Improved image quality with deep learning reconstruction – a study on a semi-anthropomorphic upper-abdomen phantom
Purpose: To assess image quality of a deep learning reconstruction (DLR) algorithm across dose levels using a semi-anthropomorphic upper-abdominal phantom, and compare with filtered back projection (FBP) and hybrid iterative reconstruction (IR). Material and methods: CT scans obtained at five dose l...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2023-03-01
|
| Series: | Research in Diagnostic and Interventional Imaging |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2772652523000017 |
| _version_ | 1797237884126756864 |
|---|---|
| author | Tormund Njølstad, MD Anselm Schulz, MD PhD Kristin Jensen, PhD Hilde K. Andersen, MSc Anne Catrine T. Martinsen, PhD |
| author_facet | Tormund Njølstad, MD Anselm Schulz, MD PhD Kristin Jensen, PhD Hilde K. Andersen, MSc Anne Catrine T. Martinsen, PhD |
| author_sort | Tormund Njølstad, MD |
| collection | DOAJ |
| description | Purpose: To assess image quality of a deep learning reconstruction (DLR) algorithm across dose levels using a semi-anthropomorphic upper-abdominal phantom, and compare with filtered back projection (FBP) and hybrid iterative reconstruction (IR). Material and methods: CT scans obtained at five dose levels (CTDIvol 5, 10, 15, 20 and 25 mGy) were reconstructed with FBP, hybrid IR (IR50, IR70 and IR90) and DLR of low (DLL), medium (DLM) and high strength (DLH) in 0.625 mm and 2.5 mm slices. CT number, homogeneity, noise, contrast, contrast-to-noise ratio (CNR), noise texture deviation (NTD; a measure of IR-specific artifacts), noise power spectrum (NPS) and task-based transfer function (TTF) were compared between reconstruction algorithms. Results: CT numbers were highly consistent across reconstruction algorithms. Image noise was significantly reduced with higher levels of DLR. Noise texture (NPS and NTD) was with DLR maintained at comparable levels to FBP, contrary to increasing levels of hybrid IR. Images reconstructed with DLR of low and high strength in 0.625 mm slices showed similar noise characteristics to 2.5 mm slice FBP and IR50, respectively. Dose-reduction potential based on image noise with IR50 as reference was estimated to 35% for DLM and 74% for DLH. Conclusions: The novel DLR algorithm demonstrates robust noise reduction with maintained noise texture characteristics despite higher algorithm strength, and may have overcome important limitations of IR. There may be potential for dose reduction and additional benefit from thin-slice reconstruction. |
| first_indexed | 2024-04-24T17:26:50Z |
| format | Article |
| id | doaj.art-472ef7f529fc44078dde000fb33f173d |
| institution | Directory Open Access Journal |
| issn | 2772-6525 |
| language | English |
| last_indexed | 2024-04-24T17:26:50Z |
| publishDate | 2023-03-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Research in Diagnostic and Interventional Imaging |
| spelling | doaj.art-472ef7f529fc44078dde000fb33f173d2024-03-28T06:39:32ZengElsevierResearch in Diagnostic and Interventional Imaging2772-65252023-03-015100022Improved image quality with deep learning reconstruction – a study on a semi-anthropomorphic upper-abdomen phantomTormund Njølstad, MD0Anselm Schulz, MD PhD1Kristin Jensen, PhD2Hilde K. Andersen, MSc3Anne Catrine T. Martinsen, PhD4Department of Radiology and Nuclear Medicine, Oslo University Hospital Ullevål, Oslo 0450, Norway; Department of Radiology, Haukeland University Hospital, Bergen, Norway; Corresponding author.Department of Radiology and Nuclear Medicine, Oslo University Hospital Ullevål, Oslo 0450, NorwayDepartment of Physics and Computational Radiology, Oslo University Hospital, Oslo, NorwayDepartment of Physics and Computational Radiology, Oslo University Hospital, Oslo, NorwayFaculty of Health Sciences, Oslo Metropolitan University, Oslo, Norway; Sunnaas Rehabilitation Hospital, Nesodden, NorwayPurpose: To assess image quality of a deep learning reconstruction (DLR) algorithm across dose levels using a semi-anthropomorphic upper-abdominal phantom, and compare with filtered back projection (FBP) and hybrid iterative reconstruction (IR). Material and methods: CT scans obtained at five dose levels (CTDIvol 5, 10, 15, 20 and 25 mGy) were reconstructed with FBP, hybrid IR (IR50, IR70 and IR90) and DLR of low (DLL), medium (DLM) and high strength (DLH) in 0.625 mm and 2.5 mm slices. CT number, homogeneity, noise, contrast, contrast-to-noise ratio (CNR), noise texture deviation (NTD; a measure of IR-specific artifacts), noise power spectrum (NPS) and task-based transfer function (TTF) were compared between reconstruction algorithms. Results: CT numbers were highly consistent across reconstruction algorithms. Image noise was significantly reduced with higher levels of DLR. Noise texture (NPS and NTD) was with DLR maintained at comparable levels to FBP, contrary to increasing levels of hybrid IR. Images reconstructed with DLR of low and high strength in 0.625 mm slices showed similar noise characteristics to 2.5 mm slice FBP and IR50, respectively. Dose-reduction potential based on image noise with IR50 as reference was estimated to 35% for DLM and 74% for DLH. Conclusions: The novel DLR algorithm demonstrates robust noise reduction with maintained noise texture characteristics despite higher algorithm strength, and may have overcome important limitations of IR. There may be potential for dose reduction and additional benefit from thin-slice reconstruction.http://www.sciencedirect.com/science/article/pii/S2772652523000017Abdominal CTDeep learning reconstructionImage quality |
| spellingShingle | Tormund Njølstad, MD Anselm Schulz, MD PhD Kristin Jensen, PhD Hilde K. Andersen, MSc Anne Catrine T. Martinsen, PhD Improved image quality with deep learning reconstruction – a study on a semi-anthropomorphic upper-abdomen phantom Research in Diagnostic and Interventional Imaging Abdominal CT Deep learning reconstruction Image quality |
| title | Improved image quality with deep learning reconstruction – a study on a semi-anthropomorphic upper-abdomen phantom |
| title_full | Improved image quality with deep learning reconstruction – a study on a semi-anthropomorphic upper-abdomen phantom |
| title_fullStr | Improved image quality with deep learning reconstruction – a study on a semi-anthropomorphic upper-abdomen phantom |
| title_full_unstemmed | Improved image quality with deep learning reconstruction – a study on a semi-anthropomorphic upper-abdomen phantom |
| title_short | Improved image quality with deep learning reconstruction – a study on a semi-anthropomorphic upper-abdomen phantom |
| title_sort | improved image quality with deep learning reconstruction a study on a semi anthropomorphic upper abdomen phantom |
| topic | Abdominal CT Deep learning reconstruction Image quality |
| url | http://www.sciencedirect.com/science/article/pii/S2772652523000017 |
| work_keys_str_mv | AT tormundnjølstadmd improvedimagequalitywithdeeplearningreconstructionastudyonasemianthropomorphicupperabdomenphantom AT anselmschulzmdphd improvedimagequalitywithdeeplearningreconstructionastudyonasemianthropomorphicupperabdomenphantom AT kristinjensenphd improvedimagequalitywithdeeplearningreconstructionastudyonasemianthropomorphicupperabdomenphantom AT hildekandersenmsc improvedimagequalitywithdeeplearningreconstructionastudyonasemianthropomorphicupperabdomenphantom AT annecatrinetmartinsenphd improvedimagequalitywithdeeplearningreconstructionastudyonasemianthropomorphicupperabdomenphantom |