Motivation

Questions of survival, outcome and recurrence in patients with brain tumors still remain critical in spite of progress in chemotherapy, neurosurgery, and intraoperative surgical navigation. The crucial point in patient’s survival is the most complete resection of neoplastic tissues. Great achievements in developing and producing new medical techniques can and at the end lead to improving results of treatment. Nevertheless, on-site pathology provides the most accurate diagnosis, which is a guide-line for every neurosurgeon. But performing of the express pathology analysis is time-consuming and expensive and as such couldn’t be using for routinely fast tumor detection. Hence, it is necessary to develop a new tool for fast real-time diagnostic of tumors with high sensitivity and specificity. We think that combined using of autofluorescence, reemission and electrical impedance for brain tumors diagnostics could resolve this question.

Objectives

Fast real-time diagnostic of tumors with high sensitivity and specificity

Principle

Spectroscopic and impedance data are collected from tissue specimens by means of LIMES device (Berlin, Germany) equipped by nitrogen laser, custom-designed optical probe, CCD-camera, tool for impedance measurement and two PC with special software. A high-pressure pulsed nitrogen laser at 337,1 nm (Berlin, Germany) is used as the excitation source for fluorescence spectroscopy. The light from a pulsed nitrogen laser is focused on the custom-designed fiber-optic probe and than direct on the sample. The optical probe is than placed lightly in direct contact with the target tissue and normal and 3D- spectra of autofluorescence, and impedance spectra are sequentially acquired. Spectra of autofluorescence are recorded in the visible range. The nitrogen laser is operated at 20 Hz repetition rate, 2 ns pulse width and average pulse energy of 100 ±5 µJ. Multivariate statistical techniques, including principal component analysis and liner discriminant analysis are used to develop diagnostic algorithms for differentiating tumors from normal tissue.

Experimental setup for in vitro measurements.

Classification Algorithm

Results

Using analysis of the autofluorescence spectra, reemission spectra and electrical impedance spectra we were able to find the crucial distinguishes between normal brain tissues and different kind of brain tumors in vitro.

Table of classification samples using diagnostic algorithms.

Applications

Cancer diagnostics of different tissues and organs, such as the cervix, esophagus, colon, bronchi, breast and brain.

Contact

Andrey N. Sobchuk

B.I.Stepanov Institute of Physics National Academy of Sciences of Belarus
Pr. Nezavisimosti Ave. 68
Minsk 220072 Belarus
Phone: +375 17 2840970
Fax: +375 17 2840879
E-mail: This email address is being protected from spam bots, you need Javascript enabled to view it  

Dr.Siegfried Greschner

Fraunhofer Institute for Nondestructive Testing,
Dresden, Germany
Phone/Fax: +49 (0)30 444 1630
Fax: +375 17 2840879
E-mail: This email address is being protected from spam bots, you need Javascript enabled to view it