Advocating for free software in healthcare: experience from bioinformatics for genetic diagnostics at the CHU, Liège

Healthcare systems are a valuable ecosystem to assess the impact of software business models. In this work, we focus on our experience with both commercial and open source software in the context of genetic diagnostics applications implemented under ISO 15189 in a public hospital setting. We hope to provide an interesting insight on how free software, by contributing back to the scientific community that develops diagnostics applications, offers a sustainable model for publicly funded innovation and for maintenance of state-of-the-art scientific software tools. We will begin by outlining our experience with commercial software, highlighting three key issues: (1) the significant power imbalance between suppliers and purchasers, heavily favoring suppliers due to the highly technical nature of scientific software, (2), the risk of lock-in due to the lack of industry support for recognized open scientific standards (BAM, VCF) and (3) how scientific developments largely developed in academia and funded by public grants are often absorbed into closed source software whose authors do not contribute back to the community. Furthermore, we will highlight our experience with open source software. We will present examples of how the fact that it allows code auditing, bug identification/fixing and implementation of new features, has positively impacted our work and has helped build stronger software communities. Moreover, by significantly reducing software acquisition costs and by developing in-house long-term expertise, open source software usage allows optimal allocation of hospital's limited budgets. Finally, scientific innovations, which support translational medicine, can be implemented faster in a diagnostics settings via open source software by circumventing market research steps typically central to commercial software development. We will also advocate for a software model where public scrutiny enhances development of more robust and secure code.