[Factors affecting in-hospital mortality in patients with sepsis: Development of a risk-adjusted model based on administrative data from German hospitals]

A Comparison of ChatGPT and Multidisciplinary Team Meeting Treatment Recommendations in 10 Consecutive Cervical Cancer Patients

Background The preparation of multidisciplinary team (MDT) meetings can be time-consuming. In addition to the clinical data being available digitally in subsystems, the preparation of more complex cases requires literature research. Several expert systems have been developed to support this process. However, the interaction with these systems has to be trained. Current development enables linguistic interaction with such artificial intelligence (AI) systems. To the best of our knowledge, these have not been tested as premedical screening tools for MDT. Methods This is a retrospective consecutive case series of 10 cervical cancer cases comparing the medical recommendations of the MDT and artificial intelligence (AI) on a low level (i.e., surgery, systemic treatment, and radiotherapy). Results The clinical cases ranged from primary diagnosis via suspected recurrence to palliative settings. The AI repeatedly stated that medical professionals need to be consulted before treatment decisions. The AI answers ranged from no agreement to overachievement by mentioning treatment options for preexisting risk factors (such as obesity). In standard cases, the AI answer matched well with the expert recommendations. In some cases, the AI answers were contrary to our treatment recommendation. Conclusion The interaction with current language AIs is temptingly easy, and the replies are very understandable. Despite the AI warning regarding medical recommendations in the majority of our cases, there was a good match with the MDT recommendations. However, in some cases, the medical evidence behind the answers was missing or in the worst case fictional. In our case series, the AI did not meet the requirements to support a clinical MDT meeting by prescreening the therapeutic options. However, it did exceed the expectations regarding the risk factors of the patients.

Predicting Sepsis Mortality in a Population-Based National Database: Machine Learning Approach

BACKGROUND

Although machine learning (ML) algorithms have been applied to point-of-care sepsis prognostication, ML has not been used to predict sepsis mortality in an administrative database. Therefore, we examined the performance of common ML algorithms in predicting sepsis mortality in adult patients with sepsis and compared it with that of the conventional context knowledge-based logistic regression approach.

OBJECTIVE

The aim of this study is to examine the performance of common ML algorithms in predicting sepsis mortality in adult patients with sepsis and compare it with that of the conventional context knowledge-based logistic regression approach.

METHODS

We examined inpatient admissions for sepsis in the US National Inpatient Sample using hospitalizations in 2010-2013 as the training data set. We developed four ML models to predict in-hospital mortality: logistic regression with least absolute shrinkage and selection operator regularization, random forest, gradient-boosted decision tree, and deep neural network. To estimate their performance, we compared our models with the Super Learner model. Using hospitalizations in 2014 as the testing data set, we examined the models' area under the receiver operating characteristic curve (AUC), confusion matrix results, and net reclassification improvement.

RESULTS

Hospitalizations of 923,759 adults were included in the analysis. Compared with the reference logistic regression (AUC: 0.786, 95% CI 0.783-0.788), all ML models showed superior discriminative ability (P<.001), including logistic regression with least absolute shrinkage and selection operator regularization (AUC: 0.878, 95% CI 0.876-0.879), random forest (AUC: 0.878, 95% CI 0.877-0.880), xgboost (AUC: 0.888, 95% CI 0.886-0.889), and neural network (AUC: 0.893, 95% CI 0.891-0.895). All 4 ML models showed higher sensitivity, specificity, positive predictive value, and negative predictive value compared with the reference logistic regression model (P<.001). We obtained similar results from the Super Learner model (AUC: 0.883, 95% CI 0.881-0.885).

CONCLUSIONS

ML approaches can improve sensitivity, specificity, positive predictive value, negative predictive value, discrimination, and calibration in predicting in-hospital mortality in patients hospitalized with sepsis in the United States. These models need further validation and could be applied to develop more accurate models to compare risk-standardized mortality rates across hospitals and geographic regions, paving the way for research and policy initiatives studying disparities in sepsis care.