Deep-learning image reconstruction significantly enhances image quality in dual-energy CT portal venography, aiding liver transplant planning.
Key Details
- 1Study compared deep-learning image reconstruction (DLIR) with adaptive statistical iterative reconstruction (ASiR-V) in dual-energy CT venography using low-dose protocols.
- 2DLIR high strength mode produced better image quality, lower noise, and higher contrast-to-noise and signal-to-noise ratios (p < 0.05).
- 3Radiologists gave DLIR high subjective ratings for image quality, vascular edge sharpness, and diagnostic confidence.
- 4Mean CT dose index volume was 9.79 mGy, DLP 326.26 mGy·cm, and effective dose 4.89 mSv; mean contrast volume was 79.5 mL.
- 555 keV virtual monoenergetic imaging improved iodine contrast effectiveness.
Why It Matters
Improving image quality in CT venography with deep-learning approaches helps deliver safer, lower-dose examinations without sacrificing diagnostic confidence. This has important implications for pre- and post-operative evaluation in liver transplantation, showcasing the clinical utility of imaging AI in challenging applications.
Source
AuntMinnie
Related News
•HealthExec
Survey Reveals Hurdles and Hopes for GenAI Adoption in U.S. Healthcare
A new survey finds high confidence in generative AI's potential among U.S. nurses but a lack of preparedness and governance impedes its impact.
•AuntMinnie
AI Analysis of CAC CT Scans Predicts COPD Risk
AI-derived myosteatosis measurements from CAC CT scans predict future COPD, outperforming emphysema-based biomarkers.
•Radiology Business
Organizational Barriers Impact Adoption of Radiology AI Support Tools
New research highlights that integrating radiology decision support tools faces significant organizational and cultural challenges.