Modeling and experimental testing of three-tip configuration tactile sensor for compensating the error due to soft tissue surface irregularities during stiffness detection

In conventional open surgeries, surgeons usually palpate soft tissues by their fingertips in order to investigate their healthiness and/or interior neoplasm. However, with the spread of minimally invasive surgery (MIS) systems, surgeons do not have access to tactile information as in open surgeries. Therefore, the need for tactile sensors became indispensable. This paper presents a new three-spring tactile sensor design for soft tissue stiffness detection, which is based on one of the most popular designs that uses two springs with different stiffness. The modified design utilizes three springs with different stiffness, instead of two springs, in order to compensate for errors arising from different contact conditions (applied load and contact angle). Laser microscope was used to scan different non-viable tissues (chicken organs) in order to identify the error arising from surface irregularities. Experimental measurements showed that, soft tissue irregularities can result in a maximum inclination angle of 3°, between a two-tip sensor and the tissue surface. After that, the effect of contact angle on the two-spring sensor output was investigated using finite element modeling (FEM). Results showed that a maximum error of 103% occurs in the sensor output. An analytical model for the modified three-tip sensor was developed, and proved that the sensor can measure tissue stiffness independent of contact conditions. The analytical model was verified using FEM, and the results showed that the error in the sensor output is not more than 0.25%, for an inclination angle up to ±6° (twice the measured value). Finally, a prototype of the three-tip configuration sensor, on the macro scale, was fabricated and tested, in order to verify the sensor performance. Experimental results showed that the new configuration has the ability to measure soft tissue stiffness independent of the applied load and contact angle (±3°) with a maximum error of 14%.