Trustworthy Machine Learning Predictions to Support Clinical Research and Decisions

Nowadays, physicians have at their hands a huge amount of data produced by a large set of diagnostic and instrumental tests integrated with data obtained by high-throughput technologies. If such data were opportunely linked and analysed, they might be used to strengthen predictions, so that to improve the prevention and the time-to-diagnosis, reduce the costs of the health system, and bring out hidden knowledge. Machine learning is the principal technique used nowadays to leverage data and gain useful information. However, it has led to various challenges, such as improving the interpretability and explainability of the employed predictive models and integrating expert knowledge into the final system. Solving those challenges is of paramount importance to enhance the trust of both clinicians and patients in the system predictions. To solve the aforementioned issues, in this paper we propose a software workflow able to cope with the trustworthiness aspects of machine learning models and considering a multitude of heterogeneous data and models.

I. Introduction

21st-century healthcare systems face a hard challenge due to the ageing population. An example comes from cardiovascular disease (CVD), which counts 17.9 million deaths per year. According to recent predictions, the situation is worsening: the prevalence of the metabolic disease will rise, and in 2030 1 out of 2 U.S. adults will suffer from obesity [25]. With the ageing population [8], high-risk cardiovascular phenotypes will prevail, negatively impacting cardiovascular mortality and morbidity. These considerations can be extended to several chronic diseases (e.g. cancer, neurological and autoimmune disease) that, if not prevented, are deemed to encumber the healthcare systems. To face this challenge means adequately utilising medical resources and data to provide accurate, feasible, and easily implementable disease risk prediction models. Machine Learning (ML) can be used to extrapolate hidden information and improve disease prediction, such as CVD [12], [19], [22].