ISMDNet: Multi-Material Decomposition Using Deep Learning and Inverse Transform Sampling in Spectral Photon Counting CT
Bibliographic record
Abstract
Spectral Photon-Counting Computed Tomography (SPCCT) is a novel imaging technique that uses photon-counting detectors to acquire images across multiple X-ray energy windows. This energy discrimination enables material decomposition (MD) using the X-ray attenuation properties of various materials. In this study, we develop a novel deep learning (DL) framework for MD in SPCCT and compare its performance with recently developed MD methods. Our approach includes three variants of an encoder-decoder network: ISMDNet (Inverse Sampling Material Decomposition Network) with dual-residual blocks, ISMDNet-AT, which incorporates a channel-wise attention mechanism, and ISMDNet-TR, which integrates multiple transformer layers. We developed a unique synthetic image generation method using inverse transform sampling to address the scarcity of spectral CT datasets and the labor-intensive nature of ground-truth annotation. Synthetic images were used to train deep learning (DL) models, which were subsequently validated and tested on real SPCCT images. We conducted two studies, one involving various known materials in a phantom and the other using a biological tissue phantom. The performance of all approaches was evaluated using the root mean square error (RMSE) and the structural similarity index measure (SSIM) for MD quality, along with precision and the F1 score to assess the effectiveness of material identification. We compared our DL models’ performance with the state-of-the-art MD methods, including direct inversion (DI), generalized dictionary learning-based image domain (GDLIMD), MARS-MD, U-Net, and InceptNet. Our MD techniques consistently outperform the other MD methods.
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 distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 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 teacher head, 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".