When Performance is not Enough -- A Multidisciplinary View on Clinical Decision Support

TL;DR

This paper emphasizes that sustainable progress in healthcare requires considering implementation, usability, and ethical factors of machine learning systems, illustrated through a nephrology risk prediction pilot project.

Contribution

It provides a multidisciplinary perspective on ML in healthcare, highlighting practical challenges and lessons learned beyond performance improvements.

Findings

Implementation and usability are critical for practical adoption.

Ethical considerations are essential in medical decision support.

A nephrology risk prediction system demonstrates real-world challenges.

Abstract

Scientific publications about machine learning in healthcare are often about implementing novel methods and boosting the performance - at least from a computer science perspective. However, beyond such often short-lived improvements, much more needs to be taken into consideration if we want to arrive at a sustainable progress in healthcare. What does it take to actually implement such a system, make it usable for the domain expert, and possibly bring it into practical usage? Targeted at Computer Scientists, this work presents a multidisciplinary view on machine learning in medical decision support systems and covers information technology, medical, as well as ethical aspects. Along with an implemented risk prediction system in nephrology, challenges and lessons learned in a pilot project are presented.