Radiomics in pulmonary neuroendocrine tumours (NETs)

Abstract

Objectives The aim of this single-centre, observational, retrospective study is to find a correlation using Radiomics between the analysis of CT texture features of primary lesion of neuroendocrine (NET) lung cancer subtypes (typical and atypical carcinoids, large and small cell neuroendocrine carcinoma), Ki-67 index and the presence of lymph nodal mediastinal metastases. Methods Twenty-seven patients (11 males and 16 females, aged between 48 and 81 years old—average age of 70,4 years) with histological diagnosis of pulmonary NET with known Ki-67 status and metastases who have performed pre-treatment CT in our department were included. All examinations were performed with the same CT scan (Sensation 16-slice, Siemens). The study protocol was a baseline scan followed by 70 s delay acquisition after administration of intravenous contrast medium. After segmentation of primary lesions, quantitative texture parameters of first and higher orders were extracted. Statistics nonparametric tests and linear correlation tests were conducted to evaluate the relationship between different textural characteristics and tumour subtypes. Results Statistically significant (p < 0.05) differences were seen in post-contrast enhanced CT in multiple first and higher-order extracted parameters regarding the correlation with classes of Ki-67 index values. Statistical analysis for direct acquisitions was not significant. Concerning the correlation with the presence of metastases, one histogram feature (Skewness) and one feature included in the Gray-Level Co-occurrence Matrix (ClusterShade) were significant on contrast-enhanced CT only. Conclusions CT texture analysis may be used as a valid tool for predicting the subtype of lung NET and its aggressiveness.

Morphological, Functional and Texture Analysis Magnetic Resonance Imaging Features in the Assessment of Radiotherapy-Induced Xerostomia in Oropharyngeal Cancer

“…Therefore, interest towards artificial intelligence and its applications to imaging is steadily growing. In the context of the ever-growing academic importance of texture analysis [17,28,29,34], the present study was an effort to investigate its role in the assessment of salivary gland alterations and, specifically, RT-induced xerostomia.…”

“…Radiomics has been rapidly developing over the last few years; it is a hybrid analytical process aimed at determining the correlation between the characteristics of tissues and their corresponding digital images [17,27]. Texture analysis is a form of radiomics, in which macroscopic heterogeneities of tissues can be non-invasively studied to infer information about their microscopic architecture beyond the possibilities of the human eye as if it were a "virtual biopsy" [17,28,29]. Texture analysis is based on the extraction of parameters representing the distribution frequency, intensity, or direction of gray levels within the ROI in order to evaluate the single pixel, its interactions with adjacent pixels, and the distribution of pixels and voxels in the image [17,29].…”

Reproducibility of CT radiomic features in lung neuroendocrine tumours (NETs) patients: analysis in a heterogeneous population

“…To the best of our knowledge, this is the first study evaluating the efficacy of radiomics in considering the aggressiveness and tumour histotype of pulmonary NETs by analysing the possibility of results’ standardization and reproducibility on different CT scanners, therefore avoiding an important bias due to the heterogeneity of the machine used for imaging. In our previous study, we have selected significant features in Ki-67 classes and aggressiveness stratification, using a small number of patients trying to have the most homogeneous group of subjects, using only one CT scanner [ 27 ]. We therefore attempted with this additional study to evaluate a larger number of patients, assessing whether with hierarchical clustering analysis there was any bias resulting from the use of different CTs.…”

“…Radiomics is therefore an innovative technique used to characterize the inhomogeneity of a given tissue, and more specifically, as in our case, the lung NET lesions, through the extraction and analysis of features obtained by investigating regions of interest (ROI) from different imaging modalities such as CT, magnetic resonance imaging (MRI) or positron emission tomography CT (PET-CT) [ 22 – 26 ]. The application of texture analysis in NET of the lung could therefore be useful in both the diagnosis and early differentiation of distinctive NET tumour histotypes [ 27 , 28 ]. In this paper, we faced the reproducibility of CT radiomics features in lung NETs using different CT scanners, trying to integrate different features or to find new metrics in assessing tumour aggressiveness and histotype, that could be reproducible in daily practice.…”

Artificial Intelligence in Lung Cancer Imaging: Unfolding the Future

“…Radiomics and radiogenomics allow for further bridging of the gap between imaging data and the biological characteristics they represent [ 84 , 85 ]. Radiomics allows non-invasive assessment of tumor behavior and phenotype that goes beyond what can be directly identified by human operators.…”