RESUMEN
Objectives: Determine if a large language model (LLM, GPT-4) can label and consolidate and analyze interventional radiology (IR) microwave ablation device safety event data into meaningful summaries similar to humans. Methods: Microwave ablation safety data from January 1, 2011 to October 31, 2023 were collected and type of failure was categorized by human readers. Using GPT-4 and iterative prompt development, the data were classified. Iterative summarization of the reports was performed using GPT-4 to generate a final summary of the large text corpus. Results: Training (n = 25), validation (n = 639), and test (n = 79) data were split to reflect real-world deployment of an LLM for this task. GPT-4 demonstrated high accuracy in the multiclass classification problem of microwave ablation device data (accuracy [95% CI]: training data 96.0% [79.7, 99.9], validation 86.4% [83.5, 89.0], test 87.3% [78.0, 93.8]). The text content was distilled through GPT-4 and iterative summarization prompts. A final summary was created which reflected the clinically relevant insights from the microwave ablation data relative to human interpretation but had inaccurate event class counts. Conclusion: The LLM emulated the human analysis, suggesting feasibility of using LLMs to process large volumes of IR safety data as a tool for clinicians. It accurately labelled microwave ablation device event data by type of malfunction through few-shot learning. Content distillation was used to analyze a large text corpus (>650 reports) and generate an insightful summary which was like the human interpretation.
RESUMEN
BACKGROUND AND AIMS: Autoimmune hepatitis (AIH) in children presenting in acute liver failure (ALF) can be fatal and often requires liver transplantation (LTx). This individual patient data meta-analysis (IPD) aims to examine management and outcomes of this population, given the lack of large cohort studies on paediatric AIH first presenting as ALF (AIH-ALF). METHODS: A systematic review was conducted in accordance with the Preferred Reporting Items for Systematic Review and Meta-Analyses of IPD statement using PubMed and Excerpta Medica dataBASE, and included English studies published between 2000 and 2020. The study included patients under 21 years of age, diagnosed with type 1 or 2 AIH and presenting with ALF. Data extracted included clinical and biochemical characteristics, interventions, and outcomes. RESULTS: Three hundred and thirty eligible patients from 61 studies were identified, with an additional five patients from our institution. The majority were female (66.8%), with a median age of 10. Overall, 59.7% achieved native liver survival (NLS), 35% underwent LTx, and 5% died before LTx. The use of corticosteroids with non-steroid immunomodulators increased the likelihood of NLS by 2.5-fold compared to corticosteroids alone. AIH-1 was associated with 3.3-fold odds for NLS, compared to AIH-2. However, on multivariate analysis, only AIH-1 was identified as an independent predictor for NLS (OR 3.8 [95% CI 1.03-14.2], p = .04). CONCLUSION: While corticosteroids and non-steroid immunomodulators treatment may offer enhanced probability of achieving NLS, treatment regimens for AIH-ALF may need to consider patient-specific factors, especially AIH type. This highlights the potential for NLS in AIH-ALF and suggest a need to identify biomarkers which predict the need for combination immunosuppression to avoid LTx.
Asunto(s)
Hepatitis Autoinmune , Fallo Hepático Agudo , Trasplante de Hígado , Humanos , Hepatitis Autoinmune/tratamiento farmacológico , Hepatitis Autoinmune/complicaciones , Hepatitis Autoinmune/diagnóstico , Fallo Hepático Agudo/mortalidad , Fallo Hepático Agudo/terapia , Niño , Preescolar , Inmunosupresores/uso terapéutico , Masculino , Femenino , Adolescente , Corticoesteroides/uso terapéuticoRESUMEN
Artificial intelligence (AI) is rapidly evolving and has transformative potential for interventional radiology (IR) clinical practice. However, formal training in AI may be limited for many clinicians and therefore presents a challenge for initial implementation and trust in AI. An understanding of the foundational concepts in AI may help familiarize the interventional radiologist with the field of AI, thus facilitating understanding and participation in the development and deployment of AI. A pragmatic classification system of AI based on the complexity of the model may guide clinicians in the assessment of AI. Finally, the current state of AI in IR and the patterns of implementation are explored (pre-procedural, intra-procedural, and post-procedural).
Asunto(s)
Inteligencia Artificial , Radiología Intervencionista , Humanos , Radiología Intervencionista/educación , Radiología Intervencionista/métodosRESUMEN
The introduction of artificial intelligence (AI) in interventional radiology (IR) will bring about new challenges and opportunities for patients and clinicians. AI may comprise software as a medical device or AI-integrated hardware and will require a rigorous evaluation that should be guided based on the level of risk of the implementation. A hierarchy of risk of harm and possible harms are described herein. A checklist to guide deployment of an AI in a clinical IR environment is provided. As AI continues to evolve, regulation and evaluation of the AI medical devices will need to continue to evolve to keep pace and ensure patient safety.
Asunto(s)
Inteligencia Artificial , Radiología Intervencionista , Humanos , Radiología Intervencionista/métodos , Seguridad del Paciente , Radiografía Intervencional/métodos , Lista de VerificaciónRESUMEN
Chronic thromboembolic pulmonary hypertension (CTEPH) is a challenging diagnosis that can occur even in the absence of a prior thrombotic event. The main screening test is ventilation-perfusion (VQ) scintigraphy. The gold standard treatment for CTEPH is pulmonary endarterectomy (PEA), however, balloon pulmonary angioplasty (BPA) is an emerging treatment, especially for CTEPH at the segmental level. We report on a case of a patient with segmental CTEPH diagnosed by lung subtraction iodine mapping (LSIM) in the context of a chest wall vascular malformation. CTEPH was treated with BPA and by embolization and ligation of their vascular malformation.
RESUMEN
Female mice display greater adipose angiogenesis and maintain healthier adipose tissue than do males upon high-fat diet feeding. Through transcriptome analysis of endothelial cells (EC) from the white adipose tissue of male and female mice high-fat-fed for 7 weeks, we found that adipose EC exhibited pronouncedly sex-distinct transcriptomes. Genes upregulated in female adipose EC were associated with proliferation, oxidative phosphorylation, and chromatin remodeling contrasting the dominant enrichment for genes related to inflammation and a senescence-associated secretory of male EC. Similar sex-biased phenotypes of adipose EC were detectable in a dataset of aged EC. The highly proliferative phenotype of female EC was observed also in culture conditions. In turn, male EC displayed greater inflammatory potential than female EC in culture, based on basal and tumor necrosis factor alpha-stimulated patterns of gene expression. Our study provides insights into molecular programs that distinguish male and female EC responses to pathophysiological conditions.