were found to be 92%, 96%, 96%, 92%, and 94%, respectively. Individual reader sensitivity varied between 87% and 92%. There was good interobserver agreement with a Fleiss' kappa of 0.76. Quantification of the whole data set showed that no significant difference was observed in the HU values for the stones between the PC images and the VNC images. A significant increase was observed on the 50-keV VMC images compared with the PC and VNC images. In the study group, 17% stones were visualized only on the VNC or/and 50-keV VMC images, and not on the PC images. On quantitative analysis of these cases, there was a significant increase of HU in the VNC images as compared with PC images and a significant decrease of HU in the 50-keV VMC images as compared with PC images. CONCLUSIONS: Low-keV images increase stone-bile contrast. Evaluation of cholelithiasis using VNC and 50-keV VMC images demonstrated a 14% increase in sensitivity relative to conventional CT.

A Simulation Screening Mammography Module Created for Instruction and Assessment: Radiology Residents vs National Benchmarks.

RATIONALE AND OBJECTIVES: To improve mammographic screening training and breast cancer detection, radiology residents participated in a simulation screening mammography module in which they interpreted an enriched set of screening mammograms with known outcomes. This pilot research study evaluates the effectiveness of the simulation module while tracking the progress, efficiency, and accuracy of radiology resident interpretations and also compares their performance against national benchmarks. MATERIALS AND METHODS: A simulation module was created with 266 digital screening mammograms enriched with high-risk breast lesions (seven cases) and breast malignancies (65 cases). Over a period of 27 months, 39 radiology residents participated in the simulation screening mammography module. Resident sensitivity and specificity were compared to Breast Cancer Surveillance Consortium (BCSC data through 2009) national benchmark and American College of Radiology (ACR) Breast Imaging Reporting and Data System (BI-RADS) acceptable screening mammography audit ranges. RESULTS: The sensitivity, the percentage of cancers with an abnormal initial interpretation (BI-RADS 0), among residents was 84.5%, similar to the BCSC benchmark sensitivity of 84.9% (sensitivity for tissue diagnosis of cancer within 1 year following the initial examination) and within the acceptable ACR BI-RADS medical audit range of ≥75%. The specificity, the percentage of noncancers that had a negative image interpretation (BI-RADS 1 or 2), among residents was 83.2% compared to 90.3% reported in the BCSC benchmark data, but lower than the suggested ACR BI-RADS range of 88%-95%. CONCLUSIONS: Using simulation modules for interpretation of screening mammograms is a promising method for training radiology residents to detect breast cancer and to help them achieve competence toward national benchmarks.

Neuroimaging of vascular dementia.

When atrophy is seen on imaging in adult patients, it does not necessarily represent Alzheimer disease. Many cases of dementia or cognitive decline could be caused by reversible or preventable diseases, such as vascular dementia. This article familiarizes the physician with various types of vascular lesions leading to dementia and cognitive decline and their imaging appearances. Neuroimaging plays an important role in identifying vascular lesions of the brain early, even before the clinical manifestation of the cognitive decline symptoms and, thus, can help to prevent or delay the symptoms related to the various vascular pathologic conditions.

Understanding the US medical school requirements and medical students' attitudes about radiology rotations.

RATIONALE AND OBJECTIVES: To learn what percentage of US medical schools require their students to complete rotations in radiology during the clinical years. A secondary goal was to survey students' opinions about radiology rotations. MATERIALS AND METHODS: Data were collected from 159 US medical schools from the Association of American Medical Colleges (AAMC) for allopathic medical schools, the American Association of Colleges of Osteopathic Medicine (AACOM) for osteopathic medical schools, and by e-mailing curriculum directors at US medical schools with a survey. The secondary goal was achieved by e-mailing curriculum directors for voluntary medical student participation with an institutional review board-approved online survey. RESULTS: Data from the 2009-2010 academic year from AAMC and AACOM showed that 25% of US medical schools required radiology as a clinical rotation. Our survey of curriculum directors corroborated the AAMC and AACOM data. Data from our medical student survey showed that 87% of students from institutions requiring radiology thought radiology should be required. From institutions not requiring radiology, 45% of students thought that radiology should be required as a standalone course. Of students not required to take radiology, 63% planned to take radiology as an elective. CONCLUSIONS: Students, regardless of requirements, think there is value in having radiology as a regular aspect of a medical school curriculum. Medical schools should consider ways of incorporating radiology into their clinical curriculum.