A feasibility study of a novel spectral method using radiofrequency ultrasound data for monitoring laser tissue ablation.

This paper presents preliminary results of a new non-invasive ultrasound monitoring method called TUV (Thermotherapy Ultrasonic View) able to investigate structural tissue modifications caused by minimally invasive percutaneous laser ablation. The method, based on the spectral analysis of the raw ultrasound radiofrequency signal, develops spectral parameters in a multidimensional space and its N dimensions are represented by the central frequencies of the sub bands the signal spectrum is decomposed into. Signal processing has been performed on the data related to 7 laser treatments performed on 4 samples of removed prostatic glands which underwent laser ablation at power of 3W, 4W and 5W and energy of 1800J. In this preliminary study, clusters of these parameters, referred to tissue areas at different distances from the light laser source, modified their shape and position in different ways, during ablation treatment. TUV results have been represented by a chromatic code superimposed to the corresponding ultrasound conventional image, in order to highlight the alteration intensities occurred in the ablated tissue. Resulting images of ablated area have been compared to histological specimens to evaluate the degree of similarity between them by means of Dice and Jaccard coefficients.