Radiomics: Principles and radiotherapy applications
Section snippets
Concept of radiomics
Medical imaging, such as computed tomography (CT), positron emission tomography (PET) using FDG radioactive glucose analogue, and magnetic resonance imaging (MRI) are used routinely in the clinical management of cancer patients. They tend to play a leading role in personalised treatment based on tumour imaging. Personalised medicine has been largely developed using invasive techniques based on genomics and proteomics. However, there is a spatial and temporal heterogeneity of tumour features (
Image features
Numerous image features have been proposed in the literature (Sollini et al., 2017). Features based on the shape and size of the lesion, histogram features based on first-order statistics, texture features, and filter- and model-based features are generally distinguished (see Fig. 1).
Medical applications
Various reviews of the literature (Marusyk et al., 2012; Kumar et al., 2012; Yip and Aerts, 2016; Sollini et al., 2017; Scalco and Rizzo, 2017) show that many studies support the clinical interest of radiomics in oncology, both in MR, CT, and FDG PET imaging, particularly for patients treated using radiotherapy for many localisations (NSCLC, H&N, oesophagus, breast, cervix, etc.). Before treatment, this concerns the staging of the disease and the identification of genetic features. For
Factors influencing radiomic features
The very encouraging results of radiomics, however, raise the issue of the robustness of features with respect to the experimental conditions and image processing performed before their extraction as well as the manner of extraction.
Methods of radiomic analysis
The ever-growing number of features studied raises the question of the statistical method of analysis used and its quality, especially as patient databases can be limited.
Conclusions and perspectives
Numerous articles support the clinical use of radiomic analysis in oncology, as well as on MRI, CT than on FDG PET images, in staging of the disease, significantly differentiating the early and advanced stages of the disease, tissue discrimination between healthy and pathological tissues, demonstrating genetic characteristics, predicting the response to treatment, patient survival, and side effects. These results from the literature mainly concern solid tumours for many localisations
Financial disclosure statement
This work was supported by grant from the Cancéropôle Nord-Ouest, France.
References (63)
- et al.
Quantifying heterogeneity in human tumours using MRI and PET
Eur. J. Cancer
(2012) - et al.
Reproducibility and prognosis of quantitative features extracted from CT images
Transl. Oncol.
(2014) - et al.
CT-based radiomic signature predicts distant metastasis in lung adenocarcinoma
Radiother. Oncol.
(2015) - et al.
Lung texture in serial thoracic computed tomography scans: correlation of radiomics-based features with radiation therapy dose and radiation pneumonitis development
Int. J. Radiat. Oncol. Biol. Phys.
(2015) - et al.
New response evaluation criteria in solid tumours: revised RECIST guideline (version 1.1)
Eur. J. Cancer
(2009) - et al.
Exploring feature-based approaches in PET images for predicting cancer treatment outcomes
Pattern Recognit.
(2009) - et al.
Preliminary investigation into sources of uncertainty in quantitative imaging features
Comput. Med. Imaging Graph.
(2015) Texture analysis using gray level run lengths
Comput Graph Image Process
(1975)- et al.
Radiomics: the process and the challenges
Magn. Reson. Imaging
(2012) - et al.
Radiomics: extracting more information from medical images using advanced feature analysis
Eur. J. Cancer
(2012)