医学                        
                
                                
                        
                            霍恩斯菲尔德秤                        
                
                                
                        
                            核医学                        
                
                                
                        
                            腹部                        
                
                                
                        
                            图像噪声                        
                
                                
                        
                            辐射剂量                        
                
                                
                        
                            骨盆                        
                
                                
                        
                            放射科                        
                
                                
                        
                            计算机断层摄影术                        
                
                                
                        
                            图像质量                        
                
                                
                        
                            计算机科学                        
                
                                
                        
                            图像(数学)                        
                
                                
                        
                            人工智能                        
                
                        
                    
            作者
            
                Josua A. Decker,Stefanie Bette,Nóra Lubina,Katharina Rippel,Franziska Braun,Franka Risch,Piotr Woźnicki,Claudia Wollny,Christian Scheurig‐Muenkler,Thomas Kroencke,Florian Schwarz            
         
                    
        
    
            
            标识
            
                                    DOI:10.1016/j.ejrad.2022.110181
                                    
                                
                                 
         
        
                
            摘要
            
            To analyze the quantitative and qualitative image quality of low-dose CT scans of the abdomen on a novel photon-counting detector CT (PCD-CT) in comparison with a traditional energy-integrating detector CT (EID-CT).Consecutive patients with clinically indicated low-dose CT were scanned on a PCD-CT and compared with a BMI-matched EID-CT-cohort scanned during the same timeframe. Radiation dose, image noise, and signal-to-noise ratio (SNR) were measured for each patient. Furthermore, image quality and conspicuity of abdominal structures (adrenal glands, mesenteric vessels, ureters, and renal pelvis) were assessed on 5-point Likert-scales (1 = very poor quality/not detectable; 5 = excellent quality/differentiability).Twenty patients (mean age 46.2 [range: 19-77]; 13 men) were included. Image noise was significantly lower (24.9 ± 3.3 vs. 31.4 ± 5.6 SD HU, p < 0.001) and SNR significantly higher (2.1 ± 0.3 vs. 1.5 ± 0.4; p < 0.001) on the PCD-CT. Subjective image quality was substantially higher (4.0 ± 0.3 vs. 3.1 ± 0.6; p < 0.001) and conspicuity better for the renal pelvis, ureters, and mesenteric vessels on the PCD-CT. There was no significant difference in the conspicuity of the adrenal glands. With increasing BMI (1st-4th BMI quartile), noise increased, and SNR decreased more strongly on the EID-CT than on the PCD-CT (ΔNoise: 39% vs. 2%, ΔSNR: -33% vs. -7% for EID-CT vs. PCD-CT, respectively) while radiation dose increased comparably (70 vs. 59%).Low-dose CT scans of the abdomen performed on a novel PCD-CT exhibit reduced noise, higher SNR, increased subjective image quality, and superior conspicuity of abdominal fine structures compared to scans in comparable patients on an EID-CT.
         
            
 
                 
                
                    
                    科研通智能强力驱动
Strongly Powered by AbleSci AI