The use of quantitative sensory testing for personalized pain medicine in the age of AI

Quantitative sensory testing (QST) is a highly structured, formalised, and standardised neurological examination of the somatosensory function. Since its establishment in the 1990s, protocols have been developed for various forms of pain, musculoskeletal, neuropathic, and others, each focusing on elements most relevant to the pathophysiology of the specific type of pain. Quantitative sensory testing is meant to aid the understanding of mechanisms of pain in individuals, contributes to the development of new treatments, and provides a guide to using existing treatments in a more targeted way than the simple trial and error that is sadly still standard of care in many scenarios. However, due mostly to the wide variation of what must be considered normal range of sensation in humans, the signal-to-noise ratio of QST is far from ideal. Here, we lay out how machine learning and computational modelling will represent a step up for QST. It will enhance our understanding of influencing factors in individual QST profiles leading to highly specific, individualised reference values; it will help in designing treatment protocols that are targeted toward individual needs and thus less time- and resource-consuming; and it will allow to integrate QST with other means of neurological assessments for a more comprehensive neurophysiological in vivo picture, and thus ultimately significantly improve the use of QST for personalized pain medicine.