Exploring the diagnostic performance of machine learning in prediction of metabolic phenotypes focusing on thyroid function.

In this study, we employed various machine learning models to predict metabolic phenotypes, focusing on thyroid function, using a dataset from the National Health and Nutrition Examination Survey (NHANES) from 2007 to 2012. Our analysis utilized laboratory parameters relevant to thyroid function or metabolic dysregulation in addition to demographic features, aiming to uncover potential associations between thyroid function and metabolic phenotypes by various machine learning methods. Multinomial Logistic Regression performed best to identify the relationship between thyroid function and metabolic phenotypes, achieving an area under receiver operating characteristic curve (AUROC) of 0.818, followed closely by Neural Network (AUROC: 0.814). Following the above, the performance of Random Forest, Boosted Trees, and K Nearest Neighbors was inferior to the first two methods (AUROC 0.811, 0.811, and 0.786, respectively). In Random Forest, homeostatic model assessment for insulin resistance, serum uric acid, serum albumin, gamma glutamyl transferase, and triiodothyronine/thyroxine ratio were positioned in the upper ranks of variable importance. These results highlight the potential of machine learning in understanding complex relationships in health data. However, it's important to note that model performance may vary depending on data characteristics and specific requirements. Furthermore, we emphasize the significance of accounting for sampling weights in complex survey data analysis and the potential benefits of incorporating additional variables to enhance model accuracy and insights. Future research can explore advanced methodologies combining machine learning, sample weights, and expanded variable sets to further advance survey data analysis.

Gastric adenocarcinoma location and postoperative complication rates in Asian patients: A 2014-2019 NSQIP analysis.

BACKGROUND: Asian gastric cancer patients have higher long-term survival rates post-gastrectomy. This study compares 30-day post-gastrectomy outcomes between Asians and non-Asians. METHODS: Gastric cancer patients undergoing elective gastrectomies were identified in 2014-2019 NSQIP datasets (n â€‹= â€‹1,438). Demographics, comorbidities, and postoperative outcomes were analyzed. RESULTS: Asians had lower odds of total gastrectomy (AOR â€‹= â€‹0.52, p â€‹= â€‹0.003), age ≥65 (AOR â€‹= â€‹0.60, p â€‹= â€‹0.006), smoking history (AOR â€‹= â€‹0.35, p â€‹< â€‹0.001), dyspnea (AOR â€‹= â€‹0.25, p â€‹= â€‹0.01), and hypoalbuminemia (AOR â€‹= â€‹0.62, p â€‹= â€‹0.025); they also had lower BMI (p â€‹< â€‹0.001). Postoperative outcomes were not significantly different aside from a shorter median length of hospital stay in days (LOS) (Asians: 7 (6, 11); non-Asians: 8 (6, 11); p â€‹< â€‹0.001). CONCLUSIONS: Asian gastric cancer patients have significantly lower odds of having select preoperative comorbidities and have shorter hospital LOS.

Effect of the COVID-19 Pandemic on Medical Student Performance and Evaluation Scores.

In March 2020, the University of Hawaii John A. Burns School of Medicine suspended in person clinical teaching due to the SARS-CoV-2 (COVID) pandemic. During this period, virtual cases, telehealth participation, and online cases were incorporated into medical education. We have examined the effects of educational outcomes of third and fourth year students throughout clerkship performance, national standardized test scores, and our local fourth year OSCE examination. We found that USMLE step 2 scores were higher in the COVID-affected group. Patient logs in the COVID-affected group were lower for internal medicine, family medicine, OBGYN, and psychiatry clerkships. Clerkship performance grades in the COVID-affected group were lower for OBGYN and higher for surgery and psychiatry, but not different in other clerkships. USNBME subject specific examination scores in the COVID-affected group were higher for internal medicine, surgery, family medicine and psychiatry, but not different in all other specialties. For the fourth year OSCE, students in the COVID-affected group performed better on note taking and worse on physical examination. Future investigations will be needed to explore how our COVID-affected medical students perform in residency and beyond.

Subtotal vs total cholecystectomy for difficult gallbladders: A systematic review and meta-analysis.

INTRODUCTION: With severely inflamed gallbladders, laparoscopic cholecystectomy can be difficult and may require procedures like subtotal cholecystectomy (SC). Few studies exist comparing SC and total cholecystectomy (TC) in the setting of severe biliary inflammation. This meta-analysis aims to compare SC and TC for difficult gallbladders. METHODS: Medline-OVID, Embase-OVID, and Cinahl were searched including only studies comparing SC to TC for difficult gallbladders. Primary outcome was CBD injury. Secondary outcomes included bile leak, duodenal injury, retained stone, bleeding, intraabdominal collection, wound infection, reoperation, and mortality. RESULTS: Ten studies were included. Compared to TC, SC significantly lowered the risk for CBD injury (0 â€‹% vs. 1.6 â€‹%, RR 0.30, 95%CI 0.10-0.87) but increased risk of bile leaks (RR 3.5, 95%CI 1.79-6.84), postoperative ERCP (RR 2.86, 95%CI 1.53-5.35), intraabdominal collections (RR 2.55, 95%CI 1.32-4.93), and reoperation (RR 2.92, 95%CI 1.14-7.47). CONCLUSION: SC is a reasonable alternative to difficult gallbladders that may decrease the risk of CBD injuries. Knowing both approaches is crucial to manage the difficult gallbladder while minimizing harm. Further studies are needed to understand the value of SC for difficult cholecystectomy.