Texture Analysis of CT Colonography for Colorectal Cancer Management
Author Information
Author(s): Mamiya Hisashi, Toru Tochigi, Koichi Hayano, Gaku Ohira, Shunsuke Imanishi, Tetsuro Maruyama, Yoshihiro Kurata, Yumiko Takahashi, Atsushi Hirata, Hisahiro Matsubara
Primary Institution: Chiba University Graduate School of Medicine
Hypothesis
Can texture analysis of computed tomographic colonography (CTC) images serve as a novel biomarker for colorectal cancer (CRC)?
Conclusion
CE-CTC-derived texture parameters may be clinically useful biomarkers for managing CRC patients.
Supporting Evidence
- Texture parameters were significantly associated with tumor depth and lymph node metastasis.
- Patients with high GLCM-correlation tumors showed a significantly worse overall survival.
- GLCM correlation was identified as an independent prognostic factor for overall survival.
Takeaway
Doctors looked at pictures of tumors to see if they could find new ways to tell how bad the cancer is before surgery. They found that certain patterns in the pictures can help predict how the cancer will behave.
Methodology
This retrospective study analyzed texture parameters from CE-CTC images of 263 CRC patients and compared them with pathological factors and overall survival.
Potential Biases
The analysis was performed by a single team leader, which may introduce bias.
Limitations
The study was retrospective and single-center, and the method of tumor ROI delineation was subjective.
Participant Demographics
156 men and 107 women with a median age of 70 years.
Statistical Information
P-Value
p=0.001 for several associations
Confidence Interval
95% CI for GLCM correlation: 1.504–6.857
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website