In:
Neuro-Oncology, Oxford University Press (OUP), Vol. 21, No. Supplement_6 ( 2019-11-11), p. vi149-vi150
Abstract:
Brain tumor resection requires careful assessment of the nature and malignancy of the tumor tissue during intervention, as this is the basis for the decision as to how radical surgery has to be. Typically, there is no tissue available for analysis prior to the surgery. Intraoperative histopathological examination of tumor samples is time consuming and puts the patient to a significant amount of strain due to ongoing anesthesia and prolonged surgery, thus there is a need to develop faster methods for thorough diagnostics. Raman spectroscopy allows for a fast analysis of samples without the need for labeling and sample preparation, making it a useful tool for peri- and intrasurgical application. We collect Raman spectra and correlated histopathological results to build a database of different tumor types. As intrasurgical use of Raman spectroscopy will typically be applied to native specimens, the respective database has to be based on native samples as well. We collected several thousands of measurements from more than 100 patients and continue to assemble a native database. Tumors that do not occur with a high frequency are not sufficiently represented in the database. Therefore, we investigate whether it is possible to integrate fixed and frozen specimens in the collection of Raman spectra and the classifier. Fixation results in characteristic changes to the biochemistry of biological tissues. We collect pairs of spectra from native and fixed as well as from native and frozen specimens and use the difference of the spectra to create an algorithm for filtering of the specific changes in the spectra. These data can be used to train a classifier for the identification of native samples. This opens the way for assessing large collections of pathological samples in fixed and frozen state to cover any type of neoplastic disease in the database.
Type of Medium:
Online Resource
ISSN:
1522-8517
,
1523-5866
DOI:
10.1093/neuonc/noz175.627
Language:
English
Publisher:
Oxford University Press (OUP)
Publication Date:
2019
detail.hit.zdb_id:
2094060-9
