A Robust and Interpretable Deep Learning-Based System for Multi-Category Skin Cancer Categorization on Resource-Constrained Edge Devices
Bibliographic record
Abstract
Skin cancer remains one of the most common and potentially fatal malignancies worldwide, particularly in regions with limited access to dermatological expertise and advanced diagnostic tools. While deep learning has shown promise in improving diagnostic accuracy, its deployment on low-power edge devices remains a significant challenge due to computational constraints and interpretability issues. In this study, we propose a robust and interpretable deep learning framework for multiclass skin cancer detection, specifically optimized for resource-constrained environments such as rural clinics. The framework leverages an ensemble of lightweight CNN architectures-MobileNetV2, EfficientNetB0, and DenseNet121—combined via a feature fusion strategy to enhance classification performance. Advanced preprocessing, data augmentation, and class-balancing techniques are applied to improve generalizability on the HAM10000 dataset. Grad-CAM is integrated to provide visual explanations of model decisions, enhancing clinical trust. The proposed model achieves an accuracy of 98.25% with an inference time of 0.01 seconds on Raspberry Pi 5, demonstrating its feasibility for real-time, portable, and reliable skin cancer screening. This approach holds promise for democratizing dermatological diagnostics and improving early detection outcomes in underserved 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 | 0.000 | 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) | 0.003 | 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".