Introduction to metrology series

Deep Learning-Based Image Conversion Improves the Reproducibility of Computed Tomography Radiomics Features: A Phantom Study

OBJECTIVES

This study aimed to evaluate the usefulness of deep learning-based image conversion to improve the reproducibility of computed tomography (CT) radiomics features.

MATERIALS AND METHODS

This study was conducted using an abdominal phantom with liver nodules. We developed an image conversion algorithm using a residual feature aggregation network to reproduce radiomics features with CT images under various CT protocols and reconstruction kernels. External validation was performed using images from different scanners, consisting of 8 different protocols. To evaluate the variability of radiomics features, regions of interest (ROIs) were drawn by targeting the liver parenchyma, vessels, paraspinal area, and liver nodules. We extracted 18 first-order, 68 second-order, and 688 wavelet radiomics features. Measurement variability was assessed using the concordance correlation coefficient (CCC), compared with the ground-truth image.

RESULTS

In the ROI-based analysis, there was an 83.3% improvement of CCC (80/96; 4 ROIs with 3 categories of radiomics features and 8 protocols) in synthetic images compared with the original images. Among them, the 56 CCC pairs showed a significant increase after image synthesis. In the radiomics feature-based analysis, 62.0% (3838 of 6192; 774 radiomics features with 8 protocols) features showed increased CCC after image synthesis, and a significant increase was noted in 26.9% (1663 of 6192) features. In particular, the first-order feature (79.9%, 115/144) showed better improvement in terms of the reproducibility of radiomics feature than the second-order (59.9%, 326/544) or wavelet feature (61.7%, 3397/5504).

CONCLUSIONS

Our study demonstrated that a deep learning model for image conversion can improve the reproducibility of radiomics features across various CT protocols, reconstruction kernels, and CT scanners.

A unified framework for simultaneous assessment of accuracy, between-, and within-reader variability of image segmentations

Medical imaging is utilized in a wide range of clinical applications. To enable a detailed quantitative analysis, medical images must often be segmented to label (delineate) structures of interest; for example, a tumor. Frequently, manual segmentation is utilized in clinical practice (e.g., in radiation oncology) to define such structures of interest. However, it can be quite time consuming and subject to substantial between-, and within-reader variability. A more reproducible, less variable, and more time efficient segmentation approach is likely to improve medical treatment. This potential has spurred the development of segmentation algorithms which harness computational power. Segmentation algorithms’ widespread use is limited due to difficulty in quantifying their performance relative to manual segmentation, which itself is subject to variation. This paper presents a statistical model which simultaneously estimates segmentation method accuracy, and between- and within-reader variability. The model is simultaneously fit for multiple segmentation methods within a unified Bayesian framework. The Bayesian model is compared to other methods used in literature via a simulation study, and application to head and neck cancer PET/CT data. The modeling framework is flexible and can be employed in numerous comparison applications. Several alternate applications are discussed in the paper.

Variability and Standardization of Quantitative Imaging: Monoparametric to Multiparametric Quantification, Radiomics, and Artificial Intelligence

Radiological images have been assessed qualitatively in most clinical settings by the expert eyes of radiologists and other clinicians. On the other hand, quantification of radiological images has the potential to detect early disease that may be difficult to detect with human eyes, complement or replace biopsy, and provide clear differentiation of disease stage. Further, objective assessment by quantification is a prerequisite of personalized/precision medicine. This review article aims to summarize and discuss how the variability of quantitative values derived from radiological images are induced by a number of factors and how these variabilities are mitigated and standardization of the quantitative values are achieved. We discuss the variabilities of specific biomarkers derived from magnetic resonance imaging and computed tomography, and focus on diffusion-weighted imaging, relaxometry, lung density evaluation, and computer-aided computed tomography volumetry. We also review the sources of variability and current efforts of standardization of the rapidly evolving techniques, which include radiomics and artificial intelligence.

Essentials of Statistical Methods for Assessing Reliability and Agreement in Quantitative Imaging

Quantitative imaging is increasing in almost all fields of radiological science. Modern quantitative imaging biomarkers measure complex parameters including metabolism, tissue microenvironment, tissue chemical properties or physical properties. In this paper, we focus on measurement reliability assessment in quantitative imaging. We review essential concepts related to measurement such as measurement variability and measurement error. We also discuss reliability study methods for intraobserver and interobserver variability, and the applicable statistical tests including: intraclass correlation coefficient, Pearson correlation coefficient, and Bland-Altman graphs and limits of agreement, standard error of measurement, and coefficient of variation.

