Effect of a novel denoising technique on image quality and diagnostic accuracy in low-dose CT in patients with suspected appendicitis

Objectives: To determine the effect of a vendor unspecific, DICOM-based denoising technique on image quality and diagnostic performance in low-dose simulated abdominal computed tomography (CT) examinations in patients with suspected appendicitis. Methods and Materials: We included 51 patients who underwent contrast-enhanced abdominal CT with Filtered Back Projection due to suspected appendicitis. Realistic Low-Dose simulation generated low-dose datasets at 25% of the original exposition. QuantaStream Denoising, a novel DICOM-based technique denoised the simulated low-Dose datasets. Original (0100), low-dose (LD25) and denoised (DN25) datasets (n = 153) were evaluated regarding subjective image quality (5-point Likert scale; overall quality/image noise/diagnostic confidence), presence/absence of acute appendicitis, free abdominal air and abscess formation and objective image quality (Signal-to-Noise Ratio (SNR) and noise level) by two independent readers. Results: Subjective image quality was rated highest for O100, followed by DN25 and LD25 with significant differences (p = 0.001). Appendicitis was correctly identified in all datasets (n = 30). Appendicitis specific diagnostic confidence was highest for 0100 (p = 0.001), followed by DN25 and LD25 without significant difference. All complications were correctly identified on 0100 and DN25. On LD25, diagnostic accuracy decreased for abscess formations (sensitivity:0.714; specificity:1.0) and for free abdominal air (sensitivity:0.750; specificity:0.936). Regarding noise levels DN25 showed non-inferiority to 0100. SNRs of 0100 and DN25 showed no significant difference (p = 0.06). Conclusion: Our findings indicate that QuantaStream Denoising allows for maintained diagnostic image quality and diagnostic accuracy of low-dose abdominal CT examinations (25% of original exposition) in patients with suspected appendicitis without the need for raw sinogram data.