Abstract
Variation in mechanical properties is a useful marker for cancer in soft tissue and has been used in clinical diagnosis for centuries. However, to develop such methods as instrumented palpation, there remain challenges in using the mechanical response during palpation to quantify tumor load. This study proposes a computational framework of identification and quantification of cancerous nodules in soft tissue without a priori knowledge of its geometry, size, and depth. The methodology, using prostate tissue as an exemplar, is based on instrumented palpation performed at positions with various indentation depths over the surface of the relevant structure (in this case, the prostate gland). The profile of force feedback results is then compared with the benchmark in silico models to estimate the size and depth of the cancerous nodule. The methodology is first demonstrated using computational models and then validated using tissue-mimicking gelatin phantoms, where the depth and volume of the tumor nodule is estimated with good accuracy. The proposed framework is capable of quantifying a tumor nodule in soft tissue without a priori information about its geometry, thus presenting great promise in clinical palpation diagnosis for a wide variety of solid tumors including breast and prostate cancer.