[Paper Review] Image biomarker standardisation initiative - feature definitions.

Alex Zwanenburg, Stefan Leger+2 more|arXiv (Cornell University)|Dec 21, 2016

Radiomics and Machine Learning in Medical Imaging114 citations

TL;DR

This paper establishes standardized definitions for a comprehensive set of medical image features within the Image Biomarker Standardisation Initiative (IBSI), aiming to improve reproducibility and consistency in radiomics. By defining features such as intensity, shape, and texture metrics, the initiative enables reliable feature extraction and biomarker correlation in oncological outcomes.

ABSTRACT

While analysis of medical images has practically taken place since the first image was recorded, high throughput analysis of medical images is a more recent phenomenon. The aim of such a radiomics process is to provide decision support based on medical imaging. Part of the radiomics process is the conversion of image data into numerical features which capture different medical image aspects, and can be subsequently correlated as biomarkers to e.g. expected oncological treatment outcome. With the growth of the radiomics field, it has become clear that results are often difficult to reproduce, that standards for image processing and feature extraction are missing, and that reporting guidelines are absent. The image biomarker standardisation initiative (IBSI) seeks to address these issues. The current document provides definitions for a large number of image features.

Motivation & Objective

• Address the lack of standardization in radiomics feature extraction across medical imaging studies.
• Improve reproducibility of radiomics results by establishing consistent definitions for image features.
• Provide a foundation for reporting guidelines and interoperability in radiomics research.
• Facilitate reliable correlation of imaging features with clinical outcomes such as treatment response.

Proposed method

• Systematically define a broad range of image features, including intensity, shape, and texture-based metrics.
• Use standardized terminology and mathematical formulations to describe each feature.
• Ensure definitions are unambiguous and applicable across different imaging modalities and scanners.
• Align feature definitions with established principles in medical image analysis and radiomics.
• Organize features into logical categories for systematic application and reporting.
• Support implementation through clear documentation for researchers and software developers.

Experimental results

Research questions

• RQ1 How can image feature definitions be standardized to improve reproducibility in radiomics?
• RQ2 What are the key components of a consistent and unambiguous feature definition for medical imaging?
• RQ3 How can standardized feature definitions support reliable biomarker discovery in oncology?
• RQ4 What challenges exist in current radiomics practices that standardization can address?

Key findings

• The IBSI provides a comprehensive, standardized nomenclature for image features used in radiomics.
• Standardized definitions reduce ambiguity in feature extraction and improve cross-study comparability.
• The initiative enables consistent application of features across different imaging platforms and research settings.
• By defining features with mathematical precision, the framework supports reliable biomarker validation.
• The work lays the groundwork for future reporting guidelines and quality control in radiomics.
• The standardization effort enhances the potential for clinical translation of imaging biomarkers.