Related Experiment Video
Updated: Nov 11, 2025
Guidelines and Experience Using Imaging Biomarker Explorer IBEX for Radiomics
Published on: January 8, 2018
Radiomics Texture Features in Advanced Colorectal Cancer: Correlation with BRAF Mutation and 5-year Overall Survival
Adrian A Negreros-Osuna1, Anushri Parakh1, Ryan B Corcoran1
1Abdominal Imaging Division, Department of Radiology (A.A.N., A. Parakh, A. Pourvaziri, A.K., D.V.S.), and Cancer Center (R.B.C., D.P.R.), Massachusetts General Hospital, 55 Fruit St, Boston, MA 02114; and Facultad de Medicina y Hospital Universitario "Dr. José E. González," Universidad Autónoma de Nuevo León, Monterrey, Mexico (A.A.N.).
Radiomics texture features show promise as biomarkers for detecting BRAF mutations and predicting 5-year survival in stage IV colorectal cancer (CRC). These CT-based features can help identify patients with BRAF mutations and those likely to survive longer.
Area of Science:
- Radiology
- Oncology
- Medical Imaging Analysis
Background:
- Stage IV colorectal cancer (CRC) poses significant treatment challenges.
- Identifying biomarkers for BRAF mutation status and overall survival (OS) is crucial for personalized treatment strategies.
Purpose of the Study:
- To evaluate radiomics texture features from CT scans as potential biomarkers for detecting BRAF mutations.
- To assess the predictive value of these features for 5-year OS in patients with stage IV CRC.
Main Methods:
- Retrospective analysis of 145 stage IV CRC patients with pretreatment contrast-enhanced CT scans.
- Extraction of radiomics texture features (mean, SD, MPP, skewness, kurtosis, entropy) at various spatial scaling factors (SSFs).
- Statistical analysis included Wilcoxon rank sum test, Kaplan-Meier estimators, and Cox proportional-hazards regression.
Main Results:
- Significantly lower SDs and MPPs of radiomic features were observed in BRAF mutant tumors compared to wild-type tumors.
- Specific skewness and mean values at different SSFs predicted better 5-year OS.
- Tumor location did not significantly impact clinical outcomes.
Conclusions:
- Radiomics texture features demonstrate potential as non-invasive biomarkers for BRAF mutation status in CRC.
- These features can also serve as predictors of 5-year OS in advanced-stage CRC patients.
- Further validation is warranted to integrate radiomics into clinical decision-making for CRC management.
More Related Videos
10:25Author Spotlight: Integrating High-Resolution Intravital Imaging and MRI to Enhance Stereotactic Body Radiation Therapy Planning
Published on: April 12, 2024
15:48Tracking the Mammary Architectural Features and Detecting Breast Cancer with Magnetic Resonance Diffusion Tensor Imaging
Published on: December 15, 2014