RESUMO
BACKGROUND: This paper presents a novel tactile sensing robot designed to characterize the viscoelastic behaviour of breast tissue during clinical breast examination (CBE). The robot, named the 'robotic tactile breast mass identifier' (Robo-Tac-BMI), mainly consists of an indentation probe controlled by a robotic system and a visualization interface to manipulate the end-effector. Major high order mechanical parameters are extracted to characterize the viscoelastic behaviour of the tissue in order to certify mass detection and judge the mass type. By fusing the measurements over both breasts, the probe can generalize a mechanical image to visualize the viscoelastic distribution within the internal tissue. The viscoelastic properties provide additional and objective information to facilitate the surgeon's task in the decision-making process. METHODS: A new computational method is presented for characterizing mass existence in a real breast model. The effect of the strain rate on the mechanical properties is considered to study the viscoelastic behaviour of different mass types. A creative method is proposed to find the optimum strain rate for different mass positions and consistencies, which improves nodule detection. The computational results would be used as the basis for the design and set-up of the Robo-Tac-BMI. Clinical tests with statistical analyses were performed on 161 cases, using the Robo-Tac-BMI. RESULTS: The results include analysis of two high-order mechanical properties. Robo-Tac-BMI's capability of nodule detection and differentiation between benign and malignant mass types are investigated. The results were validated by 'gold standard' tests. CONCLUSION: The results show that Robo-Tac-BMI has the potential to provide high-order mechanical parameters, unlike the conventional screening modality carried out by the surgeon, which is not inclusive or quantitative and lacks effectiveness and documentation. The nodule detection ability of this device is confirmed statistically in clinical breast examinations. Differentiation between different mass types is reported as the preliminary result of Robo-Tac-BMI utilization.