A Multimodal Approach for the Risk Prediction of Intensive Care and Mortality in Patients with COVID-19

Performance of artificial intelligence in predicting the prognossis of severe COVID-19: a systematic review and meta-analysis

Background

COVID-19-induced pneumonia has become a persistent health concern, with severe cases posing a significant threat to patient lives. However, the potential of artificial intelligence (AI) in assisting physicians in predicting the prognosis of severe COVID-19 patients remains unclear.

Methods

To obtain relevant studies, two researchers conducted a comprehensive search of the PubMed, Web of Science, and Embase databases, including all studies published up to October 31, 2023, that utilized AI to predict mortality rates in severe COVID-19 patients. The PROBAST 2019 tool was employed to assess the potential bias in the included studies, and Stata 16 was used for meta-analysis, publication bias assessment, and sensitivity analysis.

Results

A total of 19 studies, comprising 26 models, were included in the analysis. Among them, the models that incorporated both clinical and radiological data demonstrated the highest performance. These models achieved an overall sensitivity of 0.81 (0.64-0.91), specificity of 0.77 (0.71-0.82), and an overall area under the curve (AUC) of 0.88 (0.85-0.90). Subgroup analysis revealed notable findings. Studies conducted in developed countries exhibited significantly higher predictive specificity for both radiological and combined models (p < 0.05). Additionally, investigations involving non-intensive care unit patients demonstrated significantly greater predictive specificity (p < 0.001).

Conclusion

The current evidence suggests that artificial intelligence prediction models show promising performance in predicting the prognosis of severe COVID-19 patients. However, due to variations in the suitability of different models for specific populations, it is not yet certain whether they can be fully applied in clinical practice. There is still room for improvement in their predictive capabilities, and future research and development efforts are needed.

Systematic review registration

https://www.crd.york.ac.uk/prospero/ with the Unique Identifier CRD42023431537.

Time-series analysis of multidimensional clinical-laboratory data by dynamic Bayesian networks reveals trajectories of COVID-19 outcomes

BACKGROUND AND OBJECTIVE

COVID-19 severity spans an entire clinical spectrum from asymptomatic to fatal. Most patients who require in-hospital care are admitted to non-intensive wards, but their clinical conditions can deteriorate suddenly and some eventually die. Clinical data from patients' case series have identified pre-hospital and in-hospital risk factors for adverse COVID-19 outcomes. However, most prior studies used static variables or dynamic changes of a few selected variables of interest. In this study, we aimed at integrating the analysis of time-varying multidimensional clinical-laboratory data to describe the pathways leading to COVID-19 outcomes among patients initially hospitalised in a non-intensive care setting.

METHODS

We collected the longitudinal retrospective data of 394 patients admitted to non-intensive care units at the University Hospital of Padova (Padova, Italy) due to COVID-19. We trained a dynamic Bayesian network (DBN) to encode the conditional probability relationships over time between death and all available demographics, pre-existing conditions, and clinical laboratory variables. We applied resampling, dynamic time warping, and prototyping to describe the typical trajectories of patients who died vs. those who survived.

RESULTS

The DBN revealed that the trajectory linking demographics and pre-existing clinical conditions to death passed directly through kidney dysfunction or, more indirectly, through cardiac damage. As expected, admittance to the intensive care unit was linked to markers of respiratory function. Notably, death was linked to elevation in procalcitonin and D-dimer levels. Death was associated with persistently high levels of procalcitonin from admission and throughout the hospital stay, likely reflecting bacterial superinfection. A sudden raise in D-dimer levels 3-6 days after admission was also associated with subsequent death, possibly reflecting a worsening thrombotic microangiopathy.

CONCLUSIONS

This innovative application of DBNs and prototyping to integrated data analysis enables visualising the patient's trajectories to COVID-19 outcomes and may instruct timely and appropriate clinical decisions.

An Imaging Overview of COVID-19 ARDS in ICU Patients and Its Complications: A Pictorial Review

A significant proportion of patients with COVID-19 pneumonia could develop acute respiratory distress syndrome (ARDS), thus requiring mechanical ventilation, and resulting in a high rate of intensive care unit (ICU) admission. Several complications can arise during an ICU stay, from both COVID-19 infection and the respiratory supporting system, including barotraumas (pneumothorax and pneumomediastinum), superimposed pneumonia, coagulation disorders (pulmonary embolism, venous thromboembolism, hemorrhages and acute ischemic stroke), abdominal involvement (acute mesenteric ischemia, pancreatitis and acute kidney injury) and sarcopenia. Imaging plays a pivotal role in the detection and monitoring of ICU complications and is expanding even to prognosis prediction. The present pictorial review describes the clinicopathological and radiological findings of COVID-19 ARDS in ICU patients and discusses the imaging features of complications related to invasive ventilation support, as well as those of COVID-19 itself in this particularly fragile population. Radiologists need to be familiar with COVID-19's possible extra-pulmonary complications and, through reliable and constant monitoring, guide therapeutic decisions. Moreover, as more research is pursued and the pathophysiology of COVID-19 is increasingly understood, the role of imaging must evolve accordingly, expanding from the diagnosis and subsequent management of patients to prognosis prediction.

Explainable Artificial Intelligence Approach for the Early Prediction of Ventilator Support and Mortality in COVID-19 Patients

Early prediction of mortality and risk of deterioration in COVID-19 patients can reduce mortality and increase the opportunity for better and more timely treatment. In the current study, the DL model and explainable artificial intelligence (EAI) were combined to identify the impact of certain attributes on the prediction of mortality and ventilatory support in COVID-19 patients. Nevertheless, the DL model does not suffer from the curse of dimensionality, but in order to identify significant attributes, the EAI feature importance method was used. The DL model produced significant results; however, it lacks interpretability. The study was performed using COVID-19-hospitalized patients in King Abdulaziz Medical City, Riyadh. The dataset contains the patients’ demographic information, laboratory investigations, and chest X-ray (CXR) findings. The dataset used suffers from an imbalance; therefore, balanced accuracy, sensitivity, specificity, Youden index, and AUC measures were used to investigate the effectiveness of the proposed model. Furthermore, the experiments were conducted using original and SMOTE (over and under sampled) datasets. The proposed model outperforms the baseline study, with a balanced accuracy of 0.98 and an AUC of 0.998 for predicting mortality using the full-feature set. Meanwhile, for predicting ventilator support a highest balanced accuracy of 0.979 and an AUC of 0.981 was achieved. The proposed explainable prediction model will assist doctors in the early prediction of COVID-19 patients that are at risk of mortality or ventilatory support and improve the management of hospital resources.