ChatGPT and Patient Information in Nuclear Medicine: GPT-3.5 Versus GPT-4.

The GPT-3.5-powered ChatGPT was released in late November 2022 powered by the generative pretrained transformer (GPT) version 3.5. It has emerged as a readily accessible source of patient information ahead of medical procedures. Although ChatGPT has purported benefits for supporting patient education and information, actual capability has not been evaluated. Moreover, the March 2023 emergence of paid subscription access to GPT-4 promises further enhanced capabilities requiring evaluation. Methods: ChatGPT was used to generate patient information sheets suitable for gaining informed consent for 7 common procedures in nuclear medicine. Responses were generated independently for both GPT-3.5 and GPT-4 architectures. Specific procedures were selected that had a long-standing history of use to avoid any bias associated with the September 2021 learning cutoff that constrains both GPT-3.5 and GPT-4 architectures. Each information sheet was independently evaluated by 3 expert assessors and ranked on the basis of accuracy, appropriateness, currency, and fitness for purpose. Results: ChatGPT powered by GPT-3.5 provided patient information that was appropriate in terms of being patient-facing but lacked accuracy and currency and omitted important information. GPT-3.5 produced patient information deemed not fit for the purpose. GPT-4 provided patient information enhanced across appropriateness, accuracy, and currency, despite some omission of information. GPT-4 produced patient information that was largely fit for the purpose. Conclusion: Although ChatGPT powered by GPT-3.5 is accessible and provides plausible patient information, inaccuracies and omissions present a risk to patients and informed consent. Conversely, GPT-4 is more accurate and fit for the purpose but, at the time of writing, was available only through a paid subscription.

Potential utility of B-type natriuretic peptides in secondary prevention following percutaneous coronary intervention in remote communities of Western Australia.

INTRODUCTION: A third of all acute coronary events that present in the Australian population occur in patients with established coronary heart disease. This study assessed the prognostic value of combined B-type natriuretic peptides (BNP) measurement and quantitative myocardial perfusion scan (MPS) data for cardiac events (CE). MATERIAL AND METHODS: This retrospective cohort study involved 133 patients from rural Western Australia. The cut-off for normality was 6.0 for qualitative summed difference scores (SDS) of MPS and 400 pg/mL for BNP. RESULTS: Patients with no CE had a mean SDS and BNP (1.52 with a 95% CI of 0.34 to 2.69), (175.9 with a 95% CI of 112.7-239.1) that was lower than patients with CE (6.54 with 95% CI 4.18-9.89) (P = 0.0003), (669.1 with 95% CI 543.9-794.3) (P < 0.0001). The sensitivity and specificity of combined testing for predicting CE respectively were 79.6% and 86.3% for SDS, 84.6% and 94.1% for BNP, and 100% and 92.7% for SDS and BNP combined. DISCUSSION AND CONCLUSION: Elevated BNP is marginally superior to MPS in predicting CEs in patients who have previously undergone percutaneous coronary intervention (PCI); however, MPS can identify the region of myocardium most at risk. Routine BNP monitoring in this subgroup may serve as secondary prevention by identifying subclinical disease.

Rural communities are disproportionately affected by preventable coronary heart disease-related deaths and access to cardiac imaging techniques can be infrequent or unavailable.Secondary prevention strategies can reduce hospital readmissions and contribute to improving the management of chronic conditions.This study demonstrated that elevated B-type natriuretic peptides levels were marginally superior to myocardial perfusion scans in predicting cardiac events in patients with prior percutaneous coronary intervention.Monitoring BNP levels in rural patients with prior percutaneous coronary interventions is a relatively non-invasive and inexpensive, and may lead to improved risk estimation, identify the subclinical disease and provoke further investigation as clinically appropriate.

Concurrent measurement of serum and radiomic biomarkers in the clinical investigation of equine musculoskeletal injuries: A prospective pilot study.

The prevention of musculoskeletal injuries and their related welfare and economic impacts represent an immediate priority for the horse racing industry. This prospective pilot study aimed to evaluate a method to quantitatively analyze scintigraphic features of specific anatomical regions of the horse's appendicular skeleton in combination with secondary measures of musculoskeletal metabolism in blood. Twelve horses referred for scintigraphic assessment of lameness were enrolled. Blood samples were collected immediately prior to the administration of radiotracer. Serum concentrations associated with bone turnover were determined for the following biomarkers: C-terminal telopeptides of type I collagen, proteoglycans and sulfated glycosaminoglycans, collagen type II, osteocalcin, and procollagen II C-terminal propeptide. Scintigraphic images underwent radiomic analysis of discrete regions of the distal limbs and these data were correlated to bone turnover markers. Three lame horses demonstrated asymmetrical radiomic abnormalities. The concentration of osteocalcin in the lame horses was significantly higher when compared to the control group, while no significant changes were observed for the other screened serum biomarkers. Findings from the current study provided evidence that radiomic analysis of equine scintigraphy is feasible. This method has the potential to interrogate which serum markers are associated with musculoskeletal injuries.

User acceptance of the Web-based distribution of radiology services in regional and remote centres of Western Australia.

A large-scale, Web browser-based radiological information system integrated with a picture archiving and communication system (RIS/PACS) has been installed in regional and remote radiology centres of Western Australia. This provides patients in rural areas with local access to advanced medical imaging. User attitudes to the changes were surveyed in a six-week period starting in March 2005. A total of 53 people completed the survey (62% response rate). Respondents were eight radiologists, 21 medical imaging technologists, 10 administration officers and 14 reception staff. All groups gave average ratings of the system's quality. All staff rated the information quality of the system as good. On-site information technology support was judged to be reliable, professional and empathic towards user concerns. There was a broad range of opinion between groups, but on average they were satisfied. The lack of a modern network, comparable to what is available in the metropolitan areas, has impeded the full potential of RIS/PACS in the smaller rural centres.

A preliminary assessment of Internet-based nuclear telecardiology to support the clinical management of cardiac disease in a remote community.

A portable nuclear medicine (NM) processing system was established in Kalgoorlie and an acute myocardial perfusion scintigraphy (MPS) service was provided for the local regional hospital. After scanning the patient, the data were processed on a laptop computer and JPEG images were transmitted to a secure Web server. A secure email message, with the URL link enclosed and a provisional indication of normal or abnormal findings, was sent to the referring clinician from the NM facility. Use of the Internet allowed for a group consultation between the NM technician, the referrer and the cardiologist in Perth. During a three-month study period, 42 patients were referred for exclusion of acute coronary syndrome. Of these, 21 (50%) demonstrated abnormal perfusion studies, two of which were classified as requiring urgent medical intervention. Seventeen studies were normal (41%) and four (10%) were designated equivocal. There was an alteration in the treatment plan for 32 patients (76%), including four for whom admission or further investigation was deemed unwarranted. The results suggest that MPS findings, distributed via the Internet, allow for earlier risk stratification and have a direct affect on clinical decision making.