Clinical Applications of Machine Learning Algorithms: Beyond the Black Box

High Hopes for “Deep Medicine”? AI, Economics, and the Future of Care

“…If doctors start to rely on advice from AI, the question will arise whether we should—indeed, how we could—trust our doctors. As Watson and colleagues note, “If doctors do not understand why the algorithm made a diagnosis, then why should patients trust the recommended course of treatment?” 21 If we don't believe that it is our physicians who are really making the decisions about our health care, then it's hard to see how we could feel that they are caring for us. They might care about us, but that's not the same as caring for us.…”

Precision medicine in psoriatic arthritis: how should we select targeted therapies?

“…With such diverse clinical phenotype and complex genetic and immunological assessments it is likely that newer statistical methodologies such as machine learning and artificial intelligence may be useful to combine potential predictors of response 64 . Algorithms such as deep neural networks can synthesise data from millions of inputs using combinations far more complex than humans could compute.…”

Does artificial intelligence (AI) constitute an opportunity or a threat to the future of medicine as we know it?

“…For example, an algorithm designed to predict probability of death among hospital patients with pneumonia, spuriously classified asthmatics as low risk. 7 This was because in the derivation cohort setting, all asthmatic pneumonia care was optimised in an intensive care setting, thus making asthmatics appear to have an above average survival. Therefore, a machine algorithm's validity and causal inferences must be proved through proper prospective trials and randomisation, before acceptance into widespread clinical practice.…”