Radiomics in Lung Diseases Imaging: State-of-the-Art for Clinicians

Abstract

Artificial intelligence (AI) has increasingly been serving the field of radiology over the last 50 years. As modern medicine is evolving towards precision medicine, offering personalized patient care and treatment, the requirement for robust imaging biomarkers has gradually increased. Radiomics, a specific method generating high-throughput extraction of a tremendous amount of quantitative imaging data using data-characterization algorithms, has shown great potential in individuating imaging biomarkers. Radiomic analysis can be implemented through the following two methods: hand-crafted radiomic features extraction or deep learning algorithm. Its application in lung diseases can be used in clinical decision support systems, regarding its ability to develop descriptive and predictive models in many respiratory pathologies. The aim of this article is to review the recent literature on the topic, and briefly summarize the interest of radiomics in chest Computed Tomography (CT) and its pertinence in the field of pulmonary diseases, from a clinician’s perspective.

Artificial Intelligence in Thyroid Field—A Comprehensive Review

“…From medical image such as US, CT, MR and/or PET images, the region of interest (ROI) is selected and subsequently the lesion is manually segmented, i.e., delineated with computer-assisted contouring, by an experienced clinician [ 7 ]. Subsequently, image data undergoes preprocessing operations, e.g., gray-level discretization, which enable a higher reproducibility of results [ 6 ]. The extraction of quantitative imaging features involves descriptors of spatial relationships between the various intensity level, heterogeneity patterns, shape and relations of the tissue lesion with surrounding tissues.…”

“…Typically, radiomics features are divided into [ 2 , 6 , 44 ]: Morphological, that are based on the geometric properties of the ROI, e.g. : volume, maximum surface area, maximum diameter.…”

“…Images, in addition, are characterized also by a high quantity of numerical information and recently, a quantitative evaluation has been developed in order to identify possible correlations between the numerical data contained in the digital images and the pathophysiology of the tissue [ 3 ]. The quantitative analysis has the aim to achieve information from standard-of-care images, e.g., ultrasound imaging (US), computer tomography (CT), magnetic resonance imaging (MRI) and positron emission tomography (PET), which are not easily quantifiable by means of naked-eye observations for clinical outcomes [ 5 , 6 ].…”

“…These applications are often referred as computer-aided detection (CADe) and computer-aided diagnosis (CADx) systems [ 3 ]. The output of the CAD analysis is used by the expert clinicians as a second opinion in detecting lesions or making diagnosis and aims at improving the accuracy of the diagnosis and reducing the time for image interpretation [ 6 ].…”

“…The above-mentioned approaches apply methodologies from the artificial intelligence (AI) field to achieve a partial or full automation of various steps of the process concerning the analysis of medical images [ 6 ]. Thorough understanding of their working principle is necessary in order to develop efficient predictive models and personalization treatment.…”