MedFusion Diff: A Diffusion-Based Framework for Brain Tumor Detection and MRI Generation
Bibliographic record
Abstract
Accurate detection and classification of brain tumors from MRI scans is still a significant challenge in medical image analysis due to a lack of labeled data, complicated tumor morphology, and high variability across patients. To address these challenges, we propose MedFusion-Diff, a new unified deep learning framework that integrates the generation of diffusion-based synthetic MRIs, tumor segmentation, and glioma subtype classification within a single task-aware framework. Unlike most existing approaches that perform these tasks independently, MedFusion-Diff uses a context-aware diffusion model to generate synthetic images that are realistic and clinically interpretable,thus improving model learning and generalization. The segmentation module effectively delineates tumor boundaries using attention networks, while the dual-stream classifier also enriches the feature set through both real and generated MR images. When benchmarked on a brain tumor dataset, MedFusion-Diff attained a Dice score of 94.3% for segmentation and a classification accuracy of 95.7%, significantly outperforming standard models (U-Net, ResNet, DenseNet). The results further demonstrate the benefits of incorporating synthetic data into the learning process itself, rather than simply treating it as a static step of augmentation. Along with compelling quantitative performance, MedFusion-Diff offers modular design, clinical adaptability, and scalability to multi-modal imaging tasks. Overall, this work not only pushes the state-of-the-art in brain tumor analysis but also reconceptualizes the usage of synthetic data in medical deep learning by embedding it as part of the model’s loop of learning, and opens the door to more intelligent, data-efficient, and clinically meaningful AI solutions in neuroimaging.
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 | 0.003 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.004 | 0.002 |
| Research integrity | 0.002 | 0.002 |
| Insufficient payload (model declined to judge) | 0.004 | 0.001 |
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".