Publication detail

The use of machine learning for non-invasive classification of brain pathologies

KOSTIAL, M. MARCOŇ, P. SOLÁR, P.

Original Title

The use of machine learning for non-invasive classification of brain pathologies

Type

article in a collection out of WoS and Scopus

Language

English

Original Abstract

Accurate classification and spatial delineation of brain pathologies is crucial for correct diagnosis and effective treatment. Currently, CT (computed tomography) and MRI (magnetic resonance imaging) are used for this purpose, and when suspected, a biopsy is performed. The aim of this study is to demonstrate the potential of using machine learning to identify areas of a given pathology based on the diffusivity of individual tissues. KNN (knearest neighbours) and SVM (support vector machine) models were learned on a dataset containing data from patients with glioblastoma Multiforme and Abscessus Cerebri, and then their performance was investigated. The obtained results indicate the high accuracy of the models, which only supports their possibilities and potential for future use in automated diagnostic tools that will reduce the use of biopsy and speed up the whole process.

Keywords

brain tumour, pathology, artificial intelligence, machine learning

Authors

KOSTIAL, M.; MARCOŇ, P.; SOLÁR, P.

Released

25. 4. 2023

Location

Brno

Pages from

1

Pages to

5

Pages count

5

URL

BibTex

@inproceedings{BUT183989,
  author="Martin {Kostial} and Petr {Marcoň} and Peter {Solár}",
  title="The use of machine learning for non-invasive classification of brain pathologies",
  year="2023",
  pages="1--5",
  address="Brno",
  url="https://www.eeict.cz/download"
}