Impact of model-based iterative reconstruction on low-contrast lesion detection and image quality in abdominal CT: a 12-reader-based comparative phantom study with filtered back projection at different tube voltages
| Autor: | Clemens Reisinger, Bram Stieltjes, Luigia D’Errico, Sabine Bucher, Zsolt Szucs-Farkas, Matthias Streif, Markus Wilhelm, Achim Kircher, André Euler, Anna Hirschmann, Reto Eichenberger, David Buergler, Caroline Zaehringer, Sebastian T. Schindera, Sebastién Kopp |
|---|---|
| Přispěvatelé: | University of Zurich, Schindera, Sebastian T |
| Rok vydání: | 2016 |
| Předmět: |
medicine.medical_specialty
based iterative reconstruction Image quality 610 Medicine & health Iterative reconstruction Radiation Dosage Imaging phantom 030218 nuclear medicine & medical imaging Lesion Multidetector computed tomography 03 medical and health sciences 0302 clinical medicine medicine Image noise 2741 Radiology Nuclear Medicine and Imaging Humans Radiology Nuclear Medicine and imaging Tomographic reconstruction Radon transform contrast detection business.industry Phantoms Imaging Ultrasound General Medicine 10020 Clinic for Cardiac Surgery Low 030220 oncology & carcinogenesis Filtered back projection Radiographic Image Interpretation Computer-Assisted Radiology medicine.symptom Radiological phantom business Nuclear medicine Tomography X-Ray Computed Algorithms Model |
| Zdroj: | European radiology. 27(12) |
| ISSN: | 1432-1084 |
| Popis: | To evaluate the impact of model-based iterative reconstruction (MBIR) on image quality and low-contrast lesion detection compared with filtered back projection (FBP) in abdominal computed tomography (CT) of simulated medium and large patients at different tube voltages. A phantom with 45 hypoattenuating lesions was placed in two water containers and scanned at 70, 80, 100, and 120 kVp. The 120-kVp protocol served as reference, and the volume CT dose index (CTDIvol) was kept constant for all protocols. The datasets were reconstructed with MBIR and FBP. Image noise and contrast-to-noise-ratio (CNR) were assessed. Low-contrast lesion detectability was evaluated by 12 radiologists. MBIR decreased the image noise by 24% and 27%, and increased the CNR by 30% and 29% for the medium and large phantoms, respectively. Lower tube voltages increased the CNR by 58%, 46%, and 16% at 70, 80, and 100 kVp, respectively, compared with 120 kVp in the medium phantom and by 9%, 18% and 12% in the large phantom. No significant difference in lesion detection rate was observed (medium: 79-82%; large: 57-65%; P > 0.37). Although MBIR improved quantitative image quality compared with FBP, it did not result in increased low-contrast lesion detection in abdominal CT at different tube voltages in simulated medium and large patients. • MBIR improved quantitative image quality but not lesion detection compared with FBP. • Increased CNR by low tube voltages did not improve lesion detection. • Changes in image noise and CNR do not directly influence diagnostic accuracy. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full text from SpringerLink