Diagnosis AI

Rachel Free and Jane Hollywood explore the patenting implications of AI applied to diagnostics

Rachel Free and Jane Hollywood, CMS Cameron McKenna Nabarro Olswang.

Many diagnosticapproaches are based on “deep learning”, which is inspired by the human brain and uses algorithms to train neural networks to recognise patterns. In general, a neural network is trained using historical patient data for which patient outcomes are known. Once the neural network has been trained it is able to generalise its knowledge so that when presented with a new example of patient data it has not seen before, it is able to compute a prediction of the patient outcome.