Samenvatting

Consider the clinical reasoning process of the physiotherapist: the physiotherapist analyses the information (data) available from different sources (the client’s narrative/perspective, context, medical history/examination), formulates hypotheses (rationalisation) on the basis of prior knowledge (natural algorithms), identifies the most likely underlying explanation (diagnosis/classification) on the basis of elimination (feature selection), and decides what course of treatment fits this specific case (prognosis, shared decision, action). This results in clinical experience and is adapted into skill/knowledge (learning). The above basic description of a natural clinical reasoning workflow shows many similarities with the workflow of computer algorithms or artificial intelligence (AI); however, when venturing deeper into the essence of this ‘seemingly identical twin’ there are differences. To aid in its responsible use, it is critical for clients and modern physiotherapists to develop a deeper understanding of AI. This Editorial aims to: provide an overview of the relevant developments in AI for healthcare providers, focusing specifically on physiotherapists; and provide an introductory practical guide to responsible use of AI in clinical practice, with an emphasis on clinical reasoning.