In our recent work on collaborations between healthcare providers and AI & machine-learning companies, we're seeing more realism about what the technology can realistically achieve at the moment in a healthcare setting.
That is very welcome. Rather than promising the total revolution of clinical decision-making, both parties are taking a prudent approach - choosing to work together to conduct research with a "wait-and-see" approach to what real-world use cases may arise.
Our work has seen us advise on data access and collection, the development of new IP when algorithms are trained on healthcare data, and potential value-capture. By thinking about these collaborations in a phased way, both the healthcare provider and the tech company are able to have open conversations about what is possible now, while developing a framework to generate and capture health outcome benefits in the future.
As this Nature article bears out, it is this type of approach that is more likely to see AI produce tangible improvements in clinical decision-making and, ultimately, patient outcomes.
We'll discuss this, and more, at our upcoming Life Sciences Summit on AI in health tech. Register your interest here.
There is great excitement that medical artificial intelligence (AI) based on machine learning (ML) can be used to improve decision making at the patient level in a variety of healthcare settings. However, the quantification and communication of uncertainty for individual predictions is often neglected even though uncertainty estimates could lead to more principled decision-making