Tri-Model Integration: Advancing Breast Cancer Immunohistochemical Image Generation through Multi-Method Fusion
Bibliographic record
Abstract
Immunohistochemical (IHC) staining is a crucial technique for diagnosing and formulating treatment plans for breast cancer, particularly by evaluating the expression of biomarkers like human epidermal growth factor receptor-2. However, the high cost and complexity of IHC staining procedures have driven research toward generating IHC-stained images directly from more readily available Hematoxylin and Eosin-stained images using image-to-image (I2I) translation methods. In this work, we propose a novel approach that combines the predictive capabilities of three state-of-the-art I2I models to enhance the quality and reliability of synthetic IHC images. Specifically, we designed a Convolutional Neural Network that takes as input a four-dimensional input comprising the outputs of three distinct models (each contributing an IHC prediction, which is an RGB three-dimensional output for each) and produces a final consensus image through a fusion mechanism. This ensemble method leverages the strengths of each model, leading to more robust and accurate IHC image generation. Extensive experiments on the BCI dataset demonstrate that our approach outperforms existing single-model methods, achieving superior Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity Index (SSIM) metrics. All of our code is available at: https://github.com/arshamhaq/BCI-fusion.Clinical RelevanceImproving the quality of synthetic IHC images can potentially reduce costs and streamline the diagnostic process, ultimately benefiting patient outcomes.
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 | 0.002 | 0.004 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.002 | 0.002 |
| Research integrity | 0.002 | 0.002 |
| Insufficient payload (model declined to judge) | 0.003 | 0.002 |
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".