Hybrid Ensemble DL Model for Breast Cancer Detection and Classification with Enhanced Breast Lesion Segmentation using U-Net Model
Bibliographic record
Abstract
Early and accurate detection of breast cancer is a major challenge in medical diagnostics. Mammograms and ultrasound images often face issues such as poor contrast and unclear lesion boundaries, which can hinder the performance of deep learning diagnostic models. To tackle these problems, a new model called the Hybrid Ensemble DL Model for Breast Cancer Detection and Classification with Enhanced Breast Lesion Segmentation using U-Net Model (HEBEBU) is introduced. This HEBEBU model enhances image quality and diagnostic accuracy through advanced preprocessing, segmentation, and classification techniques. Initially, the model uses morphological erosion to reduce structural noise, making complex breast tissue patterns more understandable. It then applies CLAHE to improve local contrast and highlight micro-calcifications. A Laplacian of Gaussian (LoG) edge detection and Unsharp Masking sharpen the lesion borders and enhance structural visibility. For image segmentation, a unique U-Net design is employed, which maintains spatial resolution through skip connections and utilizes binary cross-entropy for training. An Active Contour Model (ACM) further refines segmentation results by accurately defining irregular lesion borders. An RVFL neural network ensures fast, non-iterative training while achieving high accuracy. The HEBEBU model was tested and showed impressive results, with training accuracy of 99.2% and testing accuracy of 99.0%. Other metrics fell consistently above 99%.
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 | 0.000 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 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".