Journal: Frontiers in Aging

Article Title: The Usefulness of Radiomics Methodology for Developing Descriptive and Prognostic Image-Based Phenotyping in the Aging Population: Results From a Small Feasibility Study

doi: 10.3389/fragi.2022.853671

Figure Lengend Snippet: Prognostic value of radiomic data. (A) Heatmap of Rho of Spearman Correlation coefficients for an association of Radiomic Features and Incidence of new diseases and risk factors ( n = 101). On the x -axis, radiomics features are shown, and on the y -axis are the incidence of comorbidities and risk factors. The elements of the heatmap are color-coded depending on the value of the correlation coefficient. Red is for the highest value and green for the lowest, with 5 different colors in between. Abbreviations: DM: Diabetes Mellitus; COPD: Chronic Obstructive Pulmonary Disease; #: number; PC: primary care; ED: emergency department; IHF: Intensity Histogram Features; GLCM: Gray-Level Co-occurrence Matrix; GLRLM: Gray-Level run-Length Matrix; GLSZM: Gray-Level Size Zone Matrix; NGTDM: Neighborhood Gray-Tone Difference Matrix. Note. Tau B of Kendal was used for the statistical analysis. (B) Manhattan plot of p -values for associations between radiomic features and incidence of new diseases and risk factors ( n = 101). p -values for univariate associations between each radiomic feature and the incidence of new disease and risk factors after 2 years of following from baseline ultrasound. Radiomic features are situated on the x -axis in the same order as the heatmap, while the corresponding p -values are located on the y -axis and graph with a -LOG10 ( p -value) scale. Points above the red line ( p = <0.05) indicate radiomic features in which case the incidence of new diseases or risk factors showed significant association. (C) Hierarchical cluster dendrogram ( n = 44). Hierarchical cluster dendrogram of radiomic features significantly associated with hearing impairment, stroke, myocardial infarction, dementia or memory loss, and falls. Three independent clusters are identified for the radiomic phenotype ( p = 0.001).

Article Snippet: The features were extracted using the MATLAB toolbox Radiomics implemented by Vallières and others ( ).

Techniques:

Journal: Frontiers in Aging

Article Title: The Usefulness of Radiomics Methodology for Developing Descriptive and Prognostic Image-Based Phenotyping in the Aging Population: Results From a Small Feasibility Study

doi: 10.3389/fragi.2022.853671

Figure Lengend Snippet: Mitochondrial radiomic signature of ultrasound images. Radiomics aims to capture the informative content hidden in medical images, overcoming the limitations of the human eyes and human cognitive patterns. These patterns can be expressed in terms of macroscopic image-based radiomic features and carry information about their underlying pathophysiological processes and pinpoint specific biological mechanisms. This allows us to infer phenotypes or signatures, including prognostic information. Here we graphically showed that a radiomic phenotype, capturing the muscle heterogeneity, was strongly prognostic of the development of hearing impairment, stroke, myocardial infarction, dementia/memory loss, and/or falls. Based on the type of disease associated with the muscle ultrasound changes, we also believe this identified group of diseases shares a mitochondrial link. Icons utilized in this figure were obtain from the Noun Project from the following authors: Gorkem Oner (mitochondria), Gregor Cresnar (ear), Artem Kovyazin (brain), Tatina Vazest (heart), Luis Padra (fading head) and Visual Language Company (slipping person).

Article Snippet: The features were extracted using the MATLAB toolbox Radiomics implemented by Vallières and others ( ).

Techniques:

Journal: Physics in medicine and biology

Article Title: 4D radiomics: impact of 4D-CBCT image quality on radiomic analysis

doi: 10.1088/1361-6560/abd668

Figure Lengend Snippet: Histogram and GLCM radiomics errors across 10 phases of patient 3.

Article Snippet: In total, 540 radiomic features were extracted from the GTV of images based on the MATLAB based radiomics toolbox ( Vallières et al 2015 ).

Techniques:

Journal: Physics in medicine and biology

Article Title: 4D radiomics: impact of 4D-CBCT image quality on radiomic analysis

doi: 10.1088/1361-6560/abd668

Figure Lengend Snippet: Average radiomics features of the histogram, GLCM, GLRLM, GLSZM, NGTDM and wavelet of patient 3.

Article Snippet: In total, 540 radiomic features were extracted from the GTV of images based on the MATLAB based radiomics toolbox ( Vallières et al 2015 ).

Techniques:

Journal: Physics in medicine and biology

Article Title: 4D radiomics: impact of 4D-CBCT image quality on radiomic analysis

doi: 10.1088/1361-6560/abd668

Figure Lengend Snippet: Radiomics errors of all three testing patients with different training data and different projection numbers.

Article Snippet: In total, 540 radiomic features were extracted from the GTV of images based on the MATLAB based radiomics toolbox ( Vallières et al 2015 ).

Techniques:

Journal: Radiotherapy and oncology : journal of the European Society for Therapeutic Radiology and Oncology

Article Title: Predicting hypoxia status using a combination of contrast-enhanced computed tomography and [18F]-Fluorodeoxyglucose positron emission tomography radiomics features

doi: 10.1016/j.radonc.2017.11.025

Figure Lengend Snippet: Representative slices from the CT and 18F-FDG PET scans of three lesions with a diverse range of radiomics features and TBRmax. Each subfigure corresponds to a different lesion, with the CT scan on the left panel and the 18F-FDG PET scan on the right panel. The main features for each lesion are: (a) TBRmax=1.5, V>70↓ =35 cc, ℰ↑ =9735, P90%FDG=18.2, μ∼3FDG =0.48; (b) TBRmax=1.8, V>70↓ =6 cc, ℰ↑ =10980, P90%FDG =11.5, μ∼3FDG = 0.17; (c) TBRmax=1.0, V>70↓=0 cc, ℰ↑ =5702, P90%FDG=1.9, μ∼3FDG =0.65. The white contour indicates v↓, and the dark grey contour indicates v↑.