Brown Adipose Reporting Criteria in Imaging STudies (BARCIST 1.0): Recommendations for Standardized FDG-PET/CT Experiments in Humans

Human brown adipose tissue (BAT) presence, metabolic activity, and estimated mass are typically measured by imaging [18F]fluorodeoxyglucose (FDG) uptake in response to cold exposure in regions of the body expected to contain BAT, using positron emission tomography combined with X-ray computed tomography (FDG-PET/CT). Efforts to describe the epidemiology and biology of human BAT are hampered by diverse experimental practices, making it difficult to directly compare results among laboratories. An expert panel was assembled by the National Institute of Diabetes and Digestive and Kidney Diseases on November 4, 2014 to discuss minimal requirements for conducting FDG-PET/CT experiments of human BAT, data analysis, and publication of results. This resulted in Brown Adipose Reporting Criteria in Imaging STudies (BARCIST 1.0). Since there are no fully validated best practices at this time, panel recommendations are meant to enhance comparability across experiments, but not to constrain experimental design or the questions that can be asked.

Role of the Quantitative Imaging Biomarker Alliance in optimizing CT for the evaluation of lung cancer screen-detected nodules

The Quantitative Imaging Biomarker Alliance (QIBA) is a multidisciplinary consortium sponsored by the RSNA to define processes that enable the implementation and advancement of quantitative imaging methods described in a QIBA profile document that outlines the process to reliably and accurately measure imaging features. A QIBA profile includes factors such as technical (product-specific) standards, user activities, and relationship to a clinically meaningful metric, such as with nodule measurement in the course of CT screening for lung cancer. In this report, the authors describe how the QIBA approach is being applied to the measurement of small pulmonary nodules such as those found during low-dose CT-based lung cancer screening. All sources of variance with imaging measurement were defined for this process. Through a process of experimentation, literature review, and assembly of expert opinion, the strongest evidence was used to define how to best implement each step in the imaging acquisition and evaluation process. This systematic approach to implementing a quantitative imaging biomarker with standardized specifications for image acquisition and postprocessing for a specific quantitative measurement of a pulmonary nodule results in consistent performance characteristics of the measurement (eg, bias and variance). Implementation of the QIBA small nodule profile may allow more efficient and effective clinical management of the diagnostic workup of individuals found to have suspicious pulmonary nodules in the course of lung cancer screening evaluation.

Lung Cancer Workshop XI: Tobacco-Induced Disease: Advances in Policy, Early Detection and Management

The Prevent Cancer Foundation Lung Cancer Workshop XI: Tobacco-Induced Disease: Advances in Policy, Early Detection and Management was held in New York, NY on May 16 and 17, 2014. The two goals of the Workshop were to define strategies to drive innovation in precompetitive quantitative research on the use of imaging to assess new therapies for management of early lung cancer and to discuss a process to implement a national program to provide high quality computed tomography imaging for lung cancer and other tobacco-induced disease. With the central importance of computed tomography imaging for both early detection and volumetric lung cancer assessment, strategic issues around the development of imaging and ensuring its quality are critical to ensure continued progress against this most lethal cancer.

Quantitative imaging biomarkers: a review of statistical methods for technical performance assessment

Technological developments and greater rigor in the quantitative measurement of biological features in medical images have given rise to an increased interest in using quantitative imaging biomarkers to measure changes in these features. Critical to the performance of a quantitative imaging biomarker in preclinical or clinical settings are three primary metrology areas of interest: measurement linearity and bias, repeatability, and the ability to consistently reproduce equivalent results when conditions change, as would be expected in any clinical trial. Unfortunately, performance studies to date differ greatly in designs, analysis method, and metrics used to assess a quantitative imaging biomarker for clinical use. It is therefore difficult or not possible to integrate results from different studies or to use reported results to design studies. The Radiological Society of North America and the Quantitative Imaging Biomarker Alliance with technical, radiological, and statistical experts developed a set of technical performance analysis methods, metrics, and study designs that provide terminology, metrics, and methods consistent with widely accepted metrological standards. This document provides a consistent framework for the conduct and evaluation of quantitative imaging biomarker performance studies so that results from multiple studies can be compared, contrasted, or combined.