Enhancing Skin Cancer Detection with Hybrid Attention Networks and Triple-Force Deep Learning Model
Bibliographic record
Abstract
In this research, present a Hybrid Attention and Triple-Force Deep Learning (TFDL) framework for automated skin cancer detection and classification via dermoscopic images. Our proposed framework combines the Hybrid Attention U-Net (HAU-Net) for precise segmentation, combining Channel Attention (CAM), and Spatial Attention (SAM) modules to increase feature localization and delineate lesion boundaries. The Adaptive Ensemble Learning model consisting of three state-of-the-art convolutional networks (VGG16, EfficientNet-B3, and ResNet50) is then used for classification, where feature aggregation is dynamically weighted, referring to the acronym AEL. The design of AEL enables improved learning efficiency, generalization, and interpretability of the model. The experiments using benchmark skin cancer datasets (HAM10000, ISIC, and DermNet) for evaluation purposes and show that we achieve 98.1% accurate diagnostic accuracy, 97.5% sensitivity, and 98.6% specificity when combined. These accuracy metrics are 3–5% better than that of the original CNN or even existing ensemble models in diagnostic accuracy. The findings of this research confirm that the proposed triple force deep learning framework provides a clinically viable, explainable, and high-precision AI system for real-time skin cancer screening and early diagnosis of skin cancer, reducing diagnostic errors, and assisting in the clinical decision-making process for dermatologists.
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 | 0.001 |
| 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 | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 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".