A systematic evaluation of the performance of GPT-4 and PaLM2 to diagnose comorbidities in MIMIC-IV patients.

Background: Given the strikingly high diagnostic error rate in hospitals, and the recent development of Large Language Models (LLMs), we set out to measure the diagnostic sensitivity of two popular LLMs: GPT-4 and PaLM2. Small-scale studies to evaluate the diagnostic ability of LLMs have shown promising results, with GPT-4 demonstrating high accuracy in diagnosing test cases. However, larger evaluations on real electronic patient data are needed to provide more reliable estimates. Methods: To fill this gap in the literature, we used a deidentified Electronic Health Record (EHR) data set of about 300,000 patients admitted to the Beth Israel Deaconess Medical Center in Boston. This data set contained blood, imaging, microbiology and vital sign information as well as the patients' medical diagnostic codes. Based on the available EHR data, doctors curated a set of diagnoses for each patient, which we will refer to as ground truth diagnoses. We then designed carefully-written prompts to get patient diagnostic predictions from the LLMs and compared this to the ground truth diagnoses in a random sample of 1000 patients. Results: Based on the proportion of correctly predicted ground truth diagnoses, we estimated the diagnostic hit rate of GPT-4 to be 93.9%. PaLM2 achieved 84.7% on the same data set. On these 1000 randomly selected EHRs, GPT-4 correctly identified 1116 unique diagnoses. Conclusion: The results suggest that artificial intelligence (AI) has the potential when working alongside clinicians to reduce cognitive errors which lead to hundreds of thousands of misdiagnoses every year. However, human oversight of AI remains essential: LLMs cannot replace clinicians, especially when it comes to human understanding and empathy. Furthermore, a significant number of challenges in incorporating AI into health care exist, including ethical, liability and regulatory barriers.

Impact of Gastrojejunostomy Anastomosis Diameter on Weight Loss Following Laparoscopic Gastric Bypass: A Systematic Review.

Laparoscopic Roux-en-Y gastric bypass (RYGB) is crucial for significant weight reduction and treating obesity-related issues. However, the impact of gastrojejunostomy (GJ) anastomosis diameter on weight loss remains unclear. We investigate this influence on post-RYGB weight loss outcomes. A systematic search was conducted. Six studies met the inclusion criteria, showing varied GJ diameters and follow-up durations (1-5 years). Smaller GJ diameters generally correlated with greater short-to-medium-term weight loss, with a threshold beyond which complications like stenosis increased. Studies had moderate-to-low bias risk, emphasizing the need for precise GJ area quantification post-operation. This review highlights a negative association between smaller GJ diameters and post-RYGB weight loss, advocating for standardized measurement techniques. Future research should explore intra-operative and AI-driven methods for optimizing GJ diameter determination.

Changes in social care after major emergency general surgery procedures.

BACKGROUND: Emergency general surgery (EGS) is a major part of the provision of healthcare, and patients undergoing EGS are at elevated risk of morbidity and mortality. This study aimed to determine factors contributing to patients losing their independence and being discharged to residential and nursing homes having previously lived in their own residences. METHODS: Our local data uploaded to the National Emergency Laparotomy Audit (NELA) (2014-2022) were analyzed. This national database encompasses all major EGS cases undertaken in the United Kingdom. The variables considered were patient demographics, American Society of Anesthesiologists score, admission and discharge dates, presenting pathology, operation type, and discharge destination. Comparative analyses segmented patients based on postdischarge EGS destinations. Multivariable logistic regression identified factors linked to residential/nursing home placement after discharge. Significance was set at P < .05. RESULTS: Data from all patients in the NELA database (n = 1611) were analyzed. Approximately 1 in 10 patients older than 70 years never returned home. Patients requiring additional support were on average 8.6 years older (P = .008). At older than 80 years, the need for extra social support increased substantially with each increasing year in age, and those older than 85 years were more than twice as likely to require extra support than 80-year-olds (P < .001). Patients who died were 11.4 years older than those discharged without additional support (P < .001). CONCLUSION: A significant proportion of patients, particularly the elderly, do not return to their usual place of residence and require a higher level of care postemergency surgery. These important social factors need to be considered before operating given that they may have significant quality of life and economic implications.

Game theory to enhance stock management of Personal Protective Equipment (PPE) during the COVID-19 outbreak.

Since the outbreak of the COVID-19 pandemic, many healthcare facilities have suffered from shortages in medical resources, particularly in Personal Protective Equipment (PPE). In this paper, we propose a game-theoretic approach to schedule PPE orders among healthcare facilities. In this PPE game, each independent healthcare facility optimises its own storage utilisation in order to keep its PPE cost at a minimum. Such a model can reduce peak demand considerably when applied to a variable PPE consumption profile. Experiments conducted for NHS England regions using actual data confirm that the challenge of securing PPE supply during disasters such as COVID-19 can be eased if proper stock management procedures are adopted. These procedures can include early stockpiling, increasing storage capacities and implementing measures that can prolong the time period between successive infection waves, such as social distancing measures. Simulation results suggest that the provision of PPE dedicated storage space can be a viable solution to avoid straining PPE supply chains in case a second wave of COVID-19 infections occurs.

Closed suction drainage improves clinical outcome in patients undergoing endoscopic vein harvesting for coronary artery bypass grafting.

BACKGROUND: Endoscopic vein harvesting (EVH) is a widely accepted technique for coronary artery bypass grafting, with well-reported benefits. However, EVH is associated with severe hematoma formation at incision sites, resulting in postoperative pain. We hypothesized that the use of a leg wound drain at the incision site may reduce these comorbidities. METHODS: One hundred consecutive patients were prospectively randomly allocated into two groups of 50: group 1 with leg wound drains, and group 2 without drains. Group 1 patients underwent EVH followed by closure with a size 10 high vacuum leg wound drain (20 kPa), whereas group 2 underwent EVH followed by closure without a leg wound drain. Patients were assessed for postoperative pain, wound infection, and satisfaction using validated scoring systems immediately after surgery. RESULTS: Pain at rest (p < 0.001) and with movement (p < 0.001), incidence of hematoma (p < 0.001), and patient satisfaction (p < 0.001) were significantly improved in the drain group at days 1 to 7 and remained significant at week 6 after surgery. Interestingly, the use of antibiotics (6% versus 24%, p = 0.012) and the number of general practitioner visits (6% versus 26%, p = 0.012) were lower in the drain group compared with the no-drain group. However, there were no differences in the length of hospital stay between the two groups after surgery. CONCLUSIONS: Our findings indicate that the use of a high vacuum leg drain after EVH for long saphenous vein is of clear therapeutic benefit in the early postoperative period. We also report that this technique may reduce antibiotic administration and general practitioner visits after patient discharge.