Personalized functional brain models of nociceptive and chronic pain

Pain experience is personal, and its brain representations exhibit a high degree of individual variability. This poses a challenge in developing a population-level brain biomarker of pain that works for everyone. In this talk, I will introduce our recent efforts to develop personalized neuroimaging markers of pain across two studies. We conducted an extensive sampling approach from a small number of healthy and clinical participants and developed personalized fMRI-based predictive models. Our findings demonstrate that personalized markers outperform group-level models when sufficient data are available for each individual, potentially capturing unique internal factors influencing pain. Moreover, we observed that brain representations of pain change even within individuals over time. These changes likely reflect dynamic internal changes in the brain, whether functional or structural. These findings emphasize the need to understand internal factors for pain better and incorporate them into brain models to enhance their predictive power and clinical relevance.