Building artificial intelligence and machine learning models : a primer for emergency physicians.

There has been a rise in the number of studies relating to the role of artificial intelligence (AI) in healthcare. Its potential in Emergency Medicine (EM) has been explored in recent years with operational, predictive, diagnostic and prognostic emergency department (ED) implementations being developed. For EM researchers building models de novo, collaborative working with data scientists is invaluable throughout the process. Synergism and understanding between domain (EM) and data experts increases the likelihood of realising a successful real-world model. Our linked manuscript provided a conceptual framework (including a glossary of AI terms) to support clinicians in interpreting AI research. The aim of this paper is to supplement that framework by exploring the key issues for clinicians and researchers to consider in the process of developing an AI model.

Dipstick urinalysis for the emergency department evaluation of urinary tract infections in infants aged less than 2 years.

BACKGROUND: The clinical diagnosis of urinary tract infection (UTI) in infants under the age of 2 years is challenging because of the nonspecific symptoms and signs in this age group. Prompt diagnosis and treatment is critical, and although dipstick testing allows rapid testing, there is some doubt about its use in infants. We sought to show the use of the dipstick test in identifying or excluding UTI in infants under the age of 2 years presenting to the emergency department with a febrile illness. METHODS: We conducted a retrospective diagnostic cohort study for over a 12-month period in a UK Paediatric Emergency Department, including all febrile children who had a urine dipstick and a quantitative culture as part of their diagnostic workup. The gold standard was the quantitative culture. RESULTS: Three hundred and twenty-one samples were eligible for inclusion. The mean age of the children included was 9.3 months. Sixty-three percent were female children. A test positive for nitrite, leucocyte esterase and blood gave a specificity of 97.12% [95% confidence interval (CI): 94.17-98.60] and a positive likelihood ratio of 15.13 (95% CI: 6.99-32.76). A test negative for nitrite, LE, blood and protein had a sensitivity of 97.44% (95% CI: 91.12-99.29) and a negative likelihood ratio of 0.10 (95% CI: 0.02-0.39). CONCLUSION: In febrile infants who were below 2 years of age, dipstick urinalysis shows promising results in identifying or ruling out a UTI.

It could have happened to Van Gogh: a case of fatal purple foxglove poisoning and review of the literature.

Although death owing to the toxic effects of the therapeutic Digitalis lanata extract, digoxin has been reported, there are no reported cases of fatal Digitalis purpurea (digitoxin) plant intoxication in humans in the literature. We describe a case of ingestion of Digitalis purpurea in a 64-year-old man, which was fatal despite administration of Digibind. A review of the literature and aspects of management of plant digitalis poisoning are discussed.