Radiomics and radiogenomics in gliomas: a contemporary update

Abstract

The natural history and treatment landscape of primary brain tumours are complicated by the varied tumour behaviour of primary or secondary gliomas (high-grade transformation of low-grade lesions), as well as the dilemmas with identification of radiation necrosis, tumour progression, and pseudoprogression on MRI. Radiomics and radiogenomics promise to offer precise diagnosis, predict prognosis, and assess tumour response to modern chemotherapy/immunotherapy and radiation therapy. This is achieved by a triumvirate of morphological, textural, and functional signatures, derived from a high-throughput extraction of quantitative voxel-level MR image metrics. However, the lack of standardisation of acquisition parameters and inconsistent methodology between working groups have made validations unreliable, hence multi-centre studies involving heterogenous study populations are warranted. We elucidate novel radiomic and radiogenomic workflow concepts and state-of-the-art descriptors in sub-visual MR image processing, with relevant literature on applications of such machine learning techniques in glioma management.

Artificial Intelligence in Brain Tumour Surgery—An Emerging Paradigm

“…Increasingly, radiological images are being recognised as three-dimensional data-sets amenable to algorithmic analysis, deeper than the images seen by individual clinicians [ 32 ]. So-called “radiomics” are able to compute swathes of image-based data, down to individual three-dimensional pixels known as “voxels”, to detect and identify patterns that are typically subtle [ 33 ]. Thus far, radiomics has been shown to be effective in detecting and characterising a wide range of tumours throughout the human body [ 32 , 34 , 35 , 36 , 37 ].…”

“…Thus far, radiomics has been shown to be effective in detecting and characterising a wide range of tumours throughout the human body [ 32 , 34 , 35 , 36 , 37 ]. Specific to brain tumour diagnosis, ML algorithms have been used to characterise the molecular expression of brain tumours [ 33 , 38 , 39 , 40 , 41 , 42 ], aid in the detection of central nervous system (CNS) metastases [ 43 , 44 , 45 , 46 , 47 , 48 , 49 , 50 , 51 , 52 , 53 ], discriminate between primary and metastatic CNS lesions [ 54 ], predict the brain tumour grade [ 55 , 56 , 57 , 58 , 59 , 60 , 61 , 62 ], and predict the presence of genetic mutations [ 38 , 63 , 64 , 65 , 66 ] among other applications [ 67 ]. These findings have been demonstrated in a range of CNS tumours, including meningiomas, glioblastoma, and CNS metastases.…”

Clinical applications of diffusion-weighted sequence in brain imaging: beyond stroke

“…Radiomics incorporates artificial intelligence (AI) in brain tumors such as GBM assessment using computerized data from images that provide subvisual radiological cues. Radiomics information correlates with genomic information providing information on tumor response to therapy [93]. Initial research has shown that both diffusion and perfusion imaging data included in radiomics provide better prognostication of newly diagnosed GBM [94].…”

Editorial. Long-term solutions in neurosurgery using extended reality technologies

“…23 Emerging techniques such as radiomics (subvisual characterization of radiological imaging) can also be used in conjunction with VR to provide complementary information regarding tissue morphology and texture. 24 Currently, MR has been incorporated into selected neurosurgical research labs in the United States. 25 With the advent of practical and user-friendly devices coupled with high-speed internet, artificial intelligence (AI), and robotics, extended reality technology has the potential to effectively train surgeons worldwide.…”