Evaluating the impact of deep learning‐based image denoising on low‐dose CT for lung cancer screening

Abstract Purpose Low‐dose CT (LDCT) is increasingly being adopted as a preferred method for lung cancer screening. However, the accompanying rise in image noise necessitates robust denoising strategies. Therefore, this study compared LDCT images with their denoised counterparts using objective image quality metrics and key nodule‐related features. Methods The dataset utilized in this study was chest CT scans for lung cancer screening, sourced from the LDCT and Projection Data collection. Seven deep learning‐based image denoising methods were used in this work. The denoising performance was evaluated using root‐mean‐square error (RMSE), peak signal‐to‐noise ratio (PSNR), structural similarity index measure (SSIM), nodule size, CT density, and Lung‐RADS classification. Results For solid nodules, denoising improved SSIM from 51% to 60%–64%, reduced RMSE from 137.13 HU to 62.40–78.30 HU, and increased PSNR from 23.91 dB to 28.59–30.51 dB. It also reduced the percent difference in diameter (PD dia ) from 2.05% to 1.44%–1.52%, in volume (PD vol ) from 5.95% to 4.43%–4.70%, in mean HU value (PD HU ) from 24.40% to 8.54%–15.33%. For subsolid nodules, denoising improved SSIM from 47% to 57%–61%, reduced RMSE from 110.87 HU to 54.62–63.96 HU, and increased PSNR from 25.78 dB to 30.53–31.61 dB. Before denoising, the PD dia , PD vol and PD HU were 15.41%, 40.16% and 10.69%, respectively, which were 7.54%–15.94%, 17.54%–29.29%, and 6.10%–8.25% after denoising. These improvements led to higher Lung‐RADS categorization accuracy for solid nodules, while subsolid nodules remained more affected by noise and denoising‐induced bias. Conclusion The integration of denoising techniques into LDCT workflows could potentially enhance early lung cancer detection without increasing radiation exposure. Nonetheless, validating their influence on diagnostic performance remains crucial for clinical adoption.