AI in the ED: Assessing the efficacy of GPT models vs. physicians in medical score calculation.

BACKGROUND AND AIMS: Artificial Intelligence (AI) models like GPT-3.5 and GPT-4 have shown promise across various domains but remain underexplored in healthcare. Emergency Departments (ED) rely on established scoring systems, such as NIHSS and HEART score, to guide clinical decision-making. This study aims to evaluate the proficiency of GPT-3.5 and GPT-4 against experienced ED physicians in calculating five commonly used medical scores. METHODS: This retrospective study analyzed data from 150 patients who visited the ED over one week. Both AI models and two human physicians were tasked with calculating scores for NIH Stroke Scale, Canadian Syncope Risk Score, Alvarado Score for Acute Appendicitis, Canadian CT Head Rule, and HEART Score. Cohen's Kappa statistic and AUC values were used to assess inter-rater agreement and predictive performance, respectively. RESULTS: The highest level of agreement was observed between the human physicians (Kappa = 0.681), while GPT-4 also showed moderate to substantial agreement with them (Kappa values of 0.473 and 0.576). GPT-3.5 had the lowest agreement with human scorers. These results highlight the superior predictive performance of human expertise over the currently available automated systems for this specific medical outcome. Human physicians achieved a higher ROC-AUC on 3 of the 5 scores, but none of the differences were statistically significant. CONCLUSIONS: While AI models demonstrated some level of concordance with human expertise, they fell short in emulating the complex clinical judgments that physicians make. The study suggests that current AI models may serve as supplementary tools but are not ready to replace human expertise in high-stakes settings like the ED. Further research is needed to explore the capabilities and limitations of AI in emergency medicine.

Alerting on mortality among patients discharged from the emergency department: a machine learning model.

OBJECTIVES: Physicians continuously make tough decisions when discharging patients. Alerting on poor outcomes may help in this decision. This study evaluates a machine learning model for predicting 30-day mortality in emergency department (ED) discharged patients. METHODS: We retrospectively analysed visits of adult patients discharged from a single ED (1/2014-12/2018). Data included demographics, evaluation and treatment in the ED, and discharge diagnosis. The data comprised of both structured and free-text fields. A gradient boosting model was trained to predict mortality within 30 days of release from the ED. The model was trained on data from the years 2014-2017 and validated on data from the year 2018. In order to reduce potential end-of-life bias, a subgroup analysis was performed for non-oncological patients. RESULTS: Overall, 363 635 ED visits of discharged patients were analysed. The 30-day mortality rate was 0.8%. A majority of the mortality cases (65.3%) had a known oncological disease. The model yielded an area under the curve (AUC) of 0.97 (95% CI 0.96 to 0.97) for predicting 30-day mortality. For a sensitivity of 84% (95% CI 0.81 to 0.86), this model had a false positive rate of 1:20. For patients without a known malignancy, the model yielded an AUC of 0.94 (95% CI 0.92 to 0.95). CONCLUSIONS: Although not frequent, patients may die following ED discharge. Machine learning-based tools may help ED physicians identify patients at risk. An optimised decision for hospitalisation or palliative management may improve patient care and system resource allocation.

Establishing a COVID-19 treatment centre in Israel at the initial stage of the outbreak: challenges, responses and lessons learned.

Anticipating the need for a COVID-19 treatment centre in Israel, a designated facility was established at Sheba Medical Center-a quaternary referral centre. The goals were diagnosis and treatment of patients with COVID-19 while protecting patients and staff from infection and ensuring operational continuity and treatment of patients with non-COVID. Options considered included adaptation of existing wards, building a tented facility and converting a non-medical structure. The option chosen was a non-medical structure converted to a hospitalisation facility suited for COVID-19 with appropriate logistic and organisational adaptations. Operational principles included patient isolation, unidirectional workflow from clean to contaminated zones and minimising direct contact between patients and caregivers using personal protection equipment (PPE) and a multimodal telemedicine system. The ED was modified to enable triage and treatment of patients with COVID-19 while maintaining a COVID-19-free environment in the main campus. This system enabled treatment of patients with COVID-19 while maintaining staff safety and conserving the operational continuity and the ability to continue delivery of treatment to patients with non-COVID-19.

The Pandemic's impacts on patients without Covid-19 on multidisciplinary aspects in emergency medicine care.

The emergence of Covid-19 has caused a pandemic and is a major public health concern. Covid-19 has fundamentally challenged the global health care system in all aspects. However, there is a growing concern for the subsequent detrimental effects of continuing delays or adjustments on time-dependent treatments for Covid-19 negative patients. Patients arriving to the ED with STEMIs and acute CVA are currently presumed to have delays due to Covid-19 related concerns. The objective of this paper is to evaluate the implications of the Covid-19 pandemic on non-Covid19 patients in emergency care settings. We conducted a retrospective study from February 2020 to April 2020 and compared this to a parallel period in 2019 to assess the impact of the Covid-19 pandemic on three distinct non-Covid-19 ED diagnosis that require immediate intervention. Our primary outcome measures were time to primary PCI in acute STEMI, time to fibrinolysis in acute CVA, and time to femoral hip fracture correction surgery. Our secondary outcome measure included a composite outcome of length of stay in hospital and mortality. From 1 February 2020 to 30 April 2020, the total referrals to ED diagnosed with STEMI, Hip fracture and CVA of which required intervention were 197 within Covid-19 group 2020 compared to 250 in the control group 2019. Mean duration to intervention (PCI, surgery and tPA, respectively) did not differ between COVID-19 group and 2019 group. Among femoral hip fracture patients', the referral numbers to ED were significantly lower in Covid-19 era (p = 0.040) and the hospitalization stay was significantly shorter (p = 0.003). Among CVA patients', we found statistical differences among the number of referrals and the patients' age. Coping with the Covid-19 pandemic presents a challenge for the general healthcare system. Our results suggest that with proper management, despite the obstacles of isolation policies and social distancing, any negative impact on the quality of health care for the non-Covid-19 patients can be minimized in the emergency department setting.