@inbook{2d7cffac64814e27b0e1ed5dee737ebb,
title = "MRI morphometry in brain tumors: Challenges and opportunities in expert, radiomic, and deep-learning-based analyses",
abstract = "Morphometry refers to the quantitative study of form, which has gained popularity in neurosciences for non-invasive in vivo evaluation of the normal and aging brain through the use of neuroimaging data, and hence designated as brain morphometry. In the rapidly evolving field of neuro-oncology, morphological evaluation provided by neuroimaging studies has been a cornerstone for the initial diagnosis, classification, management, and post-treatment follow-up of brain tumors. However, it has historically relied on predominantly subjective and qualitative observations made by imaging experts based on clinical experience. The wealth of knowledge obtained through visual inspection of tumor imaging has made remarkable contributions to the field and enhanced our understanding of tumor biology and natural history; however, further developments have been hampered by the lack of robust methods for more automated and quantitative evaluation. These methods are becoming more readily available and have been fueled by breakthrough developments in imaging post-processing and artificial intelligence. In this chapter, we review past contributions and evolution of the field of brain tumor morphological evaluation as it evolves into more automated computerized methods including radiomics and deep learning.",
keywords = "Brain tumor, Deep learning, Glioma, MRI, Morphometry, Radiogenomics, Radiomics",
author = "Pinho, {Marco C.} and Kaustav Bera and Niha Beig and Pallavi Tiwari",
note = "Publisher Copyright: {\textcopyright} 2021, Springer Science+Business Media, LLC, part of Springer Nature.",
year = "2021",
doi = "10.1007/978-1-0716-0856-2_14",
language = "English (US)",
series = "Neuromethods",
publisher = "Humana Press Inc.",
pages = "323--368",
booktitle = "Neuromethods",
}