Fig 1.
(a) Palpation in a tele-manipulation environment; (b) Palpation in a data-driven virtual tissue model. Haptic device Geomagic Touch is used to control the position; Fanuc robot follows the trajectory; force sensing probe scans the silicone phantom; virtual tissue displays on a computer monitor; probe position is displayed in real time; stiffness distribution color map is generated using force and indentation depth data.
Fig 2.
Schematic of the experimental haptic manipulator.
The right column shows the slave-side hardware: a camera, a robot arm, a sliding indentation probe, and a silicone phantom tissue. The left column shows the master-side configuration: a live camera image, a visual stiffness display, and force feedback via a haptic device. The middle column lists the software.
Fig 3.
Tele-operation architecture.
Fig 4.
Hermite curve interpolation trajectory generation.
Fig 5.
(a) when no force feedback is applied and (b) when force feedback is applied.
Fig 6.
(a) Mapping stiffness data to RGB value and (b) stiffness map generation.
Fig 7.
(a) the locations of the 3 embedded nodules are highlighted (A, B, C); (b) an operator remotely palpated the phantom tissue by using the same trajectory, which covers nodules A and B, guided by the two black tags.
Fig 8.
Stiffness map estimated from the perpendicular reaction force along the same trajectory in multiple trials of remote palpation with increased velocity from trials 1 to 7.
Nodules A and B are presented in red or orange, whereas other areas are presented in blue or cyan.
Table 1.
Overview of participant demographics and experience in the palpation experiment with the tele-operation system (Experiment II).
Fig 9.
Two stiffness maps generated during 2 trials by visual stiffness feedback.
The upper maps show the orientations of the artificial organ, whereas the lower maps show the corresponding stiffness.
Fig 10.
(a) Nodule detection sensitivities of visual stiffness feedback and force feedback and Wilson score intervals at a 95% confidence level; (b) Detection sensitivities of each nodule and Wilson score intervals at a 95% confidence level; (c) Time elapsed to determine nodule locations by visual stiffness feedback and force feedback in a tele-manipulation environment: data were averaged over all 10 subjects, and standard error bars are shown (The standard error of the mean is the standard deviation of the sampling distribution of a statistic [34]; it is an indicator of result precision).
Fig 11.
User interfaces of the nodule size discrimination experiment.
The buried nodule is marked using black hollow squares; the orientation of the phantom tissue is changed for each trial.
Table 2.
Overview of participant demographics and experience in the virtual palpation experiment (Experiment III).
Fig 12.
Results of nodule size discrimination.
(a) Overall discrimination rates of nodule sizes by the two stiffness feedback modes and their Wilson score intervals at a 95% confidence level; (b) Discrimination rates of the nodule sizes of each nodule pair and their Wilson score intervals at a 95% confidence level.