Table 3.

The arguments to use the Machine Learning/Big Data-AI approaches in research on multimorbidity [29,30,45].

Arguments
Managing data of different grades of diversity and complexity.
Allowing for hidden knowledge to be extracted from data.
The potential to represent real world phenomena.
Linking data of different types and of multiple data sources.
Clinical research tasks determine research methods, which is opposite to what is nowadays when clinical projects meet the criteria of the established research methods.
In predicting the behavior of the system, the method learns from data.
Making sense of all accumulated data (including data from routine medical practice).
Patterns identification or identification of temporal trends in patterns.
The crucial role of a domain expert (knowledge) in data analyzing and in interpreting the results.
Application in different areas of research on multimorbidity, including: Population management and prevention program planning. Health status prediction and prognosis. Drug safety surveillance in the context of polypharmacy and comorbidities. Health risk stratification and personalized treatments. Patient characteristics diversity. Clinical decision support. Prediction of outcomes based on multiple features. Quality of care / performance measurement. Complex problem-solving tasks.