Dynamic instrumented palpation - a new method for soft tissue quality assessment: application to prostate disease diagnosis

摘要

The objective is to establish the feasibility of using dynamic instrumented palpation, a novel technique of low-frequency mechanical testing, applied here to diagnose soft tissue condition. The technique is applied, in vitro, to samples of excised prostate gland affected by benign prostate hyperplasia and/or prostate cancer. Particular attention is paid to the relationship between the histological structure of the tissue and the dynamic mechanical properties in an attempt to separate patient-specific aspects from histopathological condition (i.e. prostate cancer or benign prostate hyperplasia). The technique is of clinical interest because it is potentially deployable in vivo. Prostate samples were obtained from a total of 36 patients who had undergone transurethral resection of the prostate to relieve prostatic obstruction and 4 patients who had undergone radical cystoprostatectomy for bladder cancer. Specimens (chips) recovered from transurethral resection of the prostate were of nominal size 5mmx8mm and thicknesses between 2 and 4mm, whereas those from the cystoprostatectomy were in the form of transverse slices of thickness approximately 6mm. Specimens were mechanically tested by a controlled strain cyclic compression technique, and the resulting dynamic mechanical properties expressed as the amplitude ratio and phase difference between the cyclic stress and cyclic strain. After mechanical testing, the percentage areas of glandular and smooth muscle were measured at each probe point. Good contrast between the dynamic modulus of chips from benign prostate hyperplasia and prostate cancer patients was demonstrated, and absolute values similar to those published by other authors are reported. For the slices, modulus values were considerably higher than for chips, and good in-patient mechanical contrast was revealed for predominantly nodular and predominantly stromal areas. Extending this classification between patients required pattern recognition techniques. Overal