Thursday 23 January 2025
A team of researchers has made a significant breakthrough in understanding how to improve the reliability and accuracy of medical imaging technologies, such as computed tomography (CT) scans. By analyzing large datasets of CT images, they have identified key factors that affect the reproducibility of radiomic features – mathematical measures extracted from images that can be used to diagnose and monitor diseases.
Radiomics is a rapidly growing field that has shown great promise in improving diagnostic accuracy and patient outcomes. However, one major challenge facing researchers is the variability in CT image acquisition parameters, such as slice thickness, which can affect the reproducibility of radiomic features.
The team’s study found that varying slice thickness had a significant impact on the reproducibility of radiomic features. They discovered that certain features were more robust to changes in slice thickness than others, and that some features were even sensitive to changes in other acquisition parameters, such as gray level discretization.
To address this challenge, the researchers developed a novel approach to feature extraction that combines multiple settings and regions of interest (ROIs) from CT images. By incorporating data from different ROIs, including tumor and liver parenchyma, they were able to create more robust radiomic features that are less affected by variations in image acquisition parameters.
The study’s findings have important implications for the development of radiomics-based diagnostic tools. By improving the reproducibility of radiomic features, researchers can increase the accuracy and reliability of disease diagnosis and monitoring. This could ultimately lead to better patient outcomes and more effective treatment strategies.
The research also highlights the need for further investigation into the effects of image acquisition parameters on radiomic feature variability. By understanding these factors, researchers can develop more robust and reliable diagnostic tools that are less susceptible to variations in imaging protocols.
Overall, this study demonstrates the importance of considering the impact of image acquisition parameters on radiomic features and highlights the potential benefits of combining multiple settings and ROIs to improve reproducibility. As radiomics continues to evolve as a field, it is essential that researchers prioritize the development of robust diagnostic tools that can be used confidently in clinical practice.
Cite this article: “Improving Radiomic Feature Reproducibility through Multi-Setting Analysis”, The Science Archive, 2025.
Computed Tomography, Radiomics, Image Acquisition Parameters, Slice Thickness, Gray Level Discretization, Feature Extraction, Reproducibility, Radiomic Features, Diagnostic Accuracy, Medical Imaging Technologies
