Development and validation of rapid intraoperative diagnosis system for malignant glioma
Description: We constructed a mass spectral database of human non-tumor and cerebral tumor tissues with labels of tumor grades (grade I-IV) and patient information. To confirm the discrimination ability of learning machine (logistic regression)-based diagnostic algorithm, stored data were divided into the sub groups, and leave-one-out cross-validation was performed. As a result, the discrimination accuracy for non-tumor/tumor or tumor grade were over 80%. This result suggest that our system is useful for diagnosis of human cerebral tumor. In addition, we constructed the real-time and remote sampling probe. It enables us to perform intraoperative diagnosis without extirpation of tissues.
Amount: ¥4,810,000
Funding type: research
Funder scheme: Grant-in-Aid for Scientific Research (C)
Start date: 2016-04-01
End date: 2019-03-31
Investigator: Kentaro YOSHIMURA (lead)
Institution: University of Yamanashi
Subfield: Spectroscopy
Field: Chemistry
Domain: Physical Sciences
Funded outputs count: 8
Funder award ID: 16K08964
OpenAlex ID: G7993773919
