Impact of 18F-FDG PET Intensity Normalization on Radiomic Features of Oropharyngeal Squamous Cell Carcinomas and Machine Learning-Generated Biomarkers

Conclusion: Only one third of the features demonstrated a high degree of reproducibility across intensity-normalization techniques, making uniform normalization a prerequisite for interindividual comparability of radiomic markers. The choice of normalization technique may affect the radiomic features’ predictive value with respect to HPV. Our results show trends that normalization to the lentiform nucleus may improve model performance, although more evidence is needed to draw a firm conclusion.

Source: Journal of Nuclear Medicine - May 1, 2024 Category: Nuclear Medicine Authors: Haider, S. P., Zeevi, T., Sharaf, K., Gross, M., Mahajan, A., Kann, B. H., Judson, B. L., Prasad, M. L., Burtness, B., Aboian, M., Canis, M., Reichel, C. A., Baumeister, P., Payabvash, S. Tags: Basic Science Investigations Source Type: research