Researchers have developed the Radiology Image Quality Evaluator (RIQE), a model designed to assess the quality of computed tomography (CT) images. RIQE is inspired by the Natural Image Quality Evaluator (NIQE) and is trained on a dataset of 3,792 full-dose CT slices from 158 patients. The model demonstrates effectiveness in ranking reduced-dose reconstructions as worse than full-dose counterparts and identifying images with increased noise. However, RIQE's preferences do not consistently align with lesion detectability, suggesting it should not be the sole criterion for selecting or tuning image processing techniques. AI
RANK_REASON The item describes a new research model and its evaluation, fitting the research bucket. [lever_c_demoted from research: ic=1 ai=1.0]
Read on Hugging Face Daily Papers →
- computed tomography
- LDCT-and-Projection-data
- Natural Image Quality Evaluator
- Radiology Image Quality Evaluator
- RIQE
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →