Related Experiment Video
Updated: Sep 13, 2025
07:15
Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model
Published on: August 16, 2020
6.9K
Predictive Modelling Using Thyroid Cartilage Segmentation and Radiomic Features: A Feasibility Study
Nivea Roy1, K Devaraja1, Prakashini Koteshwara2
1Department of Head and Neck Surgery, Kasturba Medical College, Manipal, Manipal Academy of Higher Education, Manipal, Karnataka 576104 India.
Summary
Radiomic analysis of thyroid cartilage shows promise for classifying age-related changes. This technique could potentially aid in laryngeal cancer staging by distinguishing normal from tumor-infiltrated cartilage.
Area of Science:
- Medical Imaging
- Radiology
- Machine Learning
Background:
- Laryngeal cancer diagnosis and staging are critical for effective treatment.
- Thyroid cartilage invasion signifies advanced cancer but is difficult to assess on CT due to age-related changes.
- Machine learning (ML) offers a potential solution for analyzing CT images.
Purpose of the Study:
- To investigate the feasibility of using radiomic features from manually segmented thyroid cartilage on CT images.
- To explore the potential of ML models in classifying age-related changes in thyroid cartilage.
- To assess the utility of radiomics in differentiating normal from potentially tumor-infiltrated cartilage.
Main Methods:
- Manual segmentation of thyroid cartilage from 39 CT images using 3D Slicer.
- Extraction of 107 radiomic features using Slicer Radiomics.
- Statistical and ML analyses performed using Jamovi and MATLAB.
Main Results:
- Successful manual segmentation of thyroid cartilage.
- Identification of significant gender and age-related differences in radiomic features.
- ML models achieved 100% accuracy for gender classification and 85.71% for age group classification.
- Regression models demonstrated improved accuracy with variable transformation.
Conclusions:
- Radiomic analysis of thyroid cartilage is a feasible and promising approach for classifying age-related changes.
- This methodology holds potential for improving laryngeal cancer staging by aiding in the differentiation of normal versus tumor-infiltrated cartilage.
