@article{Lalitharatne:2021:10.1109/LRA.2020.3048670,
author = {Lalitharatne, SWH and Tan, Y and He, L and Leong, F and Van, Zalk N and De, Lusignan S and Iida, F and Nanayakkara, T},
doi = {10.1109/LRA.2020.3048670},
journal = {IEEE Robotics and Automation Letters},
pages = {643--650},
title = {MorphFace: a hybrid morphable face for a robopatient},
url = {http://dx.doi.org/10.1109/LRA.2020.3048670},
volume = {6},
year = {2021}
}

Download

TY  - JOUR
AB  - Physicians use pain expressions shown in a patient’sface to regulate their palpation methods during physical examination.Training to interpret patients’ facial expressions withdifferent genders and ethnicities still remains a challenge, takingnovices a long time to learn through experience. This paperpresents MorphFace: a controllable 3D physical-virtual hybridface to represent pain expressions of patients from differentethnicity-gender backgrounds. It is also an intermediate stepto expose trainee physicians to the gender and ethnic diversityof patients. We extracted four principal components from theChicago Face Database to design a four degrees of freedom(DoF) physical face controlled via tendons to span   85% offacial variations among gender and ethnicity. Details such as skincolour, skin texture, and facial expressions are synthesized by avirtual model and projected onto the 3D physical face via a frontmountedLED projector to obtain a hybrid controllable patientface simulator. A user study revealed that certain differences inethnicity between the observer and the MorphFace lead to differentperceived pain intensity for the same pain level rendered bythe MorphFace. This highlights the value of having MorphFace asa controllable hybrid simulator to quantify perceptual differencesduring physician training.
AU  - Lalitharatne,SWH
AU  - Tan,Y
AU  - He,L
AU  - Leong,F
AU  - Van,Zalk N
AU  - De,Lusignan S
AU  - Iida,F
AU  - Nanayakkara,T
DO  - 10.1109/LRA.2020.3048670
EP  - 650
PY  - 2021///
SN  - 2377-3766
SP  - 643
TI  - MorphFace: a hybrid morphable face for a robopatient
T2  - IEEE Robotics and Automation Letters
UR  - http://dx.doi.org/10.1109/LRA.2020.3048670
UR  - http://hdl.handle.net/10044/1/86334
VL  - 6
ER  - 

Download