Leveraging Stacked Ensemble Meta-Learning and Deep Neural Networks for Improved Skin Cancer Diagnosis
Bibliographic record
Abstract
Skin cancer poses a major global health threat, with melanoma being one of the deadliest forms due to its rapid progression and high mortality rate.Timely and precise detection is crucial for enhancing patient survival rates.In this study, we propose a robust stacked ensemble meta-learning model for the classification of various skin cancer types using dermoscopic images.The proposed framework integrates four convolutional neural networks (CNNs) such as custom CNN, InceptionResNetV2, ResNet101V2, and DenseNet201 as base learners, and leverages five meta-learners: Random Forest, Decision Tree, Logistic Regression, Gradient Boosting, and XGBoost for final classification.The system is trained and fine-tuned on a curated subset of the ISIC 2020 dataset, consisting of 2,357 images across nine skin lesion categories.Comprehensive experiments demonstrate the superior performance of the ensemble approach, achieving an accuracy of 98.63%, with both precision and recall reaching 98.64%.The Random Forest-based ensemble emerges as the top-performing configuration.Additionally, the study provides a comparative evaluation with existing methods and highlights the clinical potential of the model in reducing false positives and false negatives.By leveraging transfer learning, data augmentation, and metalearning strategies, this work contributes a scalable and accurate diagnostic tool for skin cancer detection, especially suitable for deployment in primary care and resource-limited healthcare settings.
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 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| 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".