Texture analysis to assess structural modifications induced by radiotherapy

Area of Science:

• Radiotherapy
• Medical Imaging
• Oncology

Background:

• Radiotherapy (RT) for head and neck cancer can lead to high radiation doses in parotid glands.
• High doses may compromise parotid gland functionality and structure, impacting salivary function.
• Texture analysis is an emerging tool for evaluating tissue characteristics in medical imaging.

Purpose of the Study:

• To apply texture analysis on CT images to evaluate structural changes in parotid glands during RT.
• To assess the relationship between texture features, treatment time, and pre-treatment dosimetric parameters.

Main Methods:

• Texture analysis was performed on CT images of head and neck cancer patients at three time points during RT (beginning, intermediate, end).
• Features extracted included mean intensity, variance, entropy, homogeneity, local entropy, fractal dimension, and volume within the parotid gland.
• Correlations between volume/intensity variations and pre-treatment dosimetric parameters were investigated.

Main Results:

• A general decrease in parotid gland tissue complexity and heterogeneity was observed during RT.
• Different textural features exhibited distinct time trends, suggesting varied underlying biological mechanisms.
• Volume and mean intensity changes were correlated with pre-treatment dosimetric parameters.

Conclusions:

• Texture analysis effectively quantifies structural changes in parotid glands during head and neck radiotherapy.
• The observed changes indicate a relationship between radiation dose planning and post-treatment structural variations.
• This technique holds potential for monitoring treatment effects and guiding future radiotherapy strategies.