Article Snippet: Features were extracted in Matlab using the CERR Radiomics Imaging Quantification toolbox (RIQ, [ 25 , 26 ], April 2017 version).

Techniques: Computed Tomography

Journal: bioRxiv

Article Title: Hypoxia-related radiomics predict immunotherapy response: A multi-cohort study of NSCLC

doi: 10.1101/2020.04.02.020859

Figure Lengend Snippet: Pre-treatment (baseline) patient data are obtained, including: clinical covariates and computational image-based features (Radiomics). Radiomic features are extracted from standard-of-care imaging studies (yellow). Radiologists mark target lesions and lesions are automatically (or semi-automatically) segmented. Radiomic features are extracted from region-of-interest (purple). Unstable, non-reproducible and correlated radiomic features are removed. The remaining features are combined with the pre-treatment clinical covariates (green) and predictive model building approaches are applied which can be used for patient stratification and/or treatment selection.

Article Snippet: The tumor mask images (i.e., tumor delineations) were imported into in-house radiomic feature extraction toolbox that was created using MATLAB ® 2015b (The Mathworks Inc., Natick, Massachusetts) and C++ ( https://isocpp.org ).

Techniques: Imaging, Clinical Proteomics, Selection

Journal: bioRxiv

Article Title: Hypoxia-related radiomics predict immunotherapy response: A multi-cohort study of NSCLC

doi: 10.1101/2020.04.02.020859

Figure Lengend Snippet: Each column in the heat map represents a radiomic feature from the indicated feature group and region-of-interest (e.g., intratumoral or peritumoral). The features are compared between different segmentation algorithms (ALG), different initial parameters (IP) and test-retest scans (RIDER). The green boxes represent higher (CCC > 0.95), blue boxes represent moderate (CCC ≥ 0.75 & CCC ≤ 0.95) and red boxes represent lower (CCC < 0.75) CCCs.

Article Snippet: The tumor mask images (i.e., tumor delineations) were imported into in-house radiomic feature extraction toolbox that was created using MATLAB ® 2015b (The Mathworks Inc., Natick, Massachusetts) and C++ ( https://isocpp.org ).

Techniques:

Journal: bioRxiv

Article Title: Hypoxia-related radiomics predict immunotherapy response: A multi-cohort study of NSCLC

doi: 10.1101/2020.04.02.020859

Figure Lengend Snippet: The Classification and Regression Tree (CART) was used to identify patient risk groups based on a model containing one radiomic feature and two clinical features. Patients were grouped from low risk to very high risk based on the CART decision nodes and terminal nodes.

Article Snippet: The tumor mask images (i.e., tumor delineations) were imported into in-house radiomic feature extraction toolbox that was created using MATLAB ® 2015b (The Mathworks Inc., Natick, Massachusetts) and C++ ( https://isocpp.org ).

Techniques:

Journal: bioRxiv

Article Title: Hypoxia-related radiomics predict immunotherapy response: A multi-cohort study of NSCLC

doi: 10.1101/2020.04.02.020859

Figure Lengend Snippet: Kaplan-Meier survival curves estimates for overall survival between identified risk groups in the A) training (MCC 1) cohort, B) Test (MCC 2) cohorts and C) Validation (VA) cohort, and progressive-free survival in D) Training (MCC 1) cohort and E) Test (MCC 2) cohort. Test for agreement between radiomic and pathological immune response assessment.

Article Snippet: The tumor mask images (i.e., tumor delineations) were imported into in-house radiomic feature extraction toolbox that was created using MATLAB ® 2015b (The Mathworks Inc., Natick, Massachusetts) and C++ ( https://isocpp.org ).

Techniques: Biomarker Discovery

Journal: bioRxiv

Article Title: Hypoxia-related radiomics predict immunotherapy response: A multi-cohort study of NSCLC

doi: 10.1101/2020.04.02.020859

Figure Lengend Snippet: Whisker-box plots representing the association between CAIX expression on immunohistochemical staining and GLCM inverse difference CT radiomic feature. High and low GLCM inverse difference was found using novel cut-point (0.43) defined by CART analysis.

Article Snippet: The tumor mask images (i.e., tumor delineations) were imported into in-house radiomic feature extraction toolbox that was created using MATLAB ® 2015b (The Mathworks Inc., Natick, Massachusetts) and C++ ( https://isocpp.org ).

Techniques: Whisker Assay, Expressing, Immunohistochemical staining, Staining

Journal: bioRxiv

Article Title: Hypoxia-related radiomics predict immunotherapy response: A multi-cohort study of NSCLC

doi: 10.1101/2020.04.02.020859

Figure Lengend Snippet: Representative cases for testing the agreement between GLCM inverse difference and CAIX IHC expression. Correlation between high CAIX and high CT radiomic feature is seen on left side and correlation between low CAIX and low CT radiomic feature is seen on right side.

Article Snippet: The tumor mask images (i.e., tumor delineations) were imported into in-house radiomic feature extraction toolbox that was created using MATLAB ® 2015b (The Mathworks Inc., Natick, Massachusetts) and C++ ( https://isocpp.org ).

Techniques: Expressing