Discrimination of infiltrative glioma boundary based on laser-induced breakdown spectroscopy

Abstract Glioma is a kind of brain tumor which infiltrates from the primary tumor to surrounding healthy tissue. Due to the difficulty in determining the tumor infiltrative boundary during surgery, it has both high recurrence rate and high mortality rate. In this paper, an intraoperative discrimination method of infiltrative glioma boundary was proposed based on laser-induced breakdown spectroscopy (LIBS). The elemental information was used for intraoperative glioma boundary detection for the first time. Theoretical verification was performed on paraffin-embedded samples. Meanwhile, the formation mechanism of spectral lines and elemental composition were analyzed. Afterwards, fresh glioma samples and border infiltrated tissues were detected during the operation. Combined with support vector machine (SVM) and feature selection method, the correct classification rate (CCR) of the glioma and boundary tissue achieved 95%.