China 
Africa 
Explore chapters and articles related to this topic
Applying AI in Medical Imaging
Published in Lia Morra, Silvia Delsanto, Loredana Correale, Artificial Intelligence in Medical Imaging, 2019
Lia Morra, Silvia Delsanto, Loredana Correale
Similarly to CAD, radiomics have seen the most important developments in oncology applications. One of the more interesting and novel aspects of radiomics is that quantitative image features offer distinctive information on tumor phenotype and microenvironment (or habitat). Such information, currently underutilized, could provide information that is correlated to genomic and proteomics patterns. Furthermore, radiomics can be linked with the concept of radio-genomics, which assumes that imaging features are related to gene signatures [184; 186]. This hypothesis, which was postulated in the seminal work by Lambin and collagues, is increasingly supported by experimental evidence. For instance, Kuo and colleagues identified hepatocellular carcinoma imaging phenotypes that correlated with a doxorubicin drug response gene expression program [187], suggesting that radio-genomic analyses could be used to guide the selection of therapy for individual tumors [188].
Radiomics and quantitative imaging
Published in Jun Deng, Lei Xing, Big Data in Radiation Oncology, 2019
Dennis Mackin, Laurence E. Court
The primary difference between radiomics and other forms of quantitative imaging is the approach. Radiomics was originally defined as a method involving the “automated high-throughput extraction of large amounts (200+) of quantitative features of medical images” (Lambin et al. 2012). Instead of starting with a few preselected and targeted features, radiomics starts with hundreds of features and then applies statistical methods to find the most effective features for classifying, predicting, or estimating the quantity of interest. Although some studies discuss the interpretation of features after they have been determined to be prognostic (Van Dijk et al. 2017), interpretability is not a requirement in radiomics. Instead, the primary goal for most studies is to produce the most accurate and generalizable models.
Precision Imaging of Prostate Cancer
Published in Ayman El-Baz, Gyan Pareek, Jasjit S. Suri, Prostate Cancer Imaging, 2018
Radiomics is an emerging field for the quantification of tumor phenotypes by applying a large number of quantitative image features [113,114]. Radiomics can provide complementary and interchangeable information to improve individualized treatment selection and monitoring. Since medical imaging technology is routinely used in clinical practice worldwide, radiomics may have a high clinical impact on future patient management. The workflow of radiomics consists of three steps [113]. The first step is the acquisition of standardized images for diagnostic or planning purposes. On the images, the tumor regions are extracted by an algorithm or an experienced radiologist. Second, quantitative imaging features are extracted from the tumor regions. These features involve tumor image intensity, texture, and shape and size of the tumor. Last, all the extracted features are analyzed and selected by a model. The most informative features are identified and incorporated into predictive models for treatment outcome. Radiomics, as a high-dimensional mineable feature space, can be used for prostate cancer. Cameron et al. constructed a comprehensive radiomics feature model to detect tumorous regions using mpMRI [115]. New radiomics-driven texture feature models have been developed for the detection of prostate cancer and for the classification of prostate cancer Gleason scores by utilizing mpMRI data [116–118].
Brain tumour segmentation and survival prognostication using 3D radiomics features and machine learning algorithms
Published in Computer Methods in Biomechanics and Biomedical Engineering: Imaging & Visualization, 2023
J. Glory Precious, I. Keren Evangeline, S. P. Angeline Kirubha
Researchers have been focusing on radiomics attributes for tumour diagnosis and survival prognostication for the past few years. Radiomics is a broad term that utilises numerical or statistical methods to derive an immense number of quantitative features from a variety of medical imaging modalities to build and anticipate models with the goal of enabling specialised clinical management. Radiomics attributes are quantitative representations of the intensity, shape and texture of the volume of interest (VOI). The radiomics features are calculated using a variety of statistical methods. First-order statistics, based on a single voxel; second-order statistics (grey level co-occurrence matrix [GLCM] features), based on connections between two voxels and higher-order statistics, for example, neighbourhood grey-tone difference matrices features are among the methods utilised (Shboul et al. 2018; Bakas et al. 2017; Seow et al. 2018; Hussain et al. 2018; Osman 2019; Sajid et al. 2019). The texture and contour changes in the tumour and peritumoral regions might be caused by irregular and aggressive tumour invasion. The major axis, minor axis and the elongation of the shape of the segmented partitions are some of the crucial shape aspects. In the proposed work, pyradiomics tool was used for feature extraction.
Automated segmentation of osteoblastic vertebral metastasis: a radiomics approach
Published in Computer Methods in Biomechanics and Biomedical Engineering: Imaging & Visualization, 2021
Allison Clement, Cari Whyne, Michael Hardisty
Image texture features can be calculated using intensity and spatial information in medical images and have shown utility in delineating tumours and for the quantification of mechanical properties (Cendre et al. 2000; Geremia et al. 2013; Echegaray et al. 2015). The field of radiomics involves feature extraction from medical imaging data and data-mining of these features. A reference standard of radiomics features has been established and made available to the research community through Pyradiomics, an open-source library written in python which can calculate 2D and 3D image features. Pyradiomics calculates standardised features extracted from medical images for the purpose of diagnosis, treatment, development and planning (Van Griethuysen et al. 2017). Semi-automated segmentation algorithms that use image features have been shown to have higher reproducibility than manual segmentations (Parmar et al. 2014). Studies have calculated radiomic features from both entire images or segmented regions of interest to study phenotypic differences (Aerts et al. 2014; Parmar et al. 2014; Yu et al. 2019).
Non-small-cell lung cancer prediction using radiomic features and machine learning methods
Published in International Journal of Computers and Applications, 2022
Radiomics has been identified as a novel technique employing a quantitative image feature of high throughput for both diagnosis as well as prognosis. The radiomics takes the images to be data and also achieves data mining for predicting clinical phenotype and gene data. For a certain application, radiomic approach will proceed using two phases: the first will be a training or a phase of feature selection after which there is a second phase of either testing or application. The phase of training will typically proceed in the following way. The features are extracted from its large corpus belonging to the training data in which an object of interest like a tumor is described that the computer algorithm is able to extract all its quantitative features in an automatic manner [7].