Radiomics and radiogenomics in pediatric neuro-oncology: A review

Abstract

The current era of advanced computing has allowed for the development and implementation of the field of radiomics. In pediatric neuro-oncology, radiomics has been applied in determination of tumor histology, identification of disseminated disease, prognostication, and molecular classification of tumors (i.e., radiogenomics). The field also comes with many challenges, such as limitations in study sample sizes, class imbalance, generalizability of the methods, and data harmonization across imaging centers. The aim of this review paper is two-fold: first, to summarize existing literature in radiomics of pediatric neuro-oncology; second, to distill the themes and challenges of the field and discuss future directions in both a clinical and technical context.

Automated Tumor Segmentation and Brain Tissue Extraction from Multiparametric MRI of Pediatric Brain Tumors: A Multi-Institutional Study

“…Accurate whole tumor segmentation and reliable subregion differentiation is critical for volumetric measurements for surveillance of tumor progression and response assessment 6 , as well as for radiomic studies in neuro-oncology 25 . However, the process of manually generating 3D segmentations imposes high manual burden on the readers and is prone to variability and subjectivity 8 .…”

Unsupervised machine learning using K-means identifies radiomic subgroups of pediatric low-grade gliomas that correlate with key molecular markers

“…Future works should thus focus on the expansion of cohorts both in terms of number of subjects and diversity of tumors. Cohort size is a universal limitation in radiomic works in pediatric brain tumor populations [26] . Multi-institutional collaboration is the ideal solution to this challenge and is indeed the model upon which open-access data repositories, such as those available via PedCBioportal, are built.…”

Current state of pediatric neuro-oncology imaging, challenges and future directions

“…In other words, radiogenomics may serve as a “virtual biopsy” tool to portray the molecular landscape of tumors based on readily available MRI scans and bridge the gap between imaging and genomics disciplines [76 , 77] . We refer the readers interested to learn about the existing radiomic and radiogenomic studies in pediatric brain tumors to a recent review paper [79] .…”

“…While the field of radiomics in adult brain tumors has witnessed exponential growth over the past decade, such studies have been fairly limited in pediatric brain tumors [79] , which mainly could be due to significantly lower incidence rate (on the order of one fifth) of brain tumors in children compared to their adult counterparts [80] . .…”

“…Such consortiums are paving the path towards advancement of discoveries and available tools for diagnosis and treatment of pediatric brain tumors by increasing the statistical power for research studies. In the future, addition of data from multiple centers across the globe can further contribute to development of more generalizable predictive tools based on widely available MRI scans, built based on diverse datasets that are inclusive of patients from different countries with different economic and social backgrounds [79] . Such tools can allow their application in lower-income countries without access to sophisticated sequencing methods.…”