Radiomics
≈ 6 min read
Names in other languages
No names in other languages are held for this entry yet.
Article bodies are English and are not machine-translated. A name in another language is not a translation of the article.
In the field of medicine, radiomics is a method that extracts a large number of features from medical images using data-characterisation algorithms. These features, termed radiomic features, have the potential to uncover tumoral patterns and characteristics that fail to be perceived by the naked eye. The hypothesis of radiomics is that the distinctive imaging features between disease forms may be useful for predicting prognosis and therapeutic response for various cancer types, thus providing valuable information for personalized therapy. Radiomics emerged from the medical fields of radiology and oncology and is the most advanced in applications within these fields. However, the technique can be applied to any medical study where a pathological process can be imaged.
In the field of medicine, radiomics is a method that extracts a large number of features from medical images using data-characterisation algorithms. These features, termed radiomic features, have the potential to uncover tumoral patterns and characteristics that fail to be perceived by the naked eye. The hypothesis of radiomics is that the distinctive imaging features between disease forms may be useful for predicting prognosis and therapeutic response for various cancer types, thus providing valuable information for personalized therapy. Radiomics emerged from the medical fields of radiology and oncology and is the most advanced in applications within these fields. However, the technique can be applied to any medical study where a pathological process can be imaged.
Applications
Nasief et al. (2019) showed that changes of radiomic features over time in longitudinal images (delta-radiomic features, DRFs) can potentially be used as a biomarker to predict treatment response for pancreatic cancer. Their results showed that a Bayesian regularization neural network can be used to identify a subset of DRFs that demonstrated significant changes between good- and bad- responders following 2–4 weeks of treatment with an AUC = 0.94.
They also showed (Nasief et al., 2020) that DRFs are independent predictor of survival and if combined with the clinical biomarker CA19-9 can improve treatment response prediction and increase the possibility for response-based treatment adaptation . Several studies have also showed that radiomic features are better at predicting treatment response than conventional measures, such as tumor volume and diameter, and the maximum radiotracer uptake on positron emission tomography (PET) imaging.
Using this technique an algorithm has been developed, after initial training based on intra tumor lymphocyte density, to predict the probability of tumor response to immunotherapy, providing a demonstration of the clinical potential of radiomics as a powerful tool for personalized therapy in the emerging field of immunooncology. Other studies have also demonstrated the utility of radiomics for predicting immunotherapy response of NSCLC patients using pre-treatment CT and PET/CT images.
Radiomics remains inferior to conventional techniques in some applications, suggesting the necessity of continued improvement and manipulation of Radiomics features to different clinical scenarios. For instance, Ludwig et al. (2020) demonstrated that morphological Radiomics features were inferior to previously established features in the discrimination of intracranial aneurysm rupture status from 3-dimensional rotational angiography.
Prognostication
Radiomic studies have shown that image-based markers have the potential to provide information orthogonal to staging and biomarkers and improve prognostication.
Prediction of physiological events
Radiomics can also be used to identify challenging physiological events such as brain activity, which is usually studied with imaging techniques such as functional MRI "fMRI". FMRI raw images can undergo radiomic analysis to generate imaging features that can be later correlated with meaningful brain activity.
Imaging genomics
In imaging genomics, radiogenomics can be used to create imaging biomarkers that can identify the genomics of a disease, especially cancer without the use of a biopsy. Various techniques for dealing with high-dimensional data are used to find statistically significant correlations between MRI, CT, and PET imaging features and the genomics of disease, including SAM, VAMPIRE, and GSEA.
The imaging radiogenomic approach has proven successful in determining the MRI phenotype associated genetics of glioblastoma, a highly aggressive type of brain tumor with low prognosis. The first large-scale MR-imaging microRNA-mRNA correlative study in GBM was published by Zinn et al. in 2011 Similar studies in liver cancer have successfully determined much of the liver cancer genome from non-invasive imaging features.
Gevaert et al. at Stanford University have shown the potential to link image features of non-small cell lung nodules in CT scans to predict survival by leveraging publicly available gene expression data. This publication was accompanied by an editorial discussing the synergy between imaging and genomics.
More recently, Mu Zhou et al. at Stanford University have shown that multiple associations between semantic image features and metagenes that represented canonical molecular pathways, and it can result in noninvasive identification of molecular properties of non-small cell lung cancer. Several radiogenomic studies have now been carried out in prostate cancer, Some have noted that genetic features correlated with MRI signal are often also associated with more aggressive prostate cancer.
A systematic review of the genetic features found in more visible lesions on MRI identified multiple studies which had found loss of the tumour suppressor PTEN, increased gene expression linked to cell proliferation as well as cell-ECM interactions. This may indicate that certain genetic features drives cellular changes which ultimately effect fluid movement which can be seen on MRI and these features are predominantly associated with poor prognosis.
The combination of more dangerous genetic alterations, histology and clinical outcomes for patients with prostate tumours which are visible on mpMRI, has led to suggestions that the definition of 'clinically significant cancer' should be at least in part based on mpMRI findings. The radiogenomic approach has been also successfully applied in breast cancer.
In 2014, Mazurowski et al. showed that enhancement dynamics in MRI, computed using computer vision algorithms, are associated with gene expression-based tumor molecular subtype in breast cancer patients. Programs that study the connections between radiology and genomics are active at the University of Pennsylvania, UCLA, MD Anderson Cancer Center, Stanford University and at Baylor College of Medicine in Houston, Texas.
Similarly, the MPRAD features in brain stroke demonstrated increased performance in distinguishing the perfusion-diffusion mismatch compared to single parameter radiomics and there were no differences within the white and gray matter tissue. The majority of the single radiomic second order features (GLCM) did not show any significant textural difference between infarcted tissue and tissue at risk on the ADC map. Whereas the same second order multiparametric radiomic features (TSPM) were significantly different for the DWI dataset.
Similarly, multiparametric radiomic values for the TTP and PWI dataset demonstrated excellent results for the MPRAD. The MPRAD TSPM Entropy exhibited significant difference between infarcted tissue and potential tissue-at-risk: (6.6±0.5 vs 8.4±0.3, p=0.01).
Text derived from https://en.wikipedia.org/w/index.php?oldid=1372978214 under CC BY-SA 4.0; modifications have been made.