Assessing the prognostic utility of clinical and radiomic features for COVID-19 patients admitted to ICU: challenges and lessons learned.

Severe cases of COVID-19 often necessitate escalation to the Intensive Care Unit (ICU), where patients may face grave outcomes, including mortality. Chest X-rays play a crucial role in the diagnostic process for evaluating COVID-19 patients. Our collaborative efforts with Michigan Medicine in monitoring patient outcomes within the ICU have motivated us to investigate the potential advantages of incorporating clinical information and chest X-ray images for predicting patient outcomes. We propose an analytical workflow to address challenges such as the absence of standardized approaches for image pre-processing and data utilization. We then propose an ensemble learning approach designed to maximize the information derived from multiple prediction algorithms. This entails optimizing the weights within the ensemble and considering the common variability present in individual risk scores. Our simulations demonstrate the superior performance of this weighted ensemble averaging approach across various scenarios. We apply this refined ensemble methodology to analyze post-ICU COVID-19 mortality, an occurrence observed in 21% of COVID-19 patients admitted to the ICU at Michigan Medicine. Our findings reveal substantial performance improvement when incorporating imaging data compared to models trained solely on clinical risk factors. Furthermore, the addition of radiomic features yields even larger enhancements, particularly among older and more medically compromised patients. These results may carry implications for enhancing patient outcomes in similar clinical contexts.

Use of machine learning to assess the prognostic utility of radiomic features for in-hospital COVID-19 mortality.

As portable chest X-rays are an efficient means of triaging emergent cases, their use has raised the question as to whether imaging carries additional prognostic utility for survival among patients with COVID-19. This study assessed the importance of known risk factors on in-hospital mortality and investigated the predictive utility of radiomic texture features using various machine learning approaches. We detected incremental improvements in survival prognostication utilizing texture features derived from emergent chest X-rays, particularly among older patients or those with a higher comorbidity burden. Important features included age, oxygen saturation, blood pressure, and certain comorbid conditions, as well as image features related to the intensity and variability of pixel distribution. Thus, widely available chest X-rays, in conjunction with clinical information, may be predictive of survival outcomes of patients with COVID-19, especially older, sicker patients, and can aid in disease management by providing additional information.

Regimen-related Mortality Risk in Patients Undergoing Peritoneal Dialysis Using Hypertonic Glucose Solution: A Retrospective Cohort Study.

OBJECTIVES: The main purpose of this study was to quantify the risk of mortality linked to various regimens of hypertonic peritoneal dialysis (PD) solution. METHODS: A retrospective cohort study of patients using home-based PD was carried out. The prescribed regimen of glucose-based PD solution for all patients, determined on the basis of their individual conditions, was extracted from their medical chart records. The primary outcome was death. The treatment regimens were categorized into 3 groups according to the type of PD solution used: original PD (1.5% glucose), shuffle PD (1.5 and 2.5% glucose), and serialized PD (2.5 and 4.5% glucose). Multivariate analysis (using the Weibull model) was applied to comprehensively examine survival probabilities related to the explanatory variable, while adjusting for other potential confounders. RESULTS: Of 300 consecutive patients, 38% died over a median follow-up time of 30 months (interquartile range: 15-46 months). Multivariate analysis showed that a treatment regimen with continued higher-strength PD solution (serialized PD) resulted in a lower survival rate than when the conventional strength solution was used (adjusted hazard ratio, 2.6; 95% confidence interval, 1.6 to 4.6, p<0.01). Five interrelated risk factors (age, length of time on PD, hemoglobin levels, albumin levels, and oliguria) were significant predictors contributing to the outcome. CONCLUSIONS: Frequent exposure to high levels of glucose PD solution significantly contributed to a 2-fold higher rate of death, especially when hypertonic glucose was prescribed continuously.