Quantum iterative reconstruction on a photon-counting detector CT improves the quality of hepatocellular carcinoma imaging

• Published in: Cancer imaging 2023-07, Vol.23 (1), p.69-69, Article 69
• Main Authors: Graafen, Dirk, Stoehr, Fabian, Halfmann, Moritz C, Emrich, Tilman, Foerster, Friedrich, Yang, Yang, Düber, Christoph, Müller, Lukas, Kloeckner, Roman
• Format: Article
• Language: English
• Subjects: Abdomen; Algorithms; Carcinoma, Hepatocellular - diagnostic imaging; Computed tomography; Conspicuity; Contrast agents; Coronary vessels; CT imaging; Detectors; Hepatocellular carcinoma; Hepatoma; Humans; Image analysis; Image contrast; Image enhancement; Image filters; Image quality; Image reconstruction; Impact analysis; Iodine; Iterative methods; Lesions; Liver; Liver cancer; Liver cirrhosis; Liver Neoplasms - diagnostic imaging; Medical imaging; Medical research; Medicine, Experimental; Modulation transfer function; Noise levels; Noise reduction; Patients; Photon-counting detector CT; Photons; Power spectra; Qualitative analysis; Quality management; Quality standards; Quantum iterative reconstruction; Scanners; Tomography, X-Ray Computed
• ISSN: 1470-7330; 1740-5025; 1470-7330
• DOI: 10.1186/s40644-023-00592-5

Excellent image quality is crucial for workup of hepatocellular carcinoma (HCC) in patients with liver cirrhosis because a signature tumor signal allows for non-invasive diagnosis without histologic proof. Photon-counting detector computed tomography (PCD-CT) can enhance abdominal image quality, especially in combination with a novel iterative reconstruction algorithm, quantum iterative reconstruction (QIR). The purpose of this study was to analyze the impact of different QIR levels on PCD-CT imaging of HCC in both phantom and patient scans. Virtual monoenergetic images at 50 keV were reconstructed using filtered back projection and all available QIR levels (QIR 1-4). Objective image quality properties were investigated in phantom experiments. The study also included 44 patients with triple-phase liver PCD-CT scans of viable HCC lesions. Quantitative image analysis involved assessing the noise, contrast, and contrast-to-noise ratio of the lesions. Qualitative image analysis was performed by three raters evaluating noise, artifacts, lesion conspicuity, and overall image quality using a 5-point Likert scale. Noise power spectra in the phantom experiments showed increasing noise suppression with higher QIR levels without affecting the modulation transfer function. This pattern was confirmed in the in vivo scans, in which the lowest noise levels were found in QIR-4 reconstructions, with around a 50% reduction in median noise level compared with the filtered back projection images. As contrast does not change with QIR, QIR-4 also yielded the highest contrast-to-noise ratios. With increasing QIR levels, rater scores were significantly better for all qualitative image criteria (all p