A Physics-Informed Deep Neural Network for Harmonization of CT Images

Abstract

Objective: Computed Tomography (CT) quantification is affected by the variability in image acquisition and rendition. This paper aimed to reduce this variability by harmonizing the images utilizing physics-based deep neural networks (DNNs). Methods: An adversarial generative network was trained on virtual CT images acquired under various imaging conditions using a virtual imaging platform with 40 computational patient models. These models featured anthropomorphic lungs with different levels of pulmonary diseases, including nodules and emphysema. Imaging was conducted using a validated CT simulator at two dose levels and varying reconstruction kernels. The trained model was tested on an independent virtual test dataset and two clinical datasets. Results: On the virtual test set, the harmonizer improved the structural similarity index from 79.3 $\pm$ 16.4% to 95.8 $\pm$ 1.7%, normalized mean squared error from 16.7 $\pm$ 9.7% to 9.2 $\pm$ 1.7%, and peak signal-to-noise ratio from 27.7 $\pm$ 3.7 dB to 32.2 $\pm$ 1.6 dB. Moreover, the harmonized images yielded more precise quantification of emphysema-based imaging biomarkers for lung attenuation, LAA −950 from 5.6 $\pm$ 8.7% to 0.23 $\pm$ 0.16%, Perc 15 from 43.4 $\pm$ 45.4 HU to 20.0 $\pm$ 7.5 HU, and Lung Mass from 0.3 $\pm$ 0.3 g to 0.1 $\pm$ 0.2 g. In clinical data, the harmonizer reduced biomarker variability by an average of 70%. For lung nodules, harmonized images improved the detectability index by 6.5-fold and DNN-based precision by 6%. Conclusion: The proposed harmonizer significantly enhances image quality and quantification accuracy in CT imaging. Significance: The study demonstrated the potential utility of image harmonization for consistent CT image quality and reliable quantification, which is crucial for clinical applications and patient management.