Building Process-Oriented Data Science Solutions for Real-World Healthcare.

The COVID-19 pandemic has highlighted some of the opportunities, problems and barriers facing the application of Artificial Intelligence to the medical domain. It is becoming increasingly important to determine how Artificial Intelligence will help healthcare providers understand and improve the daily practice of medicine. As a part of the Artificial Intelligence research field, the Process-Oriented Data Science community has been active in the analysis of this situation and in identifying current challenges and available solutions. We have identified a need to integrate the best efforts made by the community to ensure that promised improvements to care processes can be achieved in real healthcare. In this paper, we argue that it is necessary to provide appropriate tools to support medical experts and that frequent, interactive communication between medical experts and data miners is needed to co-create solutions. Process-Oriented Data Science, and specifically concrete techniques such as Process Mining, can offer an easy to manage set of tools for developing understandable and explainable Artificial Intelligence solutions. Process Mining offers tools, methods and a data driven approach that can involve medical experts in the process of co-discovering real-world evidence in an interactive way. It is time for Process-Oriented Data scientists to collaborate more closely with healthcare professionals to provide and build useful, understandable solutions that answer practical questions in daily practice. With a shared vision, we should be better prepared to meet the complex challenges that will shape the future of healthcare.

The clinical course of pneumomediastinum in patients with SARS-CoV-2 before invasive mechanical ventilation.

Pneumomediastinum and pneumothorax are recognised complications encountered in COVID-19 before or during invasive mechanical ventilation (IMV). The clinical course of patients developing pneumomediastinum before IMV is yet to be evaluated.Four-thousand, one-hundred and thirty-one patients hospitalised with COVID-19 over a 12-month period were retrospectively reviewed to evaluate for incidence, clinical characteristics and outcomes. A subgroup analysis was done to identify any clinical traits between survivors and non-survivors. The overall incidence of pneumomediastinum prior to IMV was 0.92% (n=38) and was seen at admission or during non-invasive respiratory support. Thirty-seven per cent had associated pneumothorax most commonly unilateral (right side). The median (interquartile range (IQR)) duration from admission to developing pneumomediastinum was 7 days (3-11) and complete resolution was seen in 53% of patients; median (IQR) duration to resolution was 8 days (4-17). The in-hospital mortality associated with pneumomediastinum in patients with SARS-CoV-2 (PneumoCoV) was 55%. Increasing age (68 ± 12 years vs 56 ± 14 years; p=0.01), higher body mass index (31 ± 5 kg/m2 vs 28 ± 5 kg/m2; p=0.04), lack of resolution of pneumomediastinum (67% vs 24%; p=0.01; odds ratio (OR) 6.5; 95% confidence interval (CI) 1.5-27.5), presence of concurrent pneumothorax (65% vs 14%; p=0.002; OR 11; 95% CI 2.2-53.1) and elevated procalcitonin levels (>0.5 ng/mL; 81% vs 41%; p=0.01; OR 6; 95% CI 1.4-26) were significant features in those who did not survive.The incidence of PneumoCoV, despite being low, is associated with increased mortality. It is a hallmark of moderate to severe disease with multifaceted contributory factors. Both demographic and clinical factors predict survival.