Accurate Multi-Material Decomposition in Dual-Energy CT: A Phantom Study
Bibliographic record
Abstract
DUAL-energy computed tomography (DECT) differentiates materials by exploiting the varying material linear attenuation coefficients (LACs) for different x-ray energy spectra. Multi-material decomposition (MMD) is a particularly attractive DECT clinical application to distinguish the complicated material components within the human body. One prior material assisted (PMA) image domain MMD method was implemented, but has suffered from inaccurate decomposition, magnified noise, and expensive computation. To suppress the noise, we implemented a statistical MMD (SMMD) algorithm, which applied the statistical weight to account for the noise variance in the DECT images. Its decomposition accuracy heavily relies on the initial value. In this paper, we propose a novel method to overcome these challenges. Based on the piecewise constant property of CT images with energy-dependent LAC, we assume that the pixels with high similarity have the same material composition. We cluster pixel patches into groups using the block-matching technique. The material composition in each group is preselected according to the shortest Euclidean distance in the energy map between the center of mass of the similar patch groups and the LAC of the object with known material composition pre-assigned by the clinician. MMD is performed on the central pixel of each patch using the preselected material composition. In a preliminary study, the proposed method is evaluated using the digital and water phantoms. The proposed method increases the volume fraction by 25.2% and decreases the standard deviation by 66.2% compared with the PMA method and increases the volume fraction by 19.6% compared with the SMMD method. The proposed method achieves an overall improvement of the normalized cross-correlation matrix diagonality by 34.8% and 69.4% compared with the PMA and SMMD methods. The phantom results indicate that the proposed method has the potential to be applied to clinical practice due to its increased decomposition accuracy, and suppressed noise and cross contamination.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.004 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.000 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".