Skip to content

Robopatient

Robopatient is a medical robotic simulator which investigates how human participants use haptic exploration behaviours to detect abnormalities in a soft tissue given palpation force constraints. Research around this platform focuses on training medical students to perform physical examinations on patients under constraints imposed by pain expressions conditioned by different gender and culture backgrounds. Robopatient is currently capable of generating "pain" based on pain expressions such as facial expressions and pain sound based on palpation during diagnosis. I am using this robotic platform to study the embodied generation of pain expressions.

We examined how a personalised approach to display pain expressions of individuals affect human preferences for pain expressivity in robotic patients. We used Proximal Policy Control to adapt pain expressions based on human input.

Read full paper here

Palpation Alters Auditory Pain Expressions With Gender-Specific Variations in Robopatients
IEEE Transactions on Medical Robotics and Bionics, 2025

RoboPatient could further synthesise vocal and facial pain expressions from haptic input during palpation to support clinician training and reduce diagnostic errors. A user study with 7,680 trials shows that pain perception is primarily driven by sound pitch and amplitude, with stronger forces producing more realistic and congruent pain responses.

Read full paper here

Auditory-Tactile Congruence for Synthesis of Adaptive Pain Expressions in RoboPatients
Advanced Robotics Research, 2026 (Accepted